SlideShare a Scribd company logo
1 of 103
Department of Engineering
American University of Iraq, Sulaimani (AUIS)
1
Contents
2
ARTIFICIAL
NEURAL
NETWORKS
REGRESSION
PERFORMANCE
PREDICTION
ANN FOR
RESEARCH
SAMPLE
EXERCISE
3
ARTIFICIAL
NEURAL
NETWORKS
 GENERAL FRAMEWORK OF AI
 ARTIFICIAL INTELLIGENCE
 MACHINE LEARNING
 ARTIFICIAL NEURAL NETWORKS
 DEEP LEARNING
4
ARTIFICIAL
NEURAL
NETWORKS
5
Image Source: aiso-lab.com
WHAT – WHY - HOW
6
Artificial Intelligence
7
Artificial Intelligence
 What is AI ?
Branch of computer science that focuses on creating systems that
are intelligent and independent.
 Why use AI ?
To enhances the speed, precision and effectiveness of human
efforts.
 How is AI implemented ?
AI can be developed through learning that is symbolic-based or
data-based.
8
Machine learning
 What is Machine Learning ?
Using learning paradigms in machines to achieve pattern
recognition and other human functions.
 Why use Machine Learning ?
Machines can utilize more data, and dimensions of data, to perform
functions such as pattern recognition.
 How is Machine Learning developed ?
Machines learn through either statistical learning or deep
learning.
9
The State-of-AI
10
Text AI
Visual AI
Interactive AI
Functional AI
Analytic AI
text recognition, speech-to-text conversion
machine translation, and content generation capabilities.
Chatbots and smart personal assistants.
Computer vision or augmented reality
Sentiment analysis and supplier risk assessment
Maintenance of breakdown in machines after spotting errors, etc.
Examples:
Alex Bekker. 5 Types of AI to Propel Your Business. Science soft.
Link: https://www.scnsoft.com/blog/artificial-intelligence-types
11
Source: Alex Bekker,ScienceSoft. link: https://www.scnsoft.com/blog/artificial-intelligence-types
The State-of-AI
The State-of-AI
12
Symbolic learning
Machine learning Pattern recognition
Computer vision
Neural Networks Deep Learning
learning capabilities (or) reasoning capabilities
What about combining the two ?
Able to give explanations and can manipulate
complex data structures
Richa Bhatia. Understanding the difference between symbolic AI & non symbolic AI.
https://analyticsindiamag.com/understanding-difference-symbolic-ai-non-symbolic-ai/
13
Symbolic learning
Symbolic learning
See
Process
Computer vision
Robotics
Interact
Symbolic
learning
14
Human ability to
interact with the world
System that
can learn
Symbolic learning
Machine learning Pattern recognition
Computer vision
Neural Networks Deep Learning
Machine learning
15
Robotic system
Sensor
Actuators
Non-AI
programming
AI
algorithm
AI and Robotics
ARTIFICIAL NEURAL NETWORKS
 What are Artificial Neural Networks ?
 Why use Artificial Neural Networks ?
 How are Artificial Neural Networks developed ?
Answers in the presentation > slides 15-41
16
Deep Learning
 What is Deep Learning ?
 Why use Deep Learning ?
 How is Deep Learning achieved ?
Answers in the presentation > slide 42
17
Artificial Neural Networks (ANN)
C
18
Artificial Neural Networks (ANN)
Motivation
19
Artificial Neural Networks (ANN)
Advantages
20
Artificial Neural Networks (ANN)
Components
21
Artificial Neural Networks (ANN)
C
22
Artificial Neural Networks (ANN)
C
23
Artificial Neural Networks (ANN)
C
24
Artificial Neural Networks (ANN)
Numerical
Data
Labeling
Images
Video
Text
Time-series
Clustering
Machine
perception
Classification Clustering
C
25
Artificial Neural Networks (ANN)
Classification
SUPERVISED LEARNING
Face detection
Object
Identification
Text
classification
Voice detection
Gesture
Identification
Transcript
speech-text
Create
Numerical
Data
Assign
to
data
TRAIN
THE
ANN
Create
correlation
26
Artificial Neural Networks (ANN)
Ni, D. X. (2007). Application of neural networks to character recognition. Proceedings
of students/faculty research day, CSIS, Pace University, May 4th.
Classification
C
27
Artificial Neural Networks (ANN)
Classification
https://www.analyticsindiamag.com/how-to-create-your-first-artificial-neural-network-in-python/ C
28
Artificial Neural Networks (ANN)
Regression
SUPERVISED LEARNING
Hardware
breakdown
Employee
turnover
Customer churn
Health
breakdown
Machine output
C
29
Artificial Neural Networks (ANN)
Regression
https://dataaspirant.com/2017/03/02/how-logistic-regression-model-works/ C
30
Artificial Neural Networks (ANN)
Regression
https://dataaspirant.com/2017/03/02/how-logistic-regression-model-works/ C
31
Artificial Neural Networks (ANN)
Regression
https://dataaspirant.com/2017/03/02/how-logistic-regression-model-works/ C
32
Artificial Neural Networks (ANN)
Clustering
UNSUPERVISED
LEARNING
Comparing documents, images, etc.
to surface similar items
Detecting anomalies, or unusual
behavior
C
33
Artificial Neural Networks (ANN)
Clustering
C
34
Artificial Neural Networks (ANN)
35
https://quantumcomputingtech.blogspot.com/2019/02/machine-learning-animated-gif.html
Clustering
Artificial Neural Networks (ANN)
REINFORCEMENT
Self-driving car
Customer management
Adaptive traffic signal
control
Games
Reinforcement
LEARNING
C
36
37
Artificial Neural Networks (ANN)
REINFORCEMENT
This paradigm does not require a labelled input/output pairs to be presented and
does not need sub-optimal actions to be explicitly corrected.
Artificial Neural Networks (ANN)
Learning paradigms
SUPERVISED LEARNING
UNSUPERVISED LEARNING
REINFORCEMENT
LEARNING
38
Artificial Neural Networks (ANN)
Learning paradigms
Kaplan, S. (2017). Deep generative models for synthetic retinal image generation. C
39
Artificial Neural Networks (ANN)
Learning paradigms
Unsupervised learning Supervised learning Clustering: An Introduction to Supervised Machine Learning and Pattern
Classification: The Big Picture by Sebastian Raschka C
40
Artificial Neural Networks (ANN)
SINGLE LAYER
C
41
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
C
42
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
C
43
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
C
44
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
