LOAD FORECAST and methods of prediction

1. What is load?
 Load can be characterised electrically by: a)size of equipment(MW,KW)b) single
phase or three phase c) voltage level, frequency d) duty cycle –whether continous
or random
 Load devices may be: a) Electrical motor drives for industrial purpose or for
traction b) Heating equipment c) lighting equipment d) different electrical
equipments
 Loads vary with time in a 24 hour period and may also vary periodically
 Plot of variation of load demand with time is called a load curve

LOAD CURVE

P max
Load factor = Pave/Pmax

demand

P average

0.00 12.00 noon 24.00 mid night

2.a Example: During a period of 24hrs, the load demand is as follows.

Period 12 to 7 7 to 9 9 to 11 11 to 1 1 to 4 4 to 10 10 to 12
Load, 20 100 45 70 40 85 30
MW

Total energy consumed =20x7+100x2+45x2 +70x2+40x3+85x6 +30x2

=1260MWh

So, Average demand=1260/24= 52.5MW

Peak load= 100MW

Load factor=average demand/Peak Demand =52.5/100=52.5%

Installed capacity= sum of manufacturers rating of all installed generating equipment

Installed capacity ideally should be 100MW plus reserve capacity., say 20%

Capacity factor= Pave/ Installed capacity= 52.5/120= 43.75%

Capacity factor should be less or equal to load factor.

Firm capacity = capacity of system after allowing for maintenance

Peak load= it is the maximum simultaneous ultimate customer demand which occurs at any
period (day,week. Month or year)

Area under the load curve gives the amount of energy supplied to the system

At any instant of time sum of gross consumption of energy should be exactly equal to the sum
of demand of energy plus the losses in the system.

Peak Load: it is the simultaneous maximum demand on the system which may occur at any
period, day, week, month , year.

2. BASIC OBJECTIVE OF PLANNING

The basic objective of strengthening and improvement of Sub-Transmission and Distribution

system is to ensure quality and reliability of power supply to the Consumer, to bring down
technical losses to an optimal minimum value and arrange the system to match the growing
power demand in the area under study. The system shall be constructed so as to ensure:

