Load Forecasting-Basic Concepts,Methods and Fundamentals of Harmonics

1. Chinmaya Das
Priyank Jain
Rahul Sharma

2. TOPICS TO BE DISCUSSED
• Major objectives of Load Forecasting.
• Parameters influencing Load Forecasting .
• Load Factor & Diversity factor.
• Types of Load Forecasting based on time-frame.
• Different factors involved in Load Forecasting.
• System Peak forecasting.
• Methods used for Load Forecasting .
• G-S method of Load Flow study.
• Load Duration Curve & its significance.
• Harmonics are in a Power System & its effects on the Network .

3. BASIC DEFINITIONS [1]
Load
The power consumed by a Electrical Circuit.
Forecasting
The process of making statements about events
whose actual outcomes have not yet been observed.
Load forecasting
An estimate of power demand at some future period.

4. LOAD FORECASTING
• Load forecasting is a central and integral process in the
planning and operation of electric utilities.
• It involves the accurate prediction of both the magnitudes and
geographical locations of electric load over the different
periods (usually hours) of the planning horizon.
• Accurate load forecasting holds a great saving potential for
electric utility corporations.

5. PROGRESSIVE PATH
• The basic quantity of interest in load forecasting is typically the
hourly total system load. However, according to Gross and
Galiana (1987), load forecasting is also concerned with the
prediction of hourly, daily, weekly and monthly values of the
system load, peak system load and the system energy.
• Srinivasan and Lee (1995) classified load forecasting in terms of
the planning horizon’s duration: up to 1 day for short-term load
forecasting (STLF), 1 day to 1 year for medium-term load
forecasting (MTLF), and 1±10 years for long-term load
forecasting (LTLF).

6. FACTORS INFLUENCING LOAD FORECASTING
Population
Living Geographical
Standard Location
Cost of
Future Plan
Power

7. IMPORTANCE OF LOAD FORECASTING
Forecasting gives magnitude and location of loads.
Accurate model helps in
1)Economic size of plant and apparatus
at correct time and place.
2)Generation authorities plan their water and fuel requirements and
the generator allocation schedules.

8. IMPORTANCE OF LOAD FORECASTING
3)Load forecasting helps an
electric utility to make important
decisions including decisions on
purchasing and generating electric
power, load switching, and infrastructure
development.
4)Load forecasts are extremely
important for energy suppliers,
ISOs, financial institutions, and
other participants in electric energy
energy generation, transmission,
distribution, and markets.

9. PRESENT AIM…
“The aim of the load forecasting is to
make best use of electrical energy
and reveled the conflict between
supply and demand.”
- International Journal of Systems Science,2009

10. OBJECTIVES OF LOAD FORECASTING
• To know the peak load of system
• Energy requirement in day, month and year
• To know the load duration curve
• To estimate the proper investment requirement
• Supply side management
• Demand side management

11. FACTORS FOR ACCURATE
FORECASTS
 Weather influence
 Time factors
 Customer classes

12. WEATHER INFLUENCE
Electric load has an obvious correlation to
weather. The most important variables
responsible in load changes are:
• Dry and wet bulb temperature
• Dew point
• Humidity
• Wind Speed / Wind Direction
• Sky Cover
• Sunshine

13. [7]

14. TIME FACTORS
In the forecasting model, we should also
consider time factors such as:
• The day of the week
• The hour of the day
• Holidays

15. CUSTOMER CLASS
• Electric utilities usually serve different
types of customers such as residential,
commercial, and industrial.
• The graphs show the load behavior in the
above classes by showing the amount of
peak load per customer, and the total
energy.

16. LOAD CURVES

17. DEMAND FACTOR
• The ratio of the maximum coincident demand of a system, or
part of a system, to the total connected load of the system, or
part of the system, under consideration.
Demand Factor = maximum demand
total connected load (of consumer)

18. LOAD FACTOR
The total amount of energy the plant produced during a period of
time and divide by the amount of energy the plant would have
produced at full capacity.
Load Factor = Total amount of energy the plant produced
Plant’s Installed capacity

19. DIVERSITY FACTOR
• The ratio of the sum of the individual maximum demands of the various
subdivisions of a system to the maximum demand of the whole system.
Diversity Factor = Σ Di ( i=1 to n)
DG
Where,
• Di = maximum demand of load i, regardless of time of occurrence.
• DG = coincident maximum demand of the group of n loads

20. SYSTEM POWER FACTOR
The power factor of an AC electric power system is defined as
the ratio of the real power flowing to the apparent power in the
circuit.
OR
Measurement of cosine of angular difference between voltage
and current.
P = V* I cosφ
Pf varies from 0 to 1.
Pf = 0, when phase angle is 90.
Pf =1, when phase angle is 0.

