ELP 3205

POWER SYSTEMS I
LOAD FORECAST

INTRODUCTION

Electricity load forecasts are important to:

Electricity utilities who need to make important
decisions on purchasing and generating electric
power, load switching and infrastructure
development
Energy purchasers, Independent Service Providers
(ISPs) or Independent Power Producers (IPPs),
financial institutions and other participants in
electric energy generation, transmission and
distribution, and markets

THE DEREGULATED POWER

INDUSTRY

Traditional monopolies have been replaced

by several generation and distribution
utilities that bid to sell or buy electricity
through a wholesale market.

The expected end result is a decline in the

price of electricity to end users and a price
that better reflects the actual costs
involved.

IMPORTANCE OF LOAD FORECASTS

Electricity utilities face two complex

statistical problems: how to forecast both
electricity load and the wholesale spot price
of electricity.

Failure to implement efficient solutions to

these two forecasting problems can directly
result in multimillion dollar losses through
uninformed trades in the wholesale market.

PROBLEMS OF ACCURATE LOAD

FORECASTING

Load forecasting is however a difficult task

because the load series is complex and
exhibits several levels of seasonality (not so
much for Guyana)

There are many important exogenous

variables that must be considered,
especially weather-related variables.

LOAD FORECASTING PERIODS

SHORT TERM FORECASATS

Short term electric load forecast spans the

period from one hour up to one week and is
mainly utilized for

power system operation studies,

loss reduction,
voltage regulation,
unit commitment (how to schedule operation of
units)
maximizing the utility revenues in the
deregulated environment.

MEDIUM TERM LOAD FORECASTS

Medium term electric load forecast spans

the period from one week to several weeks
and it is mainly utilized for

predicting the necessary power to purchase or sell

from other neighboring networks (inter-tie
exchanged power)
calculation of the fuel required by the utility in the
near future.

MEDIUM TERM LOAD FORECASTS

Medium term electric load forecasts can

also mean up to five years where utilities
make investment in new generators and
build new or upgrade transmission lines

LONG TERM FORECASTS

Long Term Forecasts can be up to 20 years

and are used for policy determination e.g
type of fuel renewable or non renewable,
etc. These forecasts are continually
adjusted for the medium term

ELECTRICAL LOAD FORECASTING

MODELS

A model can be defined as a mathematical

description of how the complex elements of
a real-life situation or problem might
interplay at some future date.

In projecting electricity demand, the model

uses data on electricity prices, income,
population, the economy, and the growth
rates for each and then varies the mix
according to varying sets of assumptions

ELECTRICAL LOAD FORECASTING

THEORY

The relationships between electricity

demand and the multitude of factors that
influence or affect electricity demand are
expressed in mathematical equations called
functions.
A model is a collection of functions.
A function, in turn, is made up of variables
of those factors which change or can be
changed.

ELECTRICAL LOAD FORECASTING

THEORY
Independent variables are those factors
which influence the demand for electricity.
The dependent variable is electricity
demand itself
Elasticities describe how much the
dependent variable (electricity demand)
changes in response to small changes in the
independent variables.
Elasticities are what the modeler uses to
measure consumer behavior.

ELECTRICAL LOAD FORECASTING

THEORY

Energy planners often speak of scenarios

which are hypothetical pictures of the future
based on different assumptions about
economic or political events.

SHORT TERM LOAD FORECASTING

METHODS

Short-Term Load Forecasting is basically a

load predicting system with a leading time of
one hour to seven days, which is necessary
for adequate scheduling and operation of
power systems.

It has been an essential component of Energy

Management Systems (EMS). For proper and
profitable management in electrical utilities,
short-term load forecasting is very important.

SHORT TERM LOAD FORECASTING

METHODS

High forecasting accuracy and speed are the

two most important requirements of shortterm load forecasting and it is important to
analyze the load characteristics and identify
the main factors affecting the load.

In electricity markets, the traditional load

affecting factors such as season, day,
weather and electricity price may have a
complicated relationship with system load.

