SlideShare a Scribd company logo
1 of 75
1 
Power Production and Storage in 
Microgrids 
R. De Leone, A. Giovannelli, M. Pietrini† 
AIRO 2014, September 2014 
 
Research@Energy, Loccioni Group, Angeli di Rosora, AN, Italy 
† 
Mathematics Division, School of Science and Technologies, University of Camerino, MC, Italy
Contents 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
2 
Introduction 
Support Vector Machines 
Photovoltaic Energy Production Model 
Computational results 
Energy production, use and storage in smartgrids
Introduction 
Introduction 
Forecasting PV model 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
3
Forecasting PV model 
Introduction 
Forecasting PV model 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
4 
 Model 
Forecasting model for photovoltaic energy production
Forecasting PV model 
Introduction 
Forecasting PV model 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
4 
 Model 
Forecasting model for photovoltaic energy production 
 Aim 
To obtain an accurately daily forecast for a PV energy production plant 
located in Italy with a quarter-hour frequency
Forecasting PV model 
Introduction 
Forecasting PV model 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
4 
 Model 
Forecasting model for photovoltaic energy production 
 Aim 
To obtain an accurately daily forecast for a PV energy production plant 
located in Italy with a quarter-hour frequency 
 Technique 
Support Vector Machines, in particular -SVR
Forecasting PV model 
Introduction 
Forecasting PV model 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
4 
 Model 
Forecasting model for photovoltaic energy production 
 Aim 
To obtain an accurately daily forecast for a PV energy production plant 
located in Italy with a quarter-hour frequency 
 Technique 
Support Vector Machines, in particular -SVR 
 Scenario 
* scientific challenge: photovoltaic production depends on weather 
conditions 
* increasing development of electrical smartgrids
Support Vector Machines 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
5
-SVR (1) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
6 
Support Vector Machines (SVM) 
nonparametric technique for data classification (SVC) and regression (SVR)
-SVR (1) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
6 
Support Vector Machines (SVM) 
nonparametric technique for data classification (SVC) and regression (SVR) 
-SVR 
Given training data {(x1, y1), . . . , (xl, yl)}, where xi are input vectors and yi 
are the associated output values for xi, the support vector regression model 
requires the solution of the following optimization problem:
-SVR (1) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
6 
Support Vector Machines (SVM) 
nonparametric technique for data classification (SVC) and regression (SVR) 
-SVR 
Given training data {(x1, y1), . . . , (xl, yl)}, where xi are input vectors and yi 
are the associated output values for xi, the support vector regression model 
requires the solution of the following optimization problem: 
min 
w,b,, 
1 
2 
wTw + C 
l 
Xi=1 
i + − 
(+ 
i ) 
subject to yi − (wT (xi) + b)   + + 
i , 
(wT (xi) + b) − yi   + − 
i , 
+ 
i , − 
i  0, i = 1, . . . , l.
-SVR (2) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
7 
The parameters which control the regression quality are
-SVR (2) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
7 
The parameters which control the regression quality are 
 the cost of error C
-SVR (2) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
7 
The parameters which control the regression quality are 
 the cost of error C 
 the width of the tube
-SVR (2) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
7 
The parameters which control the regression quality are 
 the cost of error C 
 the width of the tube  
 the mapping function
-SVR (2) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
7 
The parameters which control the regression quality are 
 the cost of error C 
 the width of the tube  
 the mapping function  
In dual formulation of the problem, the kernel function is introduced 
k(xi, x) = (xi)T (x)
-SVR (2) 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
7 
The parameters which control the regression quality are 
 the cost of error C 
 the width of the tube  
 the mapping function  
In dual formulation of the problem, the kernel function is introduced 
k(xi, x) = (xi)T (x) 
Kernel function Formulation 
Linear k(x, y) = xTAy 
Polynomial k(x, y) = (xT x + c)d 
Radial Basis Function (RBF) k(x, y) = e−
kx−yk2 
Principal kernel functions.
-SVR 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
8 
-SVR 
The -Support Vector Machine problem is defined as follows:
-SVR 
Introduction 
Support Vector 
Machines 
-SVR (1) 
-SVR (2) 
-SVR 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
8 
-SVR 
The -Support Vector Machine problem is defined as follows: 
min 
w,b,, 
1 
2 
wTw + C( + 
1 
l 
l 
Xi=1 
i + − 
(+ 
i )) 
subject to yi − (wT (xi) + b)   + + 
i , 
(wT (xi) + b) − yi   + − 
i , 
+ 
i , − 
i  0, i = 1, . . . , l. 
where the parameter  allows to control the number of support vectors and 
training errors.
Photovoltaic Energy Production 
Model 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
9
Model data (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
10 
Collected measurements: historical data of an existing solar photovoltaic 
plant located in Angeli di Rosora (AN, Italy), provided by the Loccioni Group. 
Solyndra panels positioned on the roof of an industrial facility of Loccioni Group.
Model data (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
11 
 energy and power produced by the plant 
 solar irradiance 
 external environmental temperature
Model data (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
11 
 energy and power produced by the plant 
 solar irradiance 
 external environmental temperature 
Scatterplot of correlation between solar irradiance and energy production from the PV 
plant.
Model construction 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
12 
-SVR training input features:
Model construction 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
12 
-SVR training input features: 
 solar irradiance
Model construction 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
12 
-SVR training input features: 
 solar irradiance 
 external environmental temperature
Model construction 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
12 
-SVR training input features: 
 solar irradiance 
 external environmental temperature 
-SVR training output features:
Model construction 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
12 
-SVR training input features: 
 solar irradiance 
 external environmental temperature 
-SVR training output features: 
 energy production
Model construction 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
12 
-SVR training input features: 
 solar irradiance 
 external environmental temperature 
-SVR training output features: 
 energy production 
Parameters choice 
1. kernel type: Radial Basis Function (RBF) 
2. constant appearing in kernel formula: 
 = 1/features 
3. number of support vectors:  = 0.5 
4. cost of error: C = 1.8 
5. training data: 14 days
Prediction accuracy measures 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
13 
 Root Mean Square Error (RMSE) 
RMSE = rPn 
i=1(yi − ˆyi)2 
n
Prediction accuracy measures 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
13 
 Root Mean Square Error (RMSE) 
RMSE = rPn 
i=1(yi − ˆyi)2 
n 
 Mean Absolute Percentage Error (MAPE) 
MAPE = 
100% 
n 
n 
Xi=1 
| 
yi − ˆyi 
yi 
|
Prediction accuracy measures 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
13 
 Root Mean Square Error (RMSE) 
RMSE = rPn 
i=1(yi − ˆyi)2 
n 
 Mean Absolute Percentage Error (MAPE) 
MAPE = 
100% 
n 
n 
Xi=1 
| 
yi − ˆyi 
yi 
| 
 Coefficient of Determination (R2) 
R2 = Pn 
2 
i=1(ˆyi − ¯y)n 
2 
i=1(yi − ¯y)P
Prediction accuracy measures 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Model data (1) 
Model data (2) 
Model construction 
Prediction accuracy 
measures 
Computational results 
Energy production, use 
and storage in 
smartgrids 
13 
 Root Mean Square Error (RMSE) 
RMSE = rPn 
i=1(yi − ˆyi)2 
n 
 Mean Absolute Percentage Error (MAPE) 
MAPE = 
100% 
n 
n 
Xi=1 
| 
yi − ˆyi 
yi 
| 
 Coefficient of Determination (R2) 
R2 = Pn 
2 
i=1(ˆyi − ¯y)n 
2 