21. SYSTEM POWER FACTOR
When power factor (φ=0, cosφ=1). When power factor (φ=90, cosφ=0)

22. LOAD FORECAST

23. LOAD CHARACTERISTICS
• Diversity Factor 1.2-1.3
Domestic
• Load Factor 10 -15%
• Diversity Factor 1.1-1.2
Commercial
• Load Factor 25-30%

24. TYPES OF LOAD
• Diversity Factor ~1.1
Industrial • Load Factor 65-70%
• Diversity Factor 1.0
Municipal • Load Factor 25-30%

25. Type of Demand Load Utilization
Industry Factor Factor Factor
Induction furnace 0.99 0.80 0.72
Steel Rolling Mills 0.80 0.25 0.72
Textile Industry 0.50 0.80 0.40
Gas Plant Industry 0.70 0.50 0.35
College & Schools 0.50 0.20 0.10
Paper Industry 0.50 0.80 0.40
Source[Electrical engineering portal]

26. Schematic of STLF
Model
Load
Load at
Temp Current Previous Previous Demand coming
hour
hour Hour 2 Hour
Forecast
Wind
Cloud

27. DEMAND SUPPLY FORECAST : INDIA
[5]

28. TYPES OF LOAD FORECASTING
Short Term(1
hour- 1 week)
Load
Forecasting
Long
Middle Term(1
Term(Longer
week -1 year)
than Year)

29. DE-REGULATION AND FORECASTING
• Load forecasting has always been important for planning
and operational decision conducted by utility companies.
• However, with the deregulation of the energy industries,
load forecasting is even more important.
• With supply and demand fluctuating and the changes of
weather conditions and energy prices increasing by a factor of
ten or more during peak situations, load forecasting is vitally
important for utilities.

30. CONCEPTS
 Time series (based on historical datas)
 Trend analysis
 Correlation Theory
 Aritficial Neural Networks

31. TIME SERIES

32. TIME SERIES
Time Series
Model
Additive Multiplicative
(Y=T+C+S+I) (Y=T C S I)
WHERE
T = LONG TERM TREND
C= CYCLICAL TREND (MAINLY OVER MANY YEARS)
S = SEASONAL TREND (1 YEAR CYCLE)
I = IRREGULARR MOVEMENTS(NOISE)*
*IN PART DUE TO TEMPERATURE EFFECTS

33. TIME SERIES-COMPONENTS

34. REGRESSION / TREND ANALYSIS
 Trend Analysis
Study of behaviour of a time series or a process in the past and
its mathematical modelling so that future behaviour can be
extrapolated from it.
Gives nature of relationships between the variables, where as correlation
analysis measures the degree of relationship between the variables.

35. REGRESSION / TREND ANALYSIS
Approaches to Trend Analysis
1.Fitting continuous mathematical functions through actual data
2.Fitting of a sequence on discontinuous lines or curves to data
(short term forecasting)

36. REGRESSION/TREND ANALYSIS

37. TREND ANALYSIS

38. TREND ANALYSIS
 Typical regression curves used in power system forecasting
Linear y=A+Bx
Exponential y=A(1+B)x
Power y=AxB
Polynomial y=A+Bx+Cx2
Method of Least squares
 Can be used either to fit a straight line trend or a parabolic Trend

39. CORRELATION THEORY
Scatter Diagram
Karl Pearson’s
coefficient of
Correlation Correlation
Methods Spearman’s
Rank correlation
coefficient
Method of least
squares
1) Coefficient of correlation is also called “goodness of fit”
2) Karl Pearson's coefficient of correlation
∑(X-Xav). (Y-Yav)
r= -------------------------
(∑(X-Xav)2. ∑(Y-Yav)2)1/2

40. CORRELATION THEORY

41. ARTIFICIAL NEURAL NETWORKS
• Artificial Neural Network(ANN) is based
on Artificial Intelligence.
• ANN, usually called neural network (NN),
is a mathematical model or computational
model that is inspired by the structure
and/or functional aspects of biological
neural networks
• Neural networks offer the potential to
overcome the reliance on a functional
form of a forecasting model.