SHORT TERM LOAD FORECASTING

METHODS

Traditional statistical load forecasting

techniques, such as regression, time series,
pattern recognition, Kalman filters, etc., have
been used in practice for a long time, showing
the forecasting accuracy that is system
dependent.

These traditional methods can be combined

using weighted multi-model forecasting
techniques, showing adequate results in
practical systems.

SHORT TERM LOAD FORECASTING

METHODS

The short term load forecasting methods

are
Similar Day Lookup Approach
Regression Based Approach
Time Series Analysis
Artificial Neural Network
Expert System
Fuzzy logic
Support Vector Machines

Similar Day Look Up Approach

Similar day approach is based on searching

historical data of days of one, two or three
years having the similar characteristics to the
day of forecast. The characteristics include
similar weather conditions, similar day of the
week or date.
The load of the similar day is considered as
the forecast. Now, instead of taking a single
similar day, forecasting is done through linear
combinations or regression procedures by
taking several similar days.

Similar Day Look Up

Approach

The trend coefficients of the previous years

are extracted from the similar days and
forecast of the concerned day is done on
their basis.

Regression Based Approach

Linear regression is a technique which

examines the dependent variable to
specified independent variable. The
independent variables are firstly considered
because changes occur in them
unfortunately. In energy forecasting, the
dependent variable is usually demand.
Independent variables are usually weather
related, such as temperature, humidity or
wind speed.

Regression Based
Approach

Slope coefficients measure the sensitivity of the

dependent variable and how they change with the
independent variable. Also, by measuring how
significant each independent variable has
historically been in its relation to the dependent
variable the future value of the dependent variable
can be estimated.

Essentially, regression analysis attempts to

measure the degree of correlation between the
dependent and independent variables, thereby
establishing the latters predicted values.

Regression Based
Approach

Regression is the one of most widely used statistical

techniques. For electric load forecasting, regression
methods are usually used to model the relationship
of load consumption and other factors such as
weather, day type, and customer class.

There are several regression models for the next

day peak forecasting. Their models contain
deterministic influences such as holidays, random
variables influences such as average loads, and
exogenous influences such as weather.

Time Series Analysis

Time series forecasting is based on the idea

that reliable predictions can be achieved by
modeling patterns in a time series plot, and
then extrapolating those patterns to the
future. Using historical data as input, time
series analysis fits a model according to
seasonality and trend.

Time Series Analysis

Time series models can be accurate in some

situations, but are especially complex and
require large amounts of historical data.

Time series approaches are not widely used

for energy industry forecasting. Because
they typically do not take into account other
key factor, such as weather forecasts

ARTIFICIAL NEURAL NETWORKS

Artificial Neural Networks are still at a very

early stage and are electronic models based
on the neural structure of the brain. We
know that the brain basically learns from
the experience. These biological inspired
methods are thought to be the major
advancement in the computational industry.

ARTIFICIAL NEURAL NETWORKS

In a neural network, the basic processing

element is the neuron. These neurons get
input from some source, combine them,
perform all necessary operations and put the
final results on the output.

Artificial neural networks have been

developed since mid-1980s and have been
extensively applied. They have very
successful applications in pattern recognition
and many other problems.

ARTIFICIAL NEURAL NETWORKS

Forecasting is based on the pattern

observed from the past event and estimates
the values for the future. ANN is well suited
to forecasting for two reasons. First, it has
been demonstrated that ANNs are able to
approximate numerically any continuous
function to be desired accuracy. In this case
the ANN is seen as multivariate, nonlinear
and nonparametric methods.

ARTIFICIAL NEURAL NETWORKS

Secondly, ANNs are date-driven methods, in

the sense that it is not necessary for the
researcher to use tentative models and then
estimate their parameters. ANNs are able to
automatically map the relationship between
input and output, they learn this
relationship and store this learning into their
parameters.

ARTIFICIAL NEURAL NETWORKS

The first way is by repeatedly forecasting

one hourly load at a time.