i=1(yi − ¯y)Pwhere yi = actual data, ˆyi = estimated data, ¯y = mean of the actual data.
Computational results 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
14
Results (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
15 
Comparison between real and forecasted PV energy production on 15th of June 2012.
Results (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
15 
Comparison between real and forecasted PV energy production on 15th of June 2012. 
RMSE = 0.5275; MAPE = 0.0785; R2 = 0.9944.
Results (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
16 
Scatterplot of correlation between real and forecasted energy production.
Results (3) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
17 
Comparison between real and forecasted PV energy production on 16th of April 2012.
Results (3) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
17 
Comparison between real and forecasted PV energy production on 16th of April 2012. 
RMSE = 0.2810; MAPE = 0.1143; R2 = 0.9926.
Results (4) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
What does it happen when the input values of solar irradiance for the next 
day are not fully accurate? 
18
Results (4) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
What does it happen when the input values of solar irradiance for the next 
day are not fully accurate? 
18 
Comparison between real and forecasted solar irradiance on 11th of October 2013.
Results (5) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
19 
Comparison between real energy production on 11th of October 2013 and forecasted one 
using actual data of irradiance and temperature and estimated weather data.
Results (5) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
19 
Comparison between real energy production on 11th of October 2013 and forecasted one 
using actual data of irradiance and temperature and estimated weather data. 
RMSE = 3.5876; MAPE = 3.5718; R2 = 0.3616.
Observations about the model 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
 accurate results compared with the three different evaluation statistics 
20 
(RMSE, MAPE, R2);
Observations about the model 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
 accurate results compared with the three different evaluation statistics 
20 
(RMSE, MAPE, R2); 
 predictive model risk concerning with forecasted errors on input 
weather data provided by internet weather site;
Observations about the model 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Results (1) 
Results (2) 
Results (3) 
Results (4) 
Results (5) 
Observations about the 
model 
Energy production, use 
and storage in 
smartgrids 
 accurate results compared with the three different evaluation statistics 
20 
(RMSE, MAPE, R2); 
 predictive model risk concerning with forecasted errors on input 
weather data provided by internet weather site; 
 good application of -SVR model with the possibility of choosing the 
number of support vectors.
Energy production, use and storage 
in smartgrids 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
21
Leaf Community (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
22 
The study on photovoltaic energy production forecast is part of a greater 
project realized by Loccioni Group: the 
Leaf Community smartgrid. 
Leaf Community smartgrid by Loccioni Group.
Leaf Community (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
22 
The study on photovoltaic energy production forecast is part of a greater 
project realized by Loccioni Group: the 
Leaf Community smartgrid.
Leaf Community (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
23 
Leaf Community smartgrid consists of
Leaf Community (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
23 
Leaf Community smartgrid consists of 
 energy production systems from renewable sources: 
 photovoltaic plant 
 micro-hydroelectric power station
Leaf Community (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
23 
Leaf Community smartgrid consists of 
 energy production systems from renewable sources: 
 photovoltaic plant 
 micro-hydroelectric power station 
 consumption buildings: 
 industrial facility 
 office building
Leaf Community (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
23 
Leaf Community smartgrid consists of 
 energy production systems from renewable sources: 
 photovoltaic plant 
 micro-hydroelectric power station 
 consumption buildings: 
 industrial facility 
 office building 
 energy storage device.
Leaf Community (3) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
24 
The smartgrid architecture scheme.
Leaf Community System Model 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
25 
The smartgrid optimization model will manage the energy flows within the 
network in order to 
 ensure the maximum economic benefit 
 achieve the goal of energy autonomy for the most of time.
Leaf Community System Model 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
25 
The smartgrid optimization model will manage the energy flows within the 
network in order to 
 ensure the maximum economic benefit 
 achieve the goal of energy autonomy for the most of time. 
Flow chart of the smartgrid optimization problem.
Optimization Problem Formulation (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
26 
 The optimization problem is formulated as a mixed-integer linear 
programming problem.
Optimization Problem Formulation (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
26 
 The optimization problem is formulated as a mixed-integer linear 
programming problem. 
 Each component of the smartgrid is modeled separately, based on its 
characteristics and constraints.
Optimization Problem Formulation (2a) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
27 
Input data
Optimization Problem Formulation (2a) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
27 
Input data • photovoltaic energy production forecast 
Photovoltaic energy production forecast using -SVR.
Optimization Problem Formulation (2b) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
28 
Input data
Optimization Problem Formulation (2b) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
28 
Input data • energy consumption forecast 
Energy consumption forecast using linear regression.
Optimization Problem Formulation (2c) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
29 
Input data
Optimization Problem Formulation (2c) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
29 
Input data • electricity tariffs 
Selling and purchased electricity tariffs.
Optimization Problem Formulation (3a) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
30 
Problem data 
(t) energy purchased price at time t (e/kWh) 
cIND(t) energy consumption of the industrial facility at time t (kWh) 
cOF(t) energy consumption of the office building at time t (kWh) 
e(0) energy present in the battery storage at time 0 (kWh) 
pPV (t) photovoltaic energy production at time t (kWh) 
pHD(t) hydropower production at time t (kWh) 
(t) energy selling price at time t (e/kWh)
Optimization Problem Formulation (3b) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
31 
Problem parameters 
cINV energy consumption of the inverter (kWh) 
m minimum energy to charge/discharge the battery storage (kWh) 
M maximum energy to charge/discharge the battery storage (kWh) 
mESS minimum energy capacity of the battery storage (kWh) 
MESS maximum energy capacity of the battery storage (kWh) 
C charging efficiency of the battery storage 
D discharging efficiency of the battery storage 
T1 efficiency of the transformer T1 
T2 efficiency of the transformer T2
Optimization Problem Formulation (3c) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
32 
Problem variables 
b1(t) energy to buy for Block 1 at time t (kWh) 
b2(t) energy to buy for Block 2 at time t (kWh) 
e(t) amount of energy present in the battery storage at time t (kWh) 
b(t) buying/not buying energy at time t (kWh) 
s(t) selling/not selling energy at time t (kWh) 
r12(t) residual energy from Block 1 to Block 2 at time t (kWh) 
s(t) energy to sell at time t (kWh) 
x1(t) energy sent to the battery storage from Block 1 at time t (kWh) 
x(t) energy sent to the battery storage from the main grid at time t (kWh) 
y1(t) energy to discharge from the battery storage to Block 1 at time t (kWh) 
y2(t) energy to discharge from the battery storage to Block 2 at time t (kWh) 