Jain & Mao, 1996 C
Deep
45
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
A MLP consists of at least three layers of nodes: an input layer, a hidden layer and an
output layer. Except for the input nodes, each node is a neuron that uses a
nonlinear activation function.
MLP utilizes a supervised learning technique called backpropagation for training. Its
multiple layers and non-linear activation distinguish MLP from a linear perceptron. It can
distinguish data that is not linearly separable.
46
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
Changing connection weights after each piece of data is processed, based on the
amount of error in the output compared to the expected result.
47
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
MLPs are useful in research for their ability to solve problems stochastically, which
often allows approximate solutions for extremely complex problems like fitness
approximation.
MLPs are universal function approximators as shown by Cybenko's theorem, so they can
be used to create mathematical models by regression analysis. As classification is a
particular case of regression when the response variable is categorical, MLPs make good
classifier algorithms.
MLPs were a popular machine learning solution in the 1980s, finding applications in
diverse fields such as speech recognition, image recognition, and machine
translation software, but thereafter faced strong competition from much simpler (and
related) support vector machines. Interest in backpropagation networks returned due to
the successes of deep learning.
48
Artificial Neural Networks (ANN)
MULTIPLE LAYER PERCEPTRON (MLP)
49
Artificial Neural Networks (ANN)
FEED-FORWARD
Connections between the nodes do not form a cycle.
The first and simplest type of artificial neural network
devised.
50
Artificial Neural Networks (ANN)
BACKPROPAGATION
51
Artificial Neural Networks (ANN)
BACKPROPAGATION
52
Artificial Neural Networks (ANN)
BACKPROPAGATION
 Supervised learning
 Used at each layer to minimize the
error between the layer’s response
and the actual data
 The error at each hidden layer is an
average of the evaluated error
53
Artificial Neural Networks (ANN)
BACKPROPAGATION
 N is a neuron.
 Nw is one of N’s inputs weights
 Nout is N’s output.
 Nw = Nw +Δ Nw
 Δ Nw = Nout * (1‐ Nout)* NErrorFactor
 NErrorFactor = NExpectedOutput – NActualOutput
 This works only for the last layer, as we can know the actual output, and
the expected output.
54
Artificial Neural Networks (ANN)
Training stage
Validating stage
Testing stage
55
Contents
56
ARTIFICIAL
NEURAL
NETWORKS
REGRESSION
PERFORMANCE
PREDICTION
ANN FOR
RESEARCH
SAMPLE
EXERCISE
57
REGRESSION
A measure of the relation between the mean value of one variable (e.g. output)
and corresponding values of other variables (e.g. time and cost).
Dependent
variable
Independent
variables
58
REGRESSION
Dependent variable
Independent variables
Control variables
Regression and Correlation
59
Contents
60
ARTIFICIAL
NEURAL
NETWORKS
REGRESSION
PERFORMANCE
PREDICTION
ANN FOR
RESEARCH
SAMPLE
EXERCISE
PERFORMANCE
PREDICTION
Methodology
Performance
Prediction
Statistical
methods
Analysis
61
PV/T modeling and performance
prediction
• Why use artificial neural networks (ANN) in solar energy
technologies?
• How to use artificial neural networks (ANN) in solar energy
technologies?
• How to utilize regression and classification to invest in solar energy?
62
PV performance time-series prediction
Short-term power
forecasting
Estimating power loss
due to environment
Estimating the energetic
performance of PV/T
PV fault detection
Optimize PV array
inclination
Solar Energy Forecasting
Optimize maximum
power point tracking
63
PV performance time-series prediction
C
Senkal, O. & Kuleli, T. (2009). Estimation of solar radiation over Turkey using
artificial neural network and satellite data. Applied Energy, Vol. 86, pp. 1222–
64
PV performance time-series prediction
Almonacid, F., Rus, C., Hontoria, L., Fuentes, M. &
Nofuentes G. (2009). Characterisation of Si-crystalline PV
modules by artificial neural Networks. Renewable Energy, C
65
PV performance time-series prediction
Firefly algorithm (FA) and Particle Swarm Optimization (PSO) applied
to train NN
EXAMPLE 1: SHORT TERM POWER
FORECASTING
Demirdelen, T., Aksu, I. O., Esenboga, B., Aygul, K.,
Ekinci, F., & Bilgili, M. (2019). A new method for
generating short-term power forecasting based on artificial
neural networks and optimization methods for solar
photovoltaic power plants. In Solar Photovoltaic Power
Plants (pp. 165-189). Springer, Singapore.
Multi-Layer Feed Forward (MLFF) neural network structure
The data of 1MWPV power plant in Turkey is used to estimate output power
by
real-time data mining for short time prediction. 66
PV performance time-series prediction
Example 1
C
67
PV performance time-series prediction
Example 1
Ambient temperature [◦C]
Solar radiation [W/m2]
PV Panel temperature [◦C]
Input
Layer
Hidden
Layer
Output
Layer
24 8
Interconnections Interconnections
PV power (W)
68
PV/T performance time-series prediction
Particle Swarm Optimization (PSO) applied to train NN
EXAMPLE 2: ESTIMATING ENERGETIC
PERFORMANCE
Alnaqi, A. A., Moayedi, H., Shahsavar, A., & Nguyen, T.
K. (2019). Prediction of energetic performance of a
building integrated photovoltaic/thermal system thorough
artificial neural network and hybrid particle swarm
optimization models. Energy conversion and
management, 183, 137-148.
Multi-Layer Feed-Forward Back-Propagation (FFBP) neural network
structure
building integrated photovoltaic/thermal system
69
PV/T performance time-series prediction
Example 2
C
70
PV/T performance time-series prediction
Example 2
C
71
Statistical methods
• Coefficient of determination
• Mean Absolute Percentage Error
• Root Mean Square Error
• Mean Absolute Error
• Mean Square Error
difference between two continuous variables
how well a regression model is capable of describing a
data set