The second way is by using a system with

24 NNs in parallel, one for each hour of the
day.

ARTIFICIAL NEURAL NETWORKS

In Multi Layer Perceptron(MLP) structure of

neural network, the most commonly
training algorithm use is the back
propagation algorithm. These algorithms
are iterative; some criteria must be defined
to stop the iterations. For this either training
is stopped after a fixed number of iterations
or after the error decreased below some
specified tolerance.

ARTIFICIAL NEURAL NETWORKS

To use the ANN in electric load forecast

problems, distribution engineers should
decide upon a number of basic variables,
these variables include:
Input variable to the ANN (load, temperatureetc)
Number of classes (weekday, weekend, seasonetc)
What to forecast: hourly loads, next day peak load,
next day total load etc

ARTIFICIAL NEURAL NETWORKS

Neural network structure (Feed-forward, number

of hidden layer, number of neuron in the hidden
layeretc)

Training method and stopping criterion

Activation functions

Size of the training data

Size of the test data

EXPERT

SYSTEMS

Expert systems are new techniques that have

come out as a result of advances in the field of
artificial intelligence (AI) in the last two
decades.

An expert system is a computer program, which

has the ability to act as an expert.

This means this computer program can reason,

explain, and have its knowledge base expanded
as new information becomes available to it.

EXPERT SYSTEMS

The load forecast model is built using the

knowledge about the load forecast domain from
an expert in the field. The "Knowledge Engineer"
extracts this knowledge from load forecast
(domain) expert which is called the acquisition
module component of the expert system.

This knowledge is represented as facts and

rules by using the first predicate logic to
represent the facts and IF-THEN production
rules.

EXPERT SYSTEMS

This representation is built in what is called

the knowledge base component of the
expert system. The search for solution or
reasoning about the conclusion drawn by
the expert system is performed by the
"Inference Engine" component of the expert
system. Any expert system must have the
capability to trace its reasoning if asked by
the user.

EXPERT SYSTEMS

An example demonstrating this approach is

the rule-based algorithm which is based on
the work of two scientists Rahman and
Baba. This algorithm consists of functions
that have been developed for the load
forecast model based on the logical and
syntactical (structural) relationship between
the weather and prevailing daily load
shapes in the form of rules in a rule-base.

EXPERT SYSTEMS

The rule-base developed consists of the set

of relationships between the changes in the
system load and changes in natural and
forced condition factors that affect the use of
electricity. The extraction of these rules was
done off-line, and was dependent on the
operator experience and observations by the
authors in most cases. Statistical packages
were used to support or reject some of the
possible relationships that have been
observed

EXPERT SYSTEMS

The rule-base consists of all rules taking the

IF-THEN form and mathematical
expressions. This rule-base is used daily to
generate the forecasts. Some of the rules
do not change over time, some change very
slowly while others change continuously
and hence are to be updated from time to
time

FUZZY LOGIC

Fuzzy logic is based on the usual Boolean

logic which is used for digital circuit design.
In Boolean logic, the input may be the truth
value in the form of 0 and 1. In case of
fuzzy logic, the input is related to the
comparison based on qualities.

FUZZY LOGIC

For example, we can say that a transformer

load may be low and high. Fuzzy logic
allows us to deduce outputs from inputs
logically. In this sense, the fuzzy facilitate
for mapping between inputs and outputs
like curve fitting

FUZZY LOGIC

The advantage of fuzzy logic is that there is

no need of mathematical models for
mapping between inputs and outputs and
also there is no need of precise or even
noise free inputs. Based on the general
rules, properly designed fuzzy logic systems
are very strong for the electrical load
forecasting.

FUZZY LOGIC

There are many situations where we require

the precise outputs. After the whole
processing is done using the fuzzy logic, the
defuzzification is done to get the precise
outputs.

FUZZY LOGIC

We know that power system load is

influenced by many load factors such as
weather, economic and social activities and
different load components. By the analysis of
historical load data it is not easy to make the
accurate forecast. The use of these
intelligent methods like fuzzy logic and
expert systems provide advantage on other
conventional methods. The numerical
aspects and uncertainties are suitable for the
fuzzy methodologies.