C(t) charging state of the battery storage at time t 

D(t) discharging state of the battery storage at time t
Optimization Problem Formulation (4) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
33 
Objective function 
192 
Xt=1 h(t)s(t)T1 − (t)b1(t) + b2(t)i 
Maximize the difference between the income deriving from the sold energy 
to the main distribution grid and the cost paid for the purchased energy, 
whenever the generated output power from the renewable energy sources 
is insufficient to cover the load demand.
Optimization Problem Formulation (5) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
34 
Constraints
Optimization Problem Formulation (5) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
34 
Constraints 
 Energy balance constraints: 
energy purchased from utility grid + energy generated by the different 
sources = the local demand + the energy loss in the transmission lines 
+ the energy stored in the battery.
Optimization Problem Formulation (5) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
34 
Constraints 
 Energy balance constraints: 
energy purchased from utility grid + energy generated by the different 
sources = the local demand + the energy loss in the transmission lines 
+ the energy stored in the battery. 
 Battery storage constraints: 
 capacity of the battery; 
 battery charging/discharging range; 
 energy level present in the battery; 
 impossibility to simultaneously charge/discharge the storage.
Optimization Problem Formulation (5) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
34 
Constraints 
 Energy balance constraints: 
energy purchased from utility grid + energy generated by the different 
sources = the local demand + the energy loss in the transmission lines 
+ the energy stored in the battery. 
 Battery storage constraints: 
 capacity of the battery; 
 battery charging/discharging range; 
 energy level present in the battery; 
 impossibility to simultaneously charge/discharge the storage. 
 Main grid constraints: 
 only energy produced from the renewable sources can be sold; 
 impossibility to simultaneously purchase/sell energy.
Leaf Community Optimization Problem: Results (1) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
35 
The energy curves for two winter days with the storage fully charged.
Leaf Community Optimization Problem: Results (2) 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
36 
The energy curves for two summer days with the storage fully discharged.
Conclusions 
Introduction 
Support Vector 
Machines 
Photovoltaic Energy 
Production Model 
Computational results 
Energy production, use 
and storage in 
smartgrids 
Leaf Community (1) 
Leaf Community (2) 
Leaf Community (3) 
Leaf Community 
System Model 
Optimization Problem 
Formulation (1) 
Optimization Problem 
Formulation (2a) 
Optimization Problem 
Formulation (2b) 
Optimization Problem 
Formulation (2c) 
Optimization Problem 
Formulation (3a) 
Optimization Problem 
Formulation (3b) 
Optimization Problem 
Formulation (3c) 
37 
 The study on photovoltaic energy production forecast becomes an 
important tool in order to manage the energy flows in the smartgrid. 
 The correctness of the forecasts is determinant for a good result of the 
optimization algorithm. 
 Due to the time-varying nature of the consumptions and the variation 
of the battery state of charge, the optimization problem needs to 
respond to continuously changing operation conditions. 
 Good results of the optimization problem are obtained.