a measure of prediction accuracy of a forecasting
method in statistics,
a measure of the quality of an estimator
measure of the differences between
an estimator and the values observed
72
Statistical methods
COEFFICIENT OF DETERMINATION
a measure for how well a regression model is capable of
describing a data set.
used to evaluate the validity of the predictive
results compared to the actual (experimental)
real model results
𝒚𝒊= experimental value of (y)
𝒇𝒊= predicted value of (y)
𝒚𝒊= the mean of the experimental values
N = the number of values
73
Statistical methods
MEAN ABSOLUTE ERROR
a measure of difference between two continuous variables.
the average vertical distance between each point and the
identity line. MAE is also the average horizontal distance
between each point and the identity line.
74
Statistical methods
MEAN ABSOLUTE PERCENTAGE ERROR
a measure of prediction accuracy of a forecasting
method in statistics,
75
Statistical methods
MEAN SQUARE ERROR
measures the average of the squares of the errors—that is, the average
squared difference between the estimated values and what is estimated
The MSE is a measure of the quality of an estimator—it is always non-
negative, and values closer to zero are better.
76
Statistical methods
ROOT MEAN SQUARE ERROR
a measure for the difference between a data set and a corresponding fit. The RMS
asymptotically converges to the standard deviation from the model's predicted value
for sufficiently large sizes of data sets.
measure of the differences between values (sample or population values)
predicted by a model or an estimator and the values observed.
77
Applications in engineering disciplines
Civil engineering
78
Suryanita, R. (2016). The application of artificial neural
networks in predicting structural response of multistory
building in the region of Sumatra island. KnE Engineering.
Reinforced Concrete Building Models
Applications in engineering disciplines
Electrical engineering
79
Hiyama, T., Kouzuma, S., Imakubo,
T.:‘Identification of optimal operating pointof PV
modules using neural network for real time
maximum power trackingcontrol’,IEEE Trans.
Energy Convers., 1995,10, (2), pp. 360–367
Applications in engineering disciplines
Mechanical engineering
80
Ranasinghe, R. A. T. M., Jaksa, M. B., Kuo, Y. L., & Nejad, F. P. (2017). Application of artificial neural networks for
predicting the impact of rolling dynamic compaction using dynamic cone penetrometer test results. Journal of Rock
Mechanics and Geotechnical Engineering, 9(2), 340-349.
Predicting the impact
of rolling dynamic
compaction using
dynamic cone
penetrometer test
results
Applications in engineering disciplines
Control engineering
81
KACHULKOVA, S. Applications of Artificial Neural Networks in Control Systems. transfer, 1(1), 5.
82
ARTIFICIAL
NEURAL
NETWORKS
REGRESSION
PERFORMANCE
PREDICTION
ANN FOR
RESEARCH
SAMPLE
EXERCISE
Tips to start utilizing machine learning and
ANN techniques for your previous,
current and future studies in your field…
PAST – PRESENT - FUTURE
83
ANN FOR
RESEARCH
84
TIP #1 CLASSIFY THE VARIABLES
DEPENDENT
VARIABLES (DV)
INDEPENDENT
VARIABLES (IV)
CONTROL
VARIABLES (CV)
85
TIP #2 ENSURE HIGH-QUALITY OF DATA
GIGO
GARBAGE IN =
GARBAGE OUT
GIGO, VIGO, VIVO
86
87
TIP #2 ENSURE HIGH-QUALITY OF DATA
WHAT IS YOUR UNCERTAINTY
THRESHOLD?
WHAT MODEL ARE YOU USING FOR
UNCERTAINTY ANALYSIS?
WHAT IS YOUR UNCERTAINTY
THRESHOLD?
88
Example of uncertainty model and
analysis
The uncertainty associated with the result is WR and those associated
with the independent variables are W1, W2, …, Wn.
Source: Kline, S. J. (1953). Describing uncertainty in single sample experiments. Mech.
Engineering, 75, 3-8.
89
Example of uncertainty model and
analysis
Source: Sarafraz, M. M., Safaei, M. R., Leon, A. S., Tlili, I., Alkanhal, T. A., Tian, Z., ... & Arjomandi, M. (2019).
Experimental investigation on thermal performance of a PV/T-PCM (photovoltaic/thermal) system cooling with a PCM
and nanofluid. Energies, 12(13), 2572.
90
Data filtration
Height of an adult male
by fixing the typo or possibly remeasuring the item or
person. If that’s not possible, you must delete the data
point because you know it’s an incorrect value.
Source: Statistics by Jim. Link: https://statisticsbyjim.com/basics/remove-outliers/ (access
date: 25th of July 2022)
91
COLLECT AS MUCH DATA AS
POSSIBLE.
TIP #3 COLLECT BULK DATA
Once testing station, or setup, is established.
Record as much data as possible. With a
variety of design configurations.
92
Number of input, outputs and hidden
nodes.
TIP #4 Select the components
Utilize the identified variables (IV, DV, CV)
93
How much (%) of the data are you allocating for
training? Validating? Testing?
TIP #5 Choose the data splitting
percentage
94
Use MATLAB/Python to create a simple
MLP Neural Network to process the data
TIP #6 Start with a simple MLP ANN
95
ARTIFICIAL
NEURAL
NETWORKS
REGRESSION
PERFORMANCE
PREDICTION
ANN FOR
RESEARCH
SAMPLE
EXERCISE
96
Simple exercise
Perform tasks 1-5
Food for Thought
• What is the difference between Artificial Intelligence and
computational intelligence?
• How to interpret skewness and kurtosis?
• What is the significance of the activation function?
• What is the significance of the bias?
97
 Identify the dependent and independent variables in your work.
 Assign the input and output nodes to your network.
 Conduct data pre-processing.
 Decide the type of problem you have.
 Choose an activation function.
 Choose the number of hidden nodes.
 Assign weights and biases.
 Download MATLAB and implement a simple MLP code.
 Carryout the training, validation and testing.
Homework
98
Textbook in PV/T
https://www.amazon.com/Photovoltaic-Thermal-Systems-Principles-
Applications/dp/3030278239/ref=sr_1_1?keywords=PV%2FT+al-waeli&qid=1563014343&s=gateway&sr=8-1
99
Review article in PV/T
100
Original article on ANN use for PV/T
Performance Prediction
101
It’s play time…
• https://playground.tensorflow.org/
102
Thank you
103