42. SUMMARY OF FUNDAMENTAL STEPS
1)Collection of data (reliably)
2)Draw a graph
3)Construct a long term trend
4)Seasonal index if it exists and de-seasonalize the data
5)Adjust data for the trend

43. FACTORS IN LOAD FORECASTING

44. FACTORS IN POWER SYSTEM LOADING
Econometric
Single
Spatial Load
Factor
Forecasting
Modeling
Power
System
Loading
Forecast of
Strategic
System
Forecasting
Peak
Capacity
Forecasting

45. ECONOMETRIC
Growth in
population
(Long term
trend)
Growth of
GNP (Long
term
variation)
Business and
economic cycle
(cyclic Variation)
Most of these factors effect the long term trend and not effect normal
model based on past history.

46. GDP VS. ENERGY CONSUMPTION
• Relation of GDP to energy consumption is an important
indicator.
• The elasticity of consumption with respect to GDP for India
in 1980 – 1992 was 1.61.
This implies that increase in GDP of 1 % will increase
1.61 % of electricity consumption.

47. INDIA

48. SINGLE FACTOR MODELING [2]
Single factor modeling is based on a model that
assumes one dominant factor, determines the
model outcome.

49. DEFECTS IN SINGLE FACTOR MODELING
Too
General
DEFECTS
Not Biasing of
Comprehensive Forecast
as Rate of Values due to
Growth Differs Uneven
with Sectors Distribution

50. CAPACITY FORECAST MODEL
As the forecast for electrical energy is on
national level, in this the national projection is
converted to regional peaks.
From this the regional capacity requirements
are made, removing the current generation
and planned capacity addition there.
Finally addition of planned retirement /
decommissioning of units gives the net new
capacity to be added

51. STRATEGIC FORECASTING
Consumer’s Current Consumer’s Potential
Demand Demand
Strategic
Forecasting
Competitiveness in Availability of
Market Alternatives

52. STRATEGIC FORECASTING IN INDUSTRY
Strategic Strategic
Management Management
combines provides
Future
Econometric
Assessment
Technological Shaping of
Detail Future

53. SPATIAL FORECASTING
This method breaks down to geographically and
consumer oriented forecasting.
A land use map can be converted to electric load by
using kW per acre of load curves on land use class
basis.

54. Planning engineers are using GIS to visualize the
distribution system’s load and forecast “what they are likely
to see new addition to the system”.
Timeline of
Community
Development
Predictions
by Spatial
Location of
Forecasting Direction of
New infrastructure
substation investment

55. LIMITATIONS OF SPATIAL FORECASTING
 It cannot replace knowledge and experience of
area engineers.
 It cannot identify substation site to be
purchased.

56. FORECASTING FOR GROWTH IN REGIONS : -
 Urban areas – Increase in Specific Consumption
 Agriculture –
a. Projections of land irrigation
b. Prospective agricultural consumers
c. Availability of land water
 Industrial –
a. Diversification of business
b. New consumers
c. Change in production process

57. LOAD CURVE
 A Load Curve is a curve showing the variation of load on the
power station with respect to time .

58. TYPES OF LOAD CURVE
Daily
Load
Curve
Types of
Load
Curve
Yearly Weekly
Load Load
Curve Curve

59. SIGNIFICANCE OF LOAD CURVE
 Area under Load Curve = Units Generated
 Highest Point of Load Curve = Maximum Demand
 (Area Under Curve/Total No. of Hours)= Average Load
 Load Factor = Average Demand/ Maximum Demand

60. LOAD DURATION CURVE
• When the elements of a load curve are arranged in the order of
descending magnitudes.

61. SIGNIFICANCE OF LOAD DURATION CURVE
• The load duration curve gives the data in a more presentable
form.
• The area under the load duration curve is equal to that of the
corresponding load curve.
• The load duration curve can be extended to include any period
of time.

62. LOAD FLOW STUDY
• The power flow study (also known as load-flow study) is an
important tool involving numerical analysis applied to a power
system.
• A power flow study usually uses simplified notation such as
a one-line diagram and per-unit system, and focuses on various
forms of AC power (i.e. voltages, voltage angles, real power and
reactive power).

63. SIGNIFICANCE OF LOAD FLOW STUDY
• For planning future expansion of power systems as well as in
determining the best operation of existing systems.
• The principal information obtained from the power flow study is
the magnitude and phase angle of the voltage at each bus, and
the real and reactive power flowing in each line.