More Related Content

Similar to Power Production Forecasting

EES-UETP Microgrid course
 EES-UETP Microgrid course EES-UETP Microgrid course
EES-UETP Microgrid courseJuan C. Vasquez
 
IRJET- Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...
IRJET-  	  Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...IRJET-  	  Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...
IRJET- Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...IRJET Journal
 
Artificial Neural Network Applied to Estimate the Power Output of BIPV Systems
Artificial Neural Network Applied to Estimate the Power Output of BIPV SystemsArtificial Neural Network Applied to Estimate the Power Output of BIPV Systems
Artificial Neural Network Applied to Estimate the Power Output of BIPV SystemsIOSRjournaljce
 
Quaid-e-Azam solar park.pptx
Quaid-e-Azam solar park.pptxQuaid-e-Azam solar park.pptx
Quaid-e-Azam solar park.pptxSherAli260123
 
PROTOTYPE OF IOT BASED DC MICROGRID AUTOMATION
PROTOTYPE OF IOT BASED DC MICROGRID AUTOMATIONPROTOTYPE OF IOT BASED DC MICROGRID AUTOMATION
PROTOTYPE OF IOT BASED DC MICROGRID AUTOMATIONIRJET Journal
 
Solar based IOT controlled EV
Solar based IOT controlled EVSolar based IOT controlled EV
Solar based IOT controlled EVIRJET Journal
 
Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...
Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...
Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...IRJET Journal
 
Iaetsd design, engineerning and analysis
Iaetsd design, engineerning and analysisIaetsd design, engineerning and analysis
Iaetsd design, engineerning and analysisIaetsd Iaetsd
 
The examination and use of Solar Energy PV Power
The examination and use of Solar Energy PV PowerThe examination and use of Solar Energy PV Power
The examination and use of Solar Energy PV PowerIRJET Journal
 
A Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEM
A Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEMA Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEM
A Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEMVicki Cristol
 
Optimal design and static simulation of a hybrid solar vehicle
Optimal design and static simulation of a hybrid solar vehicleOptimal design and static simulation of a hybrid solar vehicle
Optimal design and static simulation of a hybrid solar vehicleIRJASH
 
Designing the virtual model of a mechatronic micro positioning and micro-meas...
Designing the virtual model of a mechatronic micro positioning and micro-meas...Designing the virtual model of a mechatronic micro positioning and micro-meas...
Designing the virtual model of a mechatronic micro positioning and micro-meas...eSAT Journals
 
IRJET- Sun Tracking Solar Panel
IRJET-  	  Sun Tracking Solar PanelIRJET-  	  Sun Tracking Solar Panel
IRJET- Sun Tracking Solar PanelIRJET Journal
 
Automated Wind and Solar Powered Water Drone Monitoring and Controlling System
Automated Wind and Solar Powered Water Drone Monitoring and Controlling SystemAutomated Wind and Solar Powered Water Drone Monitoring and Controlling System
Automated Wind and Solar Powered Water Drone Monitoring and Controlling SystemIRJET Journal
 
Designing the virtual model of a mechatronic micropositioning
Designing the virtual model of a mechatronic micropositioningDesigning the virtual model of a mechatronic micropositioning
Designing the virtual model of a mechatronic micropositioningeSAT Publishing House
 
Concept of Microgrid.pdf
Concept of Microgrid.pdfConcept of Microgrid.pdf
Concept of Microgrid.pdfhanadi40
 
Proposal and implementation of a novel perturb and observe algorithm using em...
Proposal and implementation of a novel perturb and observe algorithm using em...Proposal and implementation of a novel perturb and observe algorithm using em...
Proposal and implementation of a novel perturb and observe algorithm using em...saad motahhir
 
Proposal for 1kwp Roof-Top Solar PV Plant
Proposal for 1kwp Roof-Top Solar PV PlantProposal for 1kwp Roof-Top Solar PV Plant
Proposal for 1kwp Roof-Top Solar PV PlantIRJET Journal
 

Similar to Power Production Forecasting (20)

EES-UETP Microgrid course
 EES-UETP Microgrid course EES-UETP Microgrid course
EES-UETP Microgrid course
 
IRJET- Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...
IRJET-  	  Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...IRJET-  	  Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...
IRJET- Performance Analysis of a DC Microgrid Integrated Dynamic Voltage ...
 
Artificial Neural Network Applied to Estimate the Power Output of BIPV Systems
Artificial Neural Network Applied to Estimate the Power Output of BIPV SystemsArtificial Neural Network Applied to Estimate the Power Output of BIPV Systems
Artificial Neural Network Applied to Estimate the Power Output of BIPV Systems
 
Quaid-e-Azam solar park.pptx
Quaid-e-Azam solar park.pptxQuaid-e-Azam solar park.pptx
Quaid-e-Azam solar park.pptx
 
PROTOTYPE OF IOT BASED DC MICROGRID AUTOMATION
PROTOTYPE OF IOT BASED DC MICROGRID AUTOMATIONPROTOTYPE OF IOT BASED DC MICROGRID AUTOMATION
PROTOTYPE OF IOT BASED DC MICROGRID AUTOMATION
 
Solar based IOT controlled EV
Solar based IOT controlled EVSolar based IOT controlled EV
Solar based IOT controlled EV
 
Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...
Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...
Simulation of MPPT Controller for photovoltaic system Grid-connected using Mo...
 