More Related Content

Similar to ANN for Solar Energy Performance Prediction

RNN basics in deep learning
RNN basics in deep learningRNN basics in deep learning
RNN basics in deep learningchauhankapil
 
Dr. Syed Muhammad Ali Tirmizi - Special topics in finance lec 14
Dr. Syed Muhammad Ali Tirmizi - Special topics in finance   lec 14Dr. Syed Muhammad Ali Tirmizi - Special topics in finance   lec 14
Dr. Syed Muhammad Ali Tirmizi - Special topics in finance lec 14Dr. Muhammad Ali Tirmizi., Ph.D.
 
Artificial Neural Network for hand Gesture recognition
Artificial Neural Network for hand Gesture recognitionArtificial Neural Network for hand Gesture recognition
Artificial Neural Network for hand Gesture recognitionVigneshwer Dhinakaran
 
NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...
NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...
NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...ijaia
 
Student intervention detection using deep learning technique
Student intervention detection using deep learning techniqueStudent intervention detection using deep learning technique
Student intervention detection using deep learning techniqueVenkat Projects
 
Neural network
Neural network Neural network
Neural network Faireen
 
Artificial Neural Networks: Applications In Management
Artificial Neural Networks: Applications In ManagementArtificial Neural Networks: Applications In Management
Artificial Neural Networks: Applications In ManagementIOSR Journals
 
Ann model and its application
Ann model and its applicationAnn model and its application
Ann model and its applicationmilan107
 
Artificial Neural Network Implementation On FPGA Chip
Artificial Neural Network Implementation On FPGA ChipArtificial Neural Network Implementation On FPGA Chip
Artificial Neural Network Implementation On FPGA ChipMaria Perkins
 
Student intervention detection using deep learning technique
Student intervention detection using deep learning techniqueStudent intervention detection using deep learning technique
Student intervention detection using deep learning techniqueVenkat Projects
 
Intro to Deep learning - Autoencoders
Intro to Deep learning - Autoencoders Intro to Deep learning - Autoencoders
Intro to Deep learning - Autoencoders Akash Goel
 
Artificial Neural Network ANN
Artificial Neural Network ANNArtificial Neural Network ANN
Artificial Neural Network ANNAbdullah al Mamun
 

Similar to ANN for Solar Energy Performance Prediction (20)

Artificial Neural networks
Artificial Neural networksArtificial Neural networks
Artificial Neural networks
 
RNN basics in deep learning
RNN basics in deep learningRNN basics in deep learning
RNN basics in deep learning
 
Project Report -Vaibhav
Project Report -VaibhavProject Report -Vaibhav
Project Report -Vaibhav
 
Dr. Syed Muhammad Ali Tirmizi - Special topics in finance lec 14
Dr. Syed Muhammad Ali Tirmizi - Special topics in finance   lec 14Dr. Syed Muhammad Ali Tirmizi - Special topics in finance   lec 14
Dr. Syed Muhammad Ali Tirmizi - Special topics in finance lec 14
 
Artificial Neural Network for hand Gesture recognition
Artificial Neural Network for hand Gesture recognitionArtificial Neural Network for hand Gesture recognition
Artificial Neural Network for hand Gesture recognition
 
NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...
NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...
NETWORK LEARNING AND TRAINING OF A CASCADED LINK-BASED FEED FORWARD NEURAL NE...
 