64. FORMULATION OF THE LOAD FLOW PROBLEM

65. FORMULATION OF THE LOAD FLOW PROBLEM
where [Y] is the nodal admittance matrix

66. GAUSS-SEIDEL METHOD
• The Gauss-Seidel Method is an iterative technique for solving
the load flow problem, by successive estimation of the node
voltages.
• It is usually done with the help of MATLAB.
• Can be used for quite complex equations.

67. GS METHOD FROM MATLAB(A SCREEN SHOT)

69. HARMONICS - INTRODUCTION
• A sinusoidal component of a periodic wave or quantity having a
frequency that is an integral multiple of the fundamental
frequency.
• Typical harmonics for a 50Hz system are,
Single phase – 3rd, 6th, etc.
Three phase – 5th, 7th, 11th, 13th, etc.
• Harmonics should not be confused with spikes, dips,
impulses, oscillations or other forms of transients.

70. WHY HARMONICS
• These current result due to the fact that the device either
has an impedance which varies during each half cycle of
applied emf or it generate a back emf of non sinusoidal
shape.
Result - Distortion of the
Wave shape .

71. The power company typically supplies a reasonably
smooth sinusoidal waveform:

72. NONLINEAR DEVICES WILL DRAW DISTORTED WAVEFORMS,
WHICH ARE COMPRISED OF HARMONICS OF THE SOURCE:

74. ORDER OF HARMONICS

75. PROPAGATION OF HARMONICS IN THE NETWORK
INFLUENCE OF PHASE ANGLE OF HARMONICS[6]

76. TOTAL HARMONICS DISTORTION
• The ratio of the sum of the power of all harmonic
components to the power of the fundamental
frequency
• Pn :- Sum of all power
• P1 :- Power of fundamental frequency

77. TOTAL HARMONICS DISTORTION
• THD can be used to describe voltage or current
distortion and is calculated as follows
• where
IDn is the magnitude of the nth harmonic as a percentage
of the fundamental (individual distortion).

78. HARMONICS ARE GENERATED BY :
Rectifiers
 Inverters
Induction furnaces
Arc furnaces
Fluorescent lamps
TVs
UPS & Computers etc.

79. ADVERSE EFFECTS OF HARMONICS
• Fluctuation of voltage
• Efficiency & capacity utilization of transformers, generators
• High skin effect loss
• High I2R loss
• High failure rate in Motors, sophisticated electronics equipments.

80. MAXIMUM LIMITS OF VOLTAGE HARMONIC
DISTORTION IN HT AND EHT SYSTEMS

81. HOW ARE HARMONICS MINIMIZED ?
• Use three-phase drives wherever possible.
• Use an additional inductance.
• Make use of a harmonic filter.

82. FILTERS
• A series- tuned harmonic filter consists of a capacitor bank with
a reactor (inductor) in series with it. The series combination
provides a low impedance path for a specific harmonic
component, there by minimizing harmonic voltage distortion
problems.
• The filter is tuned slightly below the harmonic frequency of
concern.

84. SYSTEM PEAK
It is given by the formula :-
Annual System Peak = Energy Requirement
8760 x Load Factor

85. QUESTION
• The estimated Energy requirement and Load Factor of a
particular Region for the year 2004 are 668132 GWh and
70% respectively. Calculate the annual peak demand.
• Given :
Energy Requirement - 668132 GWh
Load Factor - 70%

86. • Thus putting given values in the formula:
Ann. System Peak = Energy Requirement
8760 x Load Factor
i.e. ASP= 668132 GWh = 668132 x 1000MWh
8760 x 0.7 6132
= 108958 MWh
Thus,Annual Peak Demand = 108958MWh

87. DATE
CHANGE
DATE DATE DATE DATE
DATE DATE
CHANGE CHANGE CHANGE CHANGE
CHANGECHANGE
CHANGE VALUES
HERE!!!

88. ACTUAL LOAD CURVE FOR THE WEEK
05SEPT 2010 TO 11 SEPT 2010
1200
1000
800
monday
Tuesday
Wednesday
600 Thursday
FRIDAY
Saturday
Sunday
400
dte
200
0

89. REFERENCES
• [1].International Journal of Systems Science, volume
33, number 1.
• [2]. Electrical engineering portal
• [3].US electric static schneider
• [4].Department of Electrical and Electronics Engineering, S.V.U.
College of Engineering, Tirupati, A.P., India
• [5].India Energy Handbook.
• [6] Dept. of Electrical,Electronic and Control
Engineering, Ciudad Universitaria. Madrid. Spain.
• [7] NRLDC.nic.in