CSTEP BORRERO R poster
CSTEP BORRERO R posterCSTEP BORRERO R poster
CSTEP BORRERO R poster
 
Converted ansys f
Converted ansys fConverted ansys f
Converted ansys f
 
Iaetsd design, engineerning and analysis
Iaetsd design, engineerning and analysisIaetsd design, engineerning and analysis
Iaetsd design, engineerning and analysis
 
The examination and use of Solar Energy PV Power
The examination and use of Solar Energy PV PowerThe examination and use of Solar Energy PV Power
The examination and use of Solar Energy PV Power
 
A Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEM
A Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEMA Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEM
A Seminar Project Report ARDUINO BASED SOLAR TRACKING SYSTEM
 
Optimal design and static simulation of a hybrid solar vehicle
Optimal design and static simulation of a hybrid solar vehicleOptimal design and static simulation of a hybrid solar vehicle
Optimal design and static simulation of a hybrid solar vehicle
 
Designing the virtual model of a mechatronic micro positioning and micro-meas...
Designing the virtual model of a mechatronic micro positioning and micro-meas...Designing the virtual model of a mechatronic micro positioning and micro-meas...
Designing the virtual model of a mechatronic micro positioning and micro-meas...
 
IRJET- Sun Tracking Solar Panel
IRJET-  	  Sun Tracking Solar PanelIRJET-  	  Sun Tracking Solar Panel
IRJET- Sun Tracking Solar Panel
 
Automated Wind and Solar Powered Water Drone Monitoring and Controlling System
Automated Wind and Solar Powered Water Drone Monitoring and Controlling SystemAutomated Wind and Solar Powered Water Drone Monitoring and Controlling System
Automated Wind and Solar Powered Water Drone Monitoring and Controlling System
 
Designing the virtual model of a mechatronic micropositioning
Designing the virtual model of a mechatronic micropositioningDesigning the virtual model of a mechatronic micropositioning
Designing the virtual model of a mechatronic micropositioning
 
Concept of Microgrid.pdf
Concept of Microgrid.pdfConcept of Microgrid.pdf
Concept of Microgrid.pdf
 
Proposal and implementation of a novel perturb and observe algorithm using em...
Proposal and implementation of a novel perturb and observe algorithm using em...Proposal and implementation of a novel perturb and observe algorithm using em...
Proposal and implementation of a novel perturb and observe algorithm using em...
 
Proposal for 1kwp Roof-Top Solar PV Plant
Proposal for 1kwp Roof-Top Solar PV PlantProposal for 1kwp Roof-Top Solar PV Plant
Proposal for 1kwp Roof-Top Solar PV Plant
 

More from Renato De Leone (12)

Main
MainMain
Main
 
Its
ItsIts
Its
 
Inn
InnInn
Inn
 
Main
MainMain
Main
 
Brasile
BrasileBrasile
Brasile
 
Pau 2015
Pau 2015Pau 2015
Pau 2015
 
Poster microgrid
Poster microgridPoster microgrid
Poster microgrid
 
SD approach to the boarding process
SD approach to the boarding processSD approach to the boarding process
SD approach to the boarding process
 
Main
MainMain
Main
 
Main
MainMain
Main
 
Impara la Ricerca Operativa divertendoti
Impara la Ricerca Operativa divertendotiImpara la Ricerca Operativa divertendoti
Impara la Ricerca Operativa divertendoti
 
Matematica e medicina
Matematica e medicinaMatematica e medicina
Matematica e medicina
 

Recently uploaded

Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...Sérgio Sacani
 
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...Sérgio Sacani
 
Spermiogenesis or Spermateleosis or metamorphosis of spermatid
Spermiogenesis or Spermateleosis or metamorphosis of spermatidSpermiogenesis or Spermateleosis or metamorphosis of spermatid
Spermiogenesis or Spermateleosis or metamorphosis of spermatidSarthak Sekhar Mondal
 
Boyles law module in the grade 10 science
Boyles law module in the grade 10 scienceBoyles law module in the grade 10 science
Boyles law module in the grade 10 sciencefloriejanemacaya1
 
Caco-2 cell permeability assay for drug absorption
Caco-2 cell permeability assay for drug absorptionCaco-2 cell permeability assay for drug absorption
Caco-2 cell permeability assay for drug absorptionPriyansha Singh
 
A relative description on Sonoporation.pdf
A relative description on Sonoporation.pdfA relative description on Sonoporation.pdf
A relative description on Sonoporation.pdfnehabiju2046
 
Presentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptxPresentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptxgindu3009
 
Analytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptxAnalytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptxSwapnil Therkar
 
Isotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on IoIsotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on IoSérgio Sacani
 
GFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxGFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxAleenaTreesaSaji
 
Types of different blotting techniques.pptx
Types of different blotting techniques.pptxTypes of different blotting techniques.pptx
Types of different blotting techniques.pptxkhadijarafiq2012
 
Formation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disksFormation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disksSérgio Sacani
 
Cultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptxCultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptxpradhanghanshyam7136
 
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...anilsa9823
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...RohitNehra6
 
Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )aarthirajkumar25
 
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |aasikanpl
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bSérgio Sacani
 
Nanoparticles synthesis and characterization​ ​
Nanoparticles synthesis and characterization​  ​Nanoparticles synthesis and characterization​  ​
Nanoparticles synthesis and characterization​ ​kaibalyasahoo82800
 
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCRStunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCRDelhi Call girls
 

Recently uploaded (20)

Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
 
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
 
Spermiogenesis or Spermateleosis or metamorphosis of spermatid
Spermiogenesis or Spermateleosis or metamorphosis of spermatidSpermiogenesis or Spermateleosis or metamorphosis of spermatid
Spermiogenesis or Spermateleosis or metamorphosis of spermatid
 
Boyles law module in the grade 10 science
Boyles law module in the grade 10 scienceBoyles law module in the grade 10 science
Boyles law module in the grade 10 science
 