SUPPORT VECTOR
MACHINES

Support Vector Machines (SVM) are the most

powerful and very recent techniques for the
solution of classification and regression
problems.

Vector machines use simple linear functions to

create linear decision boundaries in the new
space. In the case of neural network, the
problem is in the choosing of architecture and
in the case of support vector machine,
problems occurs in choosing a suitable kernel.

SUPPORT VECTOR
MACHINES

Mohandes applied a method of support vector

machines for short-term electrical load
forecasting. He compared this methods
performance with the autoregressive method.

The results indicate that SVMs compare

favorably against the autoregressive method.
Chen also proposed a SVM model to predict
daily load demand of a month. Lots of methods
are used in support vector machines.

SHORT AND MEDIUM TERM

FORECASTS

1. Trend Analysis

2. End Use Analysis

3. Econometrics

SHORT AND MEDIUM TERM

FORECASTS

Each of the three forecasting methods uses

a different approach to determine electricity
demand during a specific year in a
particular place. Each forecasting method is
distinctive in its handling of the four basic
forecast ingredients:

SHORT AND MEDIUM TERM

FORECASTS

1. The mathematical expressions of the

relationship between electricity demand
and the factors which influence or affect it the function

2. The factors which actually influence

electricity demand (population, income,
prices, etc.) - the independent variables

SHORT AND MEDIUM TERM

FORECASTS

3. Electricity demand itself - the dependent

variable

4. How much electricity demand changes in

response to population, income, price, etc.,
changes- the elasticities?

SHORT AND MEDIUM TERM

FORECASTS

The only way to determine the accuracy of

any load forecast is to wait until the forecast
year has ended and then compare the
actual load to the forecast load.

Even though the whole idea of forecasts is

accuracy, nothing is said in the comparison
of the three forecasting methods about
which method produces the most accurate
forecasts.

SHORT AND MEDIUM TERM

FORECASTS

The only thing certain about any long-range

forecast is that it can never be absolutely
precise. Forecasting accuracy depends on
the quality and quantity of the historical
data used, the validity of the forecasters
basic assumptions, and the accuracy of the
forecasts of the demand-influencing factors
(population, income, price, etc.).

SHORT AND MEDIUM TERM

FORECASTS

None of these is ever perfect. Consequently,

regional load forecasts are reviewed and
revised yearly. Even so, there is no case in
which the electricity demand will be exactly
as forecast, no matter what method is used
or who makes the forecast.

TREND ANALYSIS

Trend analysis (trending) extends past

growth rates of electricity demand into the
future, using techniques that range from
hand-drawn straight lines to complex
computer-produced curves. These
extensions constitute the forecast.

TREND ANALYSIS

Trend analysis focuses on past changes or

movements in electricity demand and uses
them to predict future changes in electricity
demand. Usually, there is not much
explanation of why demand acts as it does,
in the past or in the future.

TREND ANALYSIS

The advantage of trend analysis is that it is

simple, quick and inexpensive to perform. It
is useful when there is not enough data to
use more sophisticated methods or when
time and funding do not allow for a more
elaborate approach.

TREND ANALYSIS

The disadvantage of a trend forecast is that

it produces only one result - future
electricity demand. It does not help analyze
why electricity demand behaves the way it
does, and it provides no means to
accurately measure how changes in energy
prices or government policies (for instance)
influence electricity demand.

TREND ANALYSIS

Another shortcoming of trend analysis is

that it relies on past patterns of electricity
demand to project future patterns of
electricity demand. This simplified view of
electrical energy could lead to inaccurate
forecasts in times of change, especially
when new concepts such as conservation
and load management must be included in
the analysis

END-USE ANALYSIS

The basic idea of end-use analysis is that

the demand for electricity depends on what
it is used for (the end-use). For instance, by
studying historical data to find out how
much electricity is used for individual
electrical appliances in homes, then
multiplying that number by the projected
number of appliances in each home

END-USE ANALYSIS

and multiplying again by the projected

number of homes, an estimate of how much
electricity will be needed to run all
household appliances in a geographical
area during any particular year in the future
can be determined.