Student intervention detection using deep learning technique
Student intervention detection using deep learning techniqueStudent intervention detection using deep learning technique
Student intervention detection using deep learning technique
 
Neural network
Neural network Neural network
Neural network
 
Artificial Neural Networks: Applications In Management
Artificial Neural Networks: Applications In ManagementArtificial Neural Networks: Applications In Management
Artificial Neural Networks: Applications In Management
 
Ann model and its application
Ann model and its applicationAnn model and its application
Ann model and its application
 
Artificial Neural Network Implementation On FPGA Chip
Artificial Neural Network Implementation On FPGA ChipArtificial Neural Network Implementation On FPGA Chip
Artificial Neural Network Implementation On FPGA Chip
 
Student intervention detection using deep learning technique
Student intervention detection using deep learning techniqueStudent intervention detection using deep learning technique
Student intervention detection using deep learning technique
 
Neural Network
Neural NetworkNeural Network
Neural Network
 
E44082429
E44082429E44082429
E44082429
 
Jack
JackJack
Jack
 
40120140507007
4012014050700740120140507007
40120140507007
 
40120140507007
4012014050700740120140507007
40120140507007
 
Neural Networks
Neural NetworksNeural Networks
Neural Networks
 
Intro to Deep learning - Autoencoders
Intro to Deep learning - Autoencoders Intro to Deep learning - Autoencoders
Intro to Deep learning - Autoencoders
 
Artificial Neural Network ANN
Artificial Neural Network ANNArtificial Neural Network ANN
Artificial Neural Network ANN
 

Recently uploaded

Microscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxMicroscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxpurnimasatapathy1234
 
Introduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptxIntroduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptxupamatechverse
 
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...ZTE
 
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...RajaP95
 
VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130
VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130
VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130Suhani Kapoor
 
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130Suhani Kapoor
 
Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxupamatechverse
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escortsranjana rawat
 
SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )Tsuyoshi Horigome
 
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...ranjana rawat
 
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).pptssuser5c9d4b1
 
Call Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile serviceCall Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile servicerehmti665
 
Analog to Digital and Digital to Analog Converter
Analog to Digital and Digital to Analog ConverterAnalog to Digital and Digital to Analog Converter
Analog to Digital and Digital to Analog ConverterAbhinavSharma374939
 
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...ranjana rawat
 
High Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
Biology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptxBiology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptxDeepakSakkari2
 
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxDecoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxJoão Esperancinha
 

Recently uploaded (20)

Microscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxMicroscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptx
 
Introduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptxIntroduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptx
 
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
ZXCTN 5804 / ZTE PTN / ZTE POTN / ZTE 5804 PTN / ZTE POTN 5804 ( 100/200 GE Z...
 
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
 
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
 
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
 
VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130
VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130
VIP Call Girls Service Hitech City Hyderabad Call +91-8250192130
 
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
 
Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptx
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
 
SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )
 
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
 
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
 
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
 
Call Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile serviceCall Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile service
 
Analog to Digital and Digital to Analog Converter
Analog to Digital and Digital to Analog ConverterAnalog to Digital and Digital to Analog Converter
Analog to Digital and Digital to Analog Converter
 
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
 
High Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Meera Call 7001035870 Meet With Nagpur Escorts
 
Biology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptxBiology for Computer Engineers Course Handout.pptx
Biology for Computer Engineers Course Handout.pptx
 
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxDecoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
 