Caco-2 cell permeability assay for drug absorption
Caco-2 cell permeability assay for drug absorptionCaco-2 cell permeability assay for drug absorption
Caco-2 cell permeability assay for drug absorption
 
A relative description on Sonoporation.pdf
A relative description on Sonoporation.pdfA relative description on Sonoporation.pdf
A relative description on Sonoporation.pdf
 
Presentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptxPresentation Vikram Lander by Vedansh Gupta.pptx
Presentation Vikram Lander by Vedansh Gupta.pptx
 
Analytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptxAnalytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptx
 
Isotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on IoIsotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on Io
 
GFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxGFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptx
 
Types of different blotting techniques.pptx
Types of different blotting techniques.pptxTypes of different blotting techniques.pptx
Types of different blotting techniques.pptx
 
Formation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disksFormation of low mass protostars and their circumstellar disks
Formation of low mass protostars and their circumstellar disks
 
Cultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptxCultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptx
 
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...
 
Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )
 
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
 
Nanoparticles synthesis and characterization​ ​
Nanoparticles synthesis and characterization​  ​Nanoparticles synthesis and characterization​  ​
Nanoparticles synthesis and characterization​ ​
 
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCRStunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
 

Power Production Forecasting

  • 1. 1 Power Production and Storage in Microgrids R. De Leone, A. Giovannelli, M. Pietrini† AIRO 2014, September 2014 Research@Energy, Loccioni Group, Angeli di Rosora, AN, Italy † Mathematics Division, School of Science and Technologies, University of Camerino, MC, Italy
  • 2. Contents Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 2 Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids
  • 3. Introduction Introduction Forecasting PV model Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 3
  • 4. Forecasting PV model Introduction Forecasting PV model Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 4 Model Forecasting model for photovoltaic energy production
  • 5. Forecasting PV model Introduction Forecasting PV model Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 4 Model Forecasting model for photovoltaic energy production Aim To obtain an accurately daily forecast for a PV energy production plant located in Italy with a quarter-hour frequency
  • 6. Forecasting PV model Introduction Forecasting PV model Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 4 Model Forecasting model for photovoltaic energy production Aim To obtain an accurately daily forecast for a PV energy production plant located in Italy with a quarter-hour frequency Technique Support Vector Machines, in particular -SVR
  • 7. Forecasting PV model Introduction Forecasting PV model Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 4 Model Forecasting model for photovoltaic energy production Aim To obtain an accurately daily forecast for a PV energy production plant located in Italy with a quarter-hour frequency Technique Support Vector Machines, in particular -SVR Scenario * scientific challenge: photovoltaic production depends on weather conditions * increasing development of electrical smartgrids
  • 8. Support Vector Machines Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 5
  • 9. -SVR (1) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 6 Support Vector Machines (SVM) nonparametric technique for data classification (SVC) and regression (SVR)
  • 10. -SVR (1) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 6 Support Vector Machines (SVM) nonparametric technique for data classification (SVC) and regression (SVR) -SVR Given training data {(x1, y1), . . . , (xl, yl)}, where xi are input vectors and yi are the associated output values for xi, the support vector regression model requires the solution of the following optimization problem:
  • 11. -SVR (1) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 6 Support Vector Machines (SVM) nonparametric technique for data classification (SVC) and regression (SVR) -SVR Given training data {(x1, y1), . . . , (xl, yl)}, where xi are input vectors and yi are the associated output values for xi, the support vector regression model requires the solution of the following optimization problem: min w,b,, 1 2 wTw + C l Xi=1 i + − (+ i ) subject to yi − (wT (xi) + b) + + i , (wT (xi) + b) − yi + − i , + i , − i 0, i = 1, . . . , l.
  • 12. -SVR (2) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 7 The parameters which control the regression quality are
  • 13. -SVR (2) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 7 The parameters which control the regression quality are the cost of error C
  • 14. -SVR (2) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 7 The parameters which control the regression quality are the cost of error C the width of the tube
  • 15. -SVR (2) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 7 The parameters which control the regression quality are the cost of error C the width of the tube the mapping function
  • 16. -SVR (2) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 7 The parameters which control the regression quality are the cost of error C the width of the tube the mapping function In dual formulation of the problem, the kernel function is introduced k(xi, x) = (xi)T (x)
  • 17. -SVR (2) Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 7 The parameters which control the regression quality are the cost of error C the width of the tube the mapping function In dual formulation of the problem, the kernel function is introduced k(xi, x) = (xi)T (x) Kernel function Formulation Linear k(x, y) = xTAy Polynomial k(x, y) = (xT x + c)d Radial Basis Function (RBF) k(x, y) = e− kx−yk2 Principal kernel functions.
  • 18. -SVR Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 8 -SVR The -Support Vector Machine problem is defined as follows:
  • 19. -SVR Introduction Support Vector Machines -SVR (1) -SVR (2) -SVR Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids 8 -SVR The -Support Vector Machine problem is defined as follows: min w,b,, 1 2 wTw + C( + 1 l l Xi=1 i + − (+ i )) subject to yi − (wT (xi) + b) + + i , (wT (xi) + b) − yi + − i , + i , − i 0, i = 1, . . . , l. where the parameter allows to control the number of support vectors and training errors.
  • 20. Photovoltaic Energy Production Model Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 9
  • 21. Model data (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 10 Collected measurements: historical data of an existing solar photovoltaic plant located in Angeli di Rosora (AN, Italy), provided by the Loccioni Group. Solyndra panels positioned on the roof of an industrial facility of Loccioni Group.