END-USE ANALYSIS

Using similar techniques for electricity used

in business and industry, and then adding
up the totals for residential, commercial,
and industrial sectors, a total forecast of
electricity demand can be derived.

END-USE ANALYSIS

The advantages of end-use analysis is that

it identifies exactly where electricity goes,
how much is used for each purpose, and the
potential for additional conservation for
each end-use.

END-USE ANALYSIS

End-use analysis provides specific

information on how energy requirements
can be reduced over time from conservation
measures such as improved insulation
levels, increased use of storm windows,
building code changes, or improved
appliance efficiencies.

END-USE ANALYSIS

An end-use model also breaks down

electricity into residential, commercial and
industrial demands. Such a model can be
used to forecast load changes caused by
changes within one sector (residential, for
example) and load changes resulting
indirectly from changes in the other two
sectors.

END-USE ANALYSIS

Commercial sector end-use models

currently being developed have the
capability of making energy demand
forecasts by end-uses as specific as type of
business and type of building. This is a
major improvement over projecting only
sector-wide energy consumption and using
economic and demographic data for large
geographical areas [

END-USE ANALYSIS

The disadvantage of end-use analysis is

that most end-use models assume a
constant relationship between electricity
and end-use (electricity per appliance, or
electricity used per dollar of industrial
output). This might hold true over a few
years, but over a 10-or 20-year period,
energy savings technology or energy prices
will undoubtedly change, and the
relationships will not remain constant.

END-USE ANALYSIS

Another disadvantage of end-use analysis is

that it also requires extensive data, since all
relationships between electric load and all
the many end-uses must be calculated as
precisely as possible. Data on the existing
stock of energy-consuming capital
(buildings, machinery, etc.) in many cases
is very limited.

END-USE ANALYSIS

Also, if the data needed for end-use analysis

is not current, it may not accurately reflect
either present or future conditions, and this
can affect the accuracy of the forecast.

Finally, end-use analysis, without an

econometric component that is explained
above, does not take price changes
(elasticity of demand) in electricity or other
competing fuels into consideration.

END-USE ANALYSIS

Ideally this approach is very accurate.

However, it is sensitive to the amount and
quality of end-use data. For example, in this
method the distribution of equipment age is
important for particular types of appliances.

End-use forecast requires less historical

data but more information about customers
and their equipment

ECONOMETRICS

Econometrics uses economics, mathematics, and

statistics to forecast electricity demand.

Econometrics is a combination of trend analysis

and end-use analysis, but it does not make the
trend-analysts assumption that future electricity
demand can be projected based on past demand.

Econometrics can allow for variations in the

relationship between electricity input and enduse.

ECONOMETRICS

Econometrics uses complex mathematical

equations to show past relationships between
electricity demand and the factors which influence
that demand.

An equation can show how electricity demand in

the past reacted to population growth, price
changes, etc. For each influencing factor, the
equation can show whether the factor caused an
increase or decrease in electricity demand, as well
as the size (in percent) of the increase or
decrease.

ECONOMETRICS

For price changes, the equation can also

show how long it took consumers to
respond to the changes. The equation is
then tested and fine tuned to make sure
that it is as reliable a representation as
possible of the past relationships.

ECONOMETRICS

Once this is done, projected values of

demand-influencing factors (population,
income, prices) are put into the equation to
make the forecast. A similar procedure is
followed for all of the equations in the
model.

ECONOMETRICS
The advantages of econometrics are that it
provides detailed information on:
future levels of electricity demand,
why future electricity demand increases or
decreases,
how electricity demand is affected by
various factors.
In addition, it provides separate load
forecasts for residential, commercial, and
industrial sectors.