ANN for Solar Energy Performance Prediction

  • 1. Department of Engineering American University of Iraq, Sulaimani (AUIS) 1
  • 4.  GENERAL FRAMEWORK OF AI  ARTIFICIAL INTELLIGENCE  MACHINE LEARNING  ARTIFICIAL NEURAL NETWORKS  DEEP LEARNING 4 ARTIFICIAL NEURAL NETWORKS
  • 6. WHAT – WHY - HOW 6
  • 8. Artificial Intelligence  What is AI ? Branch of computer science that focuses on creating systems that are intelligent and independent.  Why use AI ? To enhances the speed, precision and effectiveness of human efforts.  How is AI implemented ? AI can be developed through learning that is symbolic-based or data-based. 8
  • 9. Machine learning  What is Machine Learning ? Using learning paradigms in machines to achieve pattern recognition and other human functions.  Why use Machine Learning ? Machines can utilize more data, and dimensions of data, to perform functions such as pattern recognition.  How is Machine Learning developed ? Machines learn through either statistical learning or deep learning. 9
  • 10. The State-of-AI 10 Text AI Visual AI Interactive AI Functional AI Analytic AI text recognition, speech-to-text conversion machine translation, and content generation capabilities. Chatbots and smart personal assistants. Computer vision or augmented reality Sentiment analysis and supplier risk assessment Maintenance of breakdown in machines after spotting errors, etc. Examples: Alex Bekker. 5 Types of AI to Propel Your Business. Science soft. Link: https://www.scnsoft.com/blog/artificial-intelligence-types
  • 11. 11 Source: Alex Bekker,ScienceSoft. link: https://www.scnsoft.com/blog/artificial-intelligence-types The State-of-AI
  • 12. The State-of-AI 12 Symbolic learning Machine learning Pattern recognition Computer vision Neural Networks Deep Learning learning capabilities (or) reasoning capabilities What about combining the two ? Able to give explanations and can manipulate complex data structures Richa Bhatia. Understanding the difference between symbolic AI & non symbolic AI. https://analyticsindiamag.com/understanding-difference-symbolic-ai-non-symbolic-ai/
  • 13. 13 Symbolic learning Symbolic learning See Process Computer vision Robotics Interact Symbolic learning
  • 14. 14 Human ability to interact with the world System that can learn Symbolic learning Machine learning Pattern recognition Computer vision Neural Networks Deep Learning Machine learning
  • 16. ARTIFICIAL NEURAL NETWORKS  What are Artificial Neural Networks ?  Why use Artificial Neural Networks ?  How are Artificial Neural Networks developed ? Answers in the presentation > slides 15-41 16
  • 17. Deep Learning  What is Deep Learning ?  Why use Deep Learning ?  How is Deep Learning achieved ? Answers in the presentation > slide 42 17
  • 19. Artificial Neural Networks (ANN) Motivation 19
  • 20. Artificial Neural Networks (ANN) Advantages 20
  • 21. Artificial Neural Networks (ANN) Components 21
  • 25. Artificial Neural Networks (ANN) Numerical Data Labeling Images Video Text Time-series Clustering Machine perception Classification Clustering C 25
  • 26. Artificial Neural Networks (ANN) Classification SUPERVISED LEARNING Face detection Object Identification Text classification Voice detection Gesture Identification Transcript speech-text Create Numerical Data Assign to data TRAIN THE ANN Create correlation 26
  • 27. Artificial Neural Networks (ANN) Ni, D. X. (2007). Application of neural networks to character recognition. Proceedings of students/faculty research day, CSIS, Pace University, May 4th. Classification C 27
  • 28. Artificial Neural Networks (ANN) Classification https://www.analyticsindiamag.com/how-to-create-your-first-artificial-neural-network-in-python/ C 28
  • 29. Artificial Neural Networks (ANN) Regression SUPERVISED LEARNING Hardware breakdown Employee turnover Customer churn Health breakdown Machine output C 29
  • 30. Artificial Neural Networks (ANN) Regression https://dataaspirant.com/2017/03/02/how-logistic-regression-model-works/ C 30
  • 31. Artificial Neural Networks (ANN) Regression https://dataaspirant.com/2017/03/02/how-logistic-regression-model-works/ C 31
  • 32. Artificial Neural Networks (ANN) Regression https://dataaspirant.com/2017/03/02/how-logistic-regression-model-works/ C 32
  • 33. Artificial Neural Networks (ANN) Clustering UNSUPERVISED LEARNING Comparing documents, images, etc. to surface similar items Detecting anomalies, or unusual behavior C 33
  • 34. Artificial Neural Networks (ANN) Clustering C 34
  • 35. Artificial Neural Networks (ANN) 35 https://quantumcomputingtech.blogspot.com/2019/02/machine-learning-animated-gif.html Clustering
  • 36. Artificial Neural Networks (ANN) REINFORCEMENT Self-driving car Customer management Adaptive traffic signal control Games Reinforcement LEARNING C 36
  • 37. 37 Artificial Neural Networks (ANN) REINFORCEMENT This paradigm does not require a labelled input/output pairs to be presented and does not need sub-optimal actions to be explicitly corrected.
  • 38. Artificial Neural Networks (ANN) Learning paradigms SUPERVISED LEARNING UNSUPERVISED LEARNING REINFORCEMENT LEARNING 38
  • 39. Artificial Neural Networks (ANN) Learning paradigms Kaplan, S. (2017). Deep generative models for synthetic retinal image generation. C 39
  • 40. Artificial Neural Networks (ANN) Learning paradigms Unsupervised learning Supervised learning Clustering: An Introduction to Supervised Machine Learning and Pattern Classification: The Big Picture by Sebastian Raschka C 40
  • 41. Artificial Neural Networks (ANN) SINGLE LAYER C 41
  • 42. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) C 42
  • 43. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) C 43
  • 44. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) C 44
  • 45. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) Jain & Mao, 1996 C Deep 45
  • 46. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) A MLP consists of at least three layers of nodes: an input layer, a hidden layer and an output layer. Except for the input nodes, each node is a neuron that uses a nonlinear activation function. MLP utilizes a supervised learning technique called backpropagation for training. Its multiple layers and non-linear activation distinguish MLP from a linear perceptron. It can distinguish data that is not linearly separable. 46
  • 47. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) Changing connection weights after each piece of data is processed, based on the amount of error in the output compared to the expected result. 47
  • 48. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) MLPs are useful in research for their ability to solve problems stochastically, which often allows approximate solutions for extremely complex problems like fitness approximation. MLPs are universal function approximators as shown by Cybenko's theorem, so they can be used to create mathematical models by regression analysis. As classification is a particular case of regression when the response variable is categorical, MLPs make good classifier algorithms. MLPs were a popular machine learning solution in the 1980s, finding applications in diverse fields such as speech recognition, image recognition, and machine translation software, but thereafter faced strong competition from much simpler (and related) support vector machines. Interest in backpropagation networks returned due to the successes of deep learning. 48
  • 49. Artificial Neural Networks (ANN) MULTIPLE LAYER PERCEPTRON (MLP) 49
  • 50. Artificial Neural Networks (ANN) FEED-FORWARD Connections between the nodes do not form a cycle. The first and simplest type of artificial neural network devised. 50
  • 51. Artificial Neural Networks (ANN) BACKPROPAGATION 51
  • 52. Artificial Neural Networks (ANN) BACKPROPAGATION 52
  • 53. Artificial Neural Networks (ANN) BACKPROPAGATION  Supervised learning  Used at each layer to minimize the error between the layer’s response and the actual data  The error at each hidden layer is an average of the evaluated error 53
  • 54. Artificial Neural Networks (ANN) BACKPROPAGATION  N is a neuron.  Nw is one of N’s inputs weights  Nout is N’s output.  Nw = Nw +Δ Nw  Δ Nw = Nout * (1‐ Nout)* NErrorFactor  NErrorFactor = NExpectedOutput – NActualOutput  This works only for the last layer, as we can know the actual output, and the expected output. 54
  • 55. Artificial Neural Networks (ANN) Training stage Validating stage Testing stage 55
  • 57. 57 REGRESSION A measure of the relation between the mean value of one variable (e.g. output) and corresponding values of other variables (e.g. time and cost). Dependent variable Independent variables