  • 22. Model data (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 11 energy and power produced by the plant solar irradiance external environmental temperature
  • 23. Model data (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 11 energy and power produced by the plant solar irradiance external environmental temperature Scatterplot of correlation between solar irradiance and energy production from the PV plant.
  • 24. Model construction Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 12 -SVR training input features:
  • 25. Model construction Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 12 -SVR training input features: solar irradiance
  • 26. Model construction Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 12 -SVR training input features: solar irradiance external environmental temperature
  • 27. Model construction Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 12 -SVR training input features: solar irradiance external environmental temperature -SVR training output features:
  • 28. Model construction Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 12 -SVR training input features: solar irradiance external environmental temperature -SVR training output features: energy production
  • 29. Model construction Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 12 -SVR training input features: solar irradiance external environmental temperature -SVR training output features: energy production Parameters choice 1. kernel type: Radial Basis Function (RBF) 2. constant appearing in kernel formula: = 1/features 3. number of support vectors: = 0.5 4. cost of error: C = 1.8 5. training data: 14 days
  • 30. Prediction accuracy measures Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 13 Root Mean Square Error (RMSE) RMSE = rPn i=1(yi − ˆyi)2 n
  • 31. Prediction accuracy measures Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 13 Root Mean Square Error (RMSE) RMSE = rPn i=1(yi − ˆyi)2 n Mean Absolute Percentage Error (MAPE) MAPE = 100% n n Xi=1 | yi − ˆyi yi |
  • 32. Prediction accuracy measures Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 13 Root Mean Square Error (RMSE) RMSE = rPn i=1(yi − ˆyi)2 n Mean Absolute Percentage Error (MAPE) MAPE = 100% n n Xi=1 | yi − ˆyi yi | Coefficient of Determination (R2) R2 = Pn 2 i=1(ˆyi − ¯y)n 2 i=1(yi − ¯y)P
  • 33. Prediction accuracy measures Introduction Support Vector Machines Photovoltaic Energy Production Model Model data (1) Model data (2) Model construction Prediction accuracy measures Computational results Energy production, use and storage in smartgrids 13 Root Mean Square Error (RMSE) RMSE = rPn i=1(yi − ˆyi)2 n Mean Absolute Percentage Error (MAPE) MAPE = 100% n n Xi=1 | yi − ˆyi yi | Coefficient of Determination (R2) R2 = Pn 2 i=1(ˆyi − ¯y)n 2 i=1(yi − ¯y)Pwhere yi = actual data, ˆyi = estimated data, ¯y = mean of the actual data.
  • 34. Computational results Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 14
  • 35. Results (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 15 Comparison between real and forecasted PV energy production on 15th of June 2012.
  • 36. Results (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 15 Comparison between real and forecasted PV energy production on 15th of June 2012. RMSE = 0.5275; MAPE = 0.0785; R2 = 0.9944.
  • 37. Results (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 16 Scatterplot of correlation between real and forecasted energy production.
  • 38. Results (3) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 17 Comparison between real and forecasted PV energy production on 16th of April 2012.
  • 39. Results (3) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 17 Comparison between real and forecasted PV energy production on 16th of April 2012. RMSE = 0.2810; MAPE = 0.1143; R2 = 0.9926.
  • 40. Results (4) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids What does it happen when the input values of solar irradiance for the next day are not fully accurate? 18
  • 41. Results (4) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids What does it happen when the input values of solar irradiance for the next day are not fully accurate? 18 Comparison between real and forecasted solar irradiance on 11th of October 2013.
  • 42. Results (5) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 19 Comparison between real energy production on 11th of October 2013 and forecasted one using actual data of irradiance and temperature and estimated weather data.
  • 43. Results (5) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids 19 Comparison between real energy production on 11th of October 2013 and forecasted one using actual data of irradiance and temperature and estimated weather data. RMSE = 3.5876; MAPE = 3.5718; R2 = 0.3616.
  • 44. Observations about the model Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids accurate results compared with the three different evaluation statistics 20 (RMSE, MAPE, R2);
  • 45. Observations about the model Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids accurate results compared with the three different evaluation statistics 20 (RMSE, MAPE, R2); predictive model risk concerning with forecasted errors on input weather data provided by internet weather site;
  • 46. Observations about the model Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Results (1) Results (2) Results (3) Results (4) Results (5) Observations about the model Energy production, use and storage in smartgrids accurate results compared with the three different evaluation statistics 20 (RMSE, MAPE, R2); predictive model risk concerning with forecasted errors on input weather data provided by internet weather site; good application of -SVR model with the possibility of choosing the number of support vectors.
  • 47. Energy production, use and storage in smartgrids Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 21
  • 48. Leaf Community (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 22 The study on photovoltaic energy production forecast is part of a greater project realized by Loccioni Group: the Leaf Community smartgrid. Leaf Community smartgrid by Loccioni Group.
  • 49. Leaf Community (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 22 The study on photovoltaic energy production forecast is part of a greater project realized by Loccioni Group: the Leaf Community smartgrid.
  • 50. Leaf Community (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 23 Leaf Community smartgrid consists of
  • 51. Leaf Community (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 23 Leaf Community smartgrid consists of energy production systems from renewable sources: photovoltaic plant micro-hydroelectric power station
  • 52. Leaf Community (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 23 Leaf Community smartgrid consists of energy production systems from renewable sources: photovoltaic plant micro-hydroelectric power station consumption buildings: industrial facility office building
  • 53. Leaf Community (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 23 Leaf Community smartgrid consists of energy production systems from renewable sources: photovoltaic plant micro-hydroelectric power station consumption buildings: industrial facility office building energy storage device.
  • 54. Leaf Community (3) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 24 The smartgrid architecture scheme.
