Load Forecasting Techniques

For Future demand

PRESENTED BY: SAMIULLAH JAN

 Main Topics

 Importance of Load Forecasting.

 Classification of Load.
 Types of Load Forecasting.
 Classification of Demand Forecasting.
 Factors Affecting Load Demand.
 Approach to Load Forecasting.
 Major Groups of Load Forecasting.
 Traditional Techniques.
 Modified Traditional Techniques
 Soft Computing Technique
 Importance of Load Forecasting

 Extremely important for Energy Suppliers.

 Planning of Generation , Transmission Network &

Distribution

 Accurate prediction of geographical locations of

electric load.
 Importance of Load Forecasting

 Electricity demand forecasting is considered as one

of the critical factors for economic operation of
power systems.

 Maximum savings when load forecasting is used to

control operations and decisions like economic
dispatch/ unit commitment and fuel allocation
 Classification of Load

 Industrial & Agricultural

 Domestic

 Commercial
 Types of Load Forecasting

 Short Term

 Medium Term

 Long Term
 Classification of Demand Forecasting

 Qualitative
Generally used by planners e.g Delphi method, curve
fitting.

 Quantitative
Regression analysis , exponential smoothing.
Load Demand
 Factors Affecting Load Demand

 Economic conditions of the people (per capita income)

 Population growth

 Cost of electricity

 Weather condition

 National policy, which involves the use of power and also grant available
for the same.

 Availability of national resources for the growth of power.

 Load, diversity and capacity factors of the power stations

 Approach to Load Forecast

 Peak Demand and Energy Forecast

 Total Component and Direct Approach

 Weather Dependence

 Average of All the Methods

 Three Major Groups of Load Forecasting

 Traditional
- Regression , multiple regression , exponential smoothing.

 Modified Traditional
- Adaptive demand Forecasting , Auto regressive AR
Model, Auto regressive moving average ARMA Model.

 Soft Computing
 Traditional Techniques

 Regression Model the relationship of load

consumption and other factors such as weather
conditions, day types and customer classes.

L(t)=Ln(t)+Σaixi(t)+e(t)
Ln= standard/normal load at time t
ai= slowly varying coefficients
Xi(t)= independent influencing factors e.g weather.
E(t) = error
n= no of observations
 Traditional Techniques

Multiple Regressions is the most popular method and

often used to forecast the load affected by a number of
factors ranging from meteorological effects, per capital
growth, electricity prices, economic growth etc.
Y(t)= v(t)a(t)+e(t)
Where , t is sampling time
Yt is total measured load system,
Vt is vector of adapted variables time, temperature,
(workday, weekend), etc., at is transposed vector of
regression coefficients
et is model error at time t.
 Traditional Techniques

 Exponential smoothing is one of the

approaches used for load forecasting. In this
method, first load is model based on previous data,
then to use this model to predict the future load.

 It is based on three smoothing constants for

stationary, trend and seasonality.
 Modified Traditional Techniques

 Adaptive demand Forecasting

Demand forecasting model parameters are
automatically corrected to keep track of the changing
load conditions (environmental). Hence Demand
forecasting is adaptive in nature and can also be used
as an on-line software package in the utilities control
system.
 Modified Traditional Techniques

 Auto-Regressive (AR) model can be used to

model the load profile, If the load is assumed to be
a linear combination of previous loads.

 Auto regressive moving average (ARMA)

model represents the current value of the time
series y(t) linearly in terms of its values at previous
periods [y(t-1), y(t-2),..] & in terms of previous
values of a white noise [a(t), a(t-1),].
 Soft Computing Techniques

 It is a fact that every system is pervasively imprecise,

uncertain and hard to be modeled precisely.

 A flexible approach called Soft Computing technique

has emerged to deal such models effectively and
most efficiently on research scenario.
 Soft Computing contd..

 The basic theme of soft computing is that precision

and certainty carry a cost and that intelligent
systems should exploit, wherever possible, the
tolerance for imprecision and uncertainty.

 Soft computing is collection of disciplines such as

Fuzzy logic, Neural Networks & Knowledge base
expert system
 Soft Computing Contd..

 In NTDC Medium term load forecast on annual basis

is carried out by Power Market Survey Software.

 In NTDC Long term load forecast on annual basis is

carried out by EVIEWS Software.

 Develops 10-years indicative Generation expansion

Plan annually by WASP Software
 Conclusion

 From the works reported so far, it can be inferred

that demand forecasting techniques based on soft
computing methods are gaining major advantages
for their effective use.
THANKS