ECONOMETRICS

A disadvantage of econometric forecasting is that

in order for an econometric forecast to be
accurate, the changes in electricity demand
caused by changes in the factors influencing that
demand must remain the same in the forecast
period as in the past.

This assumption (which is called constant

elasticities) may be hard to justify, especially
where very large electricity price changes (as
opposed to small, gradual changes) make
consumers more sensitive to electricity prices

ECONOMETRICS

Also, the econometric load forecast can only

be as accurate as the forecasts of factors
which influence demand.

Many of the demand-influencing factors

which may be treated and projected
individually in the mathematical equations
could actually depend on each other, and it
is difficult to determine the nature of these
interrelationships.

ECONOMETRICS

Econometric models work best when

forecasting at national, regional, or state
levels. For smaller geographical areas,
meeting the model can be a problem. This
is oddly shaped service areas for which
there demographic data.

THE GUYANA FORECAST

The following methodology was used to determine the GPL

load forecast.

Annual and monthly data on generation, sales, losses and

number of customers were collected as were available.

Annual historical and projected data were collected on GDP,

population, and exchange rates. Quarterly GDP series data
was not available.

An econometrics load forecast was carried out using annual

data with generation/sales as the independent variable and
GDP and population as the dependent variables. The
correlation was weak and the load forecast seemed very high.

THE GUYANA FORECAST

The power system in Guyana did not develop

under one national grid. For different reasons, the
major industries, bauxite and sugar, did not seek
to purchase electricity from the local utility but
developed their own generation and distribution
facilities.

Beginning in the early 1980s the local power utility

experienced severe generation shortfalls and
many businesses purchased emergency standby
diesel generators to augment the utilitys supply.

THE GUYANA FORECAST

More recently, as the power company was

faced with increasing commercial losses
and rising oil prices, many companies used
the standby generators as their main
electricity supplier and improved their
generating capacity to ensure higher
reliability

THE GUYANA FORECAST

A trend load forecast was carried out using

monthly data of sales by consumer category
and generation. The forecast seemed more
acceptable than the econometrics forecast for
the medium term but did not take into account
changes that may occur in the long term.

For the long term an econometrics forecast

was done using additional data provided by the
trend forecast and considering GDP growth
rates of 3.5%, 5% and 7%.

THE GUYANA FORECAST

THE GUYANA FORECAST

MONTHLY SALES AND LOSSES 2000-2008
50

40

Residential
Commercial
Small Industrial
LargeIndustrial
Street Lighting
Losses

MWh

30

20

10

0
00

01

02

03

04
Figure 2.3

05

06

07

08

THE GUYANA FORECAST

THE GUYANA FORECAST

THE GUYANA FORECAST

HISTORICAL & PROJECTED POPULATION 1980-2025
840
820
800

000s

780
760
740
720
700
85

90

95

00

05

Figure 2.7

10

15

20

25

THE GUYANA FORECAST

It was then decided to use a trend analysis
because nine years of monthly sales data
(108 points) per tariff category was
available but there was no corresponding
monthly or quarterly economic data.
A dynamic forecast method was utilized
using forecast values of lagged dependent
variables rather than the actual values of
the lagged dependent variables. The model
utilized is the simple regression equation

THE GUYANA FORECAST

The model utilized is the simple regression equation

yt = a + bxt + t

and for the lagged forecast model

yt = bxt (for t=1), and,

yt (t) = bxt - bxt-1 + yt-1 (for t>1)

where y = sales
x = time intervals
and a, b are the constants of the regression equation and is
the error calculated by the software.

THE GUYANA FORECAST

This model was used to forecast on a

monthly basis each tariff category
individually and also total sales and
generation. The monthly forecasts were
then added to give the annual load forecast

THE GUYANA FORECAST

SALES FORECASTS 2014 - 2025
900

800

GWh

700

600

500

400
14

15

16

17

18

19

20

21

22

23

24

Figure 2.8
Trend
Econometrics-5%

Econometrics-3.5%
Econometrics-7%

25

ANY

QUESTIONS?

 THANK

YOU