  • 55. Leaf Community System Model Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 25 The smartgrid optimization model will manage the energy flows within the network in order to ensure the maximum economic benefit achieve the goal of energy autonomy for the most of time.
  • 56. Leaf Community System Model Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 25 The smartgrid optimization model will manage the energy flows within the network in order to ensure the maximum economic benefit achieve the goal of energy autonomy for the most of time. Flow chart of the smartgrid optimization problem.
  • 57. Optimization Problem Formulation (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 26 The optimization problem is formulated as a mixed-integer linear programming problem.
  • 58. Optimization Problem Formulation (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 26 The optimization problem is formulated as a mixed-integer linear programming problem. Each component of the smartgrid is modeled separately, based on its characteristics and constraints.
  • 59. Optimization Problem Formulation (2a) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 27 Input data
  • 60. Optimization Problem Formulation (2a) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 27 Input data • photovoltaic energy production forecast Photovoltaic energy production forecast using -SVR.
  • 61. Optimization Problem Formulation (2b) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 28 Input data
  • 62. Optimization Problem Formulation (2b) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 28 Input data • energy consumption forecast Energy consumption forecast using linear regression.
  • 63. Optimization Problem Formulation (2c) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 29 Input data
  • 64. Optimization Problem Formulation (2c) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 29 Input data • electricity tariffs Selling and purchased electricity tariffs.
  • 65. Optimization Problem Formulation (3a) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 30 Problem data (t) energy purchased price at time t (e/kWh) cIND(t) energy consumption of the industrial facility at time t (kWh) cOF(t) energy consumption of the office building at time t (kWh) e(0) energy present in the battery storage at time 0 (kWh) pPV (t) photovoltaic energy production at time t (kWh) pHD(t) hydropower production at time t (kWh) (t) energy selling price at time t (e/kWh)
  • 66. Optimization Problem Formulation (3b) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 31 Problem parameters cINV energy consumption of the inverter (kWh) m minimum energy to charge/discharge the battery storage (kWh) M maximum energy to charge/discharge the battery storage (kWh) mESS minimum energy capacity of the battery storage (kWh) MESS maximum energy capacity of the battery storage (kWh) C charging efficiency of the battery storage D discharging efficiency of the battery storage T1 efficiency of the transformer T1 T2 efficiency of the transformer T2
  • 67. Optimization Problem Formulation (3c) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 32 Problem variables b1(t) energy to buy for Block 1 at time t (kWh) b2(t) energy to buy for Block 2 at time t (kWh) e(t) amount of energy present in the battery storage at time t (kWh) b(t) buying/not buying energy at time t (kWh) s(t) selling/not selling energy at time t (kWh) r12(t) residual energy from Block 1 to Block 2 at time t (kWh) s(t) energy to sell at time t (kWh) x1(t) energy sent to the battery storage from Block 1 at time t (kWh) x(t) energy sent to the battery storage from the main grid at time t (kWh) y1(t) energy to discharge from the battery storage to Block 1 at time t (kWh) y2(t) energy to discharge from the battery storage to Block 2 at time t (kWh) C(t) charging state of the battery storage at time t D(t) discharging state of the battery storage at time t
  • 68. Optimization Problem Formulation (4) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 33 Objective function 192 Xt=1 h(t)s(t)T1 − (t)b1(t) + b2(t)i Maximize the difference between the income deriving from the sold energy to the main distribution grid and the cost paid for the purchased energy, whenever the generated output power from the renewable energy sources is insufficient to cover the load demand.
  • 69. Optimization Problem Formulation (5) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 34 Constraints
  • 70. Optimization Problem Formulation (5) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 34 Constraints Energy balance constraints: energy purchased from utility grid + energy generated by the different sources = the local demand + the energy loss in the transmission lines + the energy stored in the battery.
  • 71. Optimization Problem Formulation (5) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 34 Constraints Energy balance constraints: energy purchased from utility grid + energy generated by the different sources = the local demand + the energy loss in the transmission lines + the energy stored in the battery. Battery storage constraints: capacity of the battery; battery charging/discharging range; energy level present in the battery; impossibility to simultaneously charge/discharge the storage.
  • 72. Optimization Problem Formulation (5) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 34 Constraints Energy balance constraints: energy purchased from utility grid + energy generated by the different sources = the local demand + the energy loss in the transmission lines + the energy stored in the battery. Battery storage constraints: capacity of the battery; battery charging/discharging range; energy level present in the battery; impossibility to simultaneously charge/discharge the storage. Main grid constraints: only energy produced from the renewable sources can be sold; impossibility to simultaneously purchase/sell energy.
  • 73. Leaf Community Optimization Problem: Results (1) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 35 The energy curves for two winter days with the storage fully charged.
  • 74. Leaf Community Optimization Problem: Results (2) Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 36 The energy curves for two summer days with the storage fully discharged.
  • 75. Conclusions Introduction Support Vector Machines Photovoltaic Energy Production Model Computational results Energy production, use and storage in smartgrids Leaf Community (1) Leaf Community (2) Leaf Community (3) Leaf Community System Model Optimization Problem Formulation (1) Optimization Problem Formulation (2a) Optimization Problem Formulation (2b) Optimization Problem Formulation (2c) Optimization Problem Formulation (3a) Optimization Problem Formulation (3b) Optimization Problem Formulation (3c) 37 The study on photovoltaic energy production forecast becomes an important tool in order to manage the energy flows in the smartgrid. The correctness of the forecasts is determinant for a good result of the optimization algorithm. Due to the time-varying nature of the consumptions and the variation of the battery state of charge, the optimization problem needs to respond to continuously changing operation conditions. Good results of the optimization problem are obtained.