  • 62. PV/T modeling and performance prediction • Why use artificial neural networks (ANN) in solar energy technologies? • How to use artificial neural networks (ANN) in solar energy technologies? • How to utilize regression and classification to invest in solar energy? 62
  • 63. PV performance time-series prediction Short-term power forecasting Estimating power loss due to environment Estimating the energetic performance of PV/T PV fault detection Optimize PV array inclination Solar Energy Forecasting Optimize maximum power point tracking 63
  • 64. PV performance time-series prediction C Senkal, O. & Kuleli, T. (2009). Estimation of solar radiation over Turkey using artificial neural network and satellite data. Applied Energy, Vol. 86, pp. 1222– 64
  • 65. PV performance time-series prediction Almonacid, F., Rus, C., Hontoria, L., Fuentes, M. & Nofuentes G. (2009). Characterisation of Si-crystalline PV modules by artificial neural Networks. Renewable Energy, C 65
  • 66. PV performance time-series prediction Firefly algorithm (FA) and Particle Swarm Optimization (PSO) applied to train NN EXAMPLE 1: SHORT TERM POWER FORECASTING Demirdelen, T., Aksu, I. O., Esenboga, B., Aygul, K., Ekinci, F., & Bilgili, M. (2019). A new method for generating short-term power forecasting based on artificial neural networks and optimization methods for solar photovoltaic power plants. In Solar Photovoltaic Power Plants (pp. 165-189). Springer, Singapore. Multi-Layer Feed Forward (MLFF) neural network structure The data of 1MWPV power plant in Turkey is used to estimate output power by real-time data mining for short time prediction. 66
  • 67. PV performance time-series prediction Example 1 C 67
  • 68. PV performance time-series prediction Example 1 Ambient temperature [◦C] Solar radiation [W/m2] PV Panel temperature [◦C] Input Layer Hidden Layer Output Layer 24 8 Interconnections Interconnections PV power (W) 68
  • 69. PV/T performance time-series prediction Particle Swarm Optimization (PSO) applied to train NN EXAMPLE 2: ESTIMATING ENERGETIC PERFORMANCE Alnaqi, A. A., Moayedi, H., Shahsavar, A., & Nguyen, T. K. (2019). Prediction of energetic performance of a building integrated photovoltaic/thermal system thorough artificial neural network and hybrid particle swarm optimization models. Energy conversion and management, 183, 137-148. Multi-Layer Feed-Forward Back-Propagation (FFBP) neural network structure building integrated photovoltaic/thermal system 69
  • 70. PV/T performance time-series prediction Example 2 C 70
  • 71. PV/T performance time-series prediction Example 2 C 71
  • 72. Statistical methods • Coefficient of determination • Mean Absolute Percentage Error • Root Mean Square Error • Mean Absolute Error • Mean Square Error difference between two continuous variables how well a regression model is capable of describing a data set a measure of prediction accuracy of a forecasting method in statistics, a measure of the quality of an estimator measure of the differences between an estimator and the values observed 72
  • 73. Statistical methods COEFFICIENT OF DETERMINATION a measure for how well a regression model is capable of describing a data set. used to evaluate the validity of the predictive results compared to the actual (experimental) real model results 𝒚𝒊= experimental value of (y) 𝒇𝒊= predicted value of (y) 𝒚𝒊= the mean of the experimental values N = the number of values 73
  • 74. Statistical methods MEAN ABSOLUTE ERROR a measure of difference between two continuous variables. the average vertical distance between each point and the identity line. MAE is also the average horizontal distance between each point and the identity line. 74
  • 75. Statistical methods MEAN ABSOLUTE PERCENTAGE ERROR a measure of prediction accuracy of a forecasting method in statistics, 75
  • 76. Statistical methods MEAN SQUARE ERROR measures the average of the squares of the errors—that is, the average squared difference between the estimated values and what is estimated The MSE is a measure of the quality of an estimator—it is always non- negative, and values closer to zero are better. 76
  • 77. Statistical methods ROOT MEAN SQUARE ERROR a measure for the difference between a data set and a corresponding fit. The RMS asymptotically converges to the standard deviation from the model's predicted value for sufficiently large sizes of data sets. measure of the differences between values (sample or population values) predicted by a model or an estimator and the values observed. 77
  • 78. Applications in engineering disciplines Civil engineering 78 Suryanita, R. (2016). The application of artificial neural networks in predicting structural response of multistory building in the region of Sumatra island. KnE Engineering. Reinforced Concrete Building Models
  • 79. Applications in engineering disciplines Electrical engineering 79 Hiyama, T., Kouzuma, S., Imakubo, T.:‘Identification of optimal operating pointof PV modules using neural network for real time maximum power trackingcontrol’,IEEE Trans. Energy Convers., 1995,10, (2), pp. 360–367
  • 80. Applications in engineering disciplines Mechanical engineering 80 Ranasinghe, R. A. T. M., Jaksa, M. B., Kuo, Y. L., & Nejad, F. P. (2017). Application of artificial neural networks for predicting the impact of rolling dynamic compaction using dynamic cone penetrometer test results. Journal of Rock Mechanics and Geotechnical Engineering, 9(2), 340-349. Predicting the impact of rolling dynamic compaction using dynamic cone penetrometer test results
  • 81. Applications in engineering disciplines Control engineering 81 KACHULKOVA, S. Applications of Artificial Neural Networks in Control Systems. transfer, 1(1), 5.
  • 83. Tips to start utilizing machine learning and ANN techniques for your previous, current and future studies in your field… PAST – PRESENT - FUTURE 83 ANN FOR RESEARCH
  • 84. 84 TIP #1 CLASSIFY THE VARIABLES DEPENDENT VARIABLES (DV) INDEPENDENT VARIABLES (IV) CONTROL VARIABLES (CV)
  • 85. 85 TIP #2 ENSURE HIGH-QUALITY OF DATA GIGO GARBAGE IN = GARBAGE OUT
  • 87. 87 TIP #2 ENSURE HIGH-QUALITY OF DATA WHAT IS YOUR UNCERTAINTY THRESHOLD? WHAT MODEL ARE YOU USING FOR UNCERTAINTY ANALYSIS? WHAT IS YOUR UNCERTAINTY THRESHOLD?
  • 88. 88 Example of uncertainty model and analysis The uncertainty associated with the result is WR and those associated with the independent variables are W1, W2, …, Wn. Source: Kline, S. J. (1953). Describing uncertainty in single sample experiments. Mech. Engineering, 75, 3-8.
  • 89. 89 Example of uncertainty model and analysis Source: Sarafraz, M. M., Safaei, M. R., Leon, A. S., Tlili, I., Alkanhal, T. A., Tian, Z., ... & Arjomandi, M. (2019). Experimental investigation on thermal performance of a PV/T-PCM (photovoltaic/thermal) system cooling with a PCM and nanofluid. Energies, 12(13), 2572.
  • 90. 90 Data filtration Height of an adult male by fixing the typo or possibly remeasuring the item or person. If that’s not possible, you must delete the data point because you know it’s an incorrect value. Source: Statistics by Jim. Link: https://statisticsbyjim.com/basics/remove-outliers/ (access date: 25th of July 2022)
  • 91. 91 COLLECT AS MUCH DATA AS POSSIBLE. TIP #3 COLLECT BULK DATA Once testing station, or setup, is established. Record as much data as possible. With a variety of design configurations.
  • 92. 92 Number of input, outputs and hidden nodes. TIP #4 Select the components Utilize the identified variables (IV, DV, CV)
  • 93. 93 How much (%) of the data are you allocating for training? Validating? Testing? TIP #5 Choose the data splitting percentage
  • 94. 94 Use MATLAB/Python to create a simple MLP Neural Network to process the data TIP #6 Start with a simple MLP ANN
  • 97. Food for Thought • What is the difference between Artificial Intelligence and computational intelligence? • How to interpret skewness and kurtosis? • What is the significance of the activation function? • What is the significance of the bias? 97
  • 98.  Identify the dependent and independent variables in your work.  Assign the input and output nodes to your network.  Conduct data pre-processing.  Decide the type of problem you have.  Choose an activation function.  Choose the number of hidden nodes.  Assign weights and biases.  Download MATLAB and implement a simple MLP code.  Carryout the training, validation and testing. Homework 98
  • 100. Review article in PV/T 100
  • 101. Original article on ANN use for PV/T Performance Prediction 101
  • 102. It’s play time… • https://playground.tensorflow.org/ 102

Editor's Notes

  1. Self driving car , GPS navigation ,
  2. Symbolists firmly believed in developing an intelligent system based on rules and knowledge and whose actions were interpretable
  3. https://www.youtube.com/watch?v=2ePf9rue1Ao Symbolic learning uses symbols to represent certain objects and concepts, and allows developers to define relationships between them explicitly.
  4. Non-AI programs simply carry out a defined sequence of instructions.
  5. Neuron (unit / node / cell)
  6. Activation function A sigmoid function is a bounded, differentiable, real function that is defined for all real input values and has a non-negative derivative at each point and exactly one inflection point. A sigmoid "function" and a sigmoid "curve" refer to the same object. hyperbolic tangent function The rectified linear activation function
  7. All classification tasks depend upon labeled datasets; that is, humans must transfer their knowledge to the dataset in order for a neural network to learn the correlation between labels and data. This is known as supervised learning.
  8. Hardware breakdowns (data centers, manufacturing, transport) Health breakdowns (strokes, heart attacks based on vital stats and data from wearables) Customer churn (predicting the likelihood that a customer will leave, based on web activity and metadata) Employee turnover (ditto, but for employees)
  9. Hardware breakdowns (data centers, manufacturing, transport) Health breakdowns (strokes, heart attacks based on vital stats and data from wearables) Customer churn (predicting the likelihood that a customer will leave, based on web activity and metadata) Employee turnover (ditto, but for employees)
  10. The agent learns to behave i36 environment depending on these rewards. To understand the limitations and merits of an algorithm and to develop efficient learning algorithms is the goal in reinforcement learning. Sparse: random labelling of individual cells
  11. One or more hidden layers
  12. This is regression
  13. Influence of the number of layers on the pattern recognition ability of MLP
  14. Stage (1) and (2) same input / (3) different input
  15. Y = m (slope) x + B (intersection)
  16. interpretation is clear
  17. of an estimator (of a procedure for estimating an unobserved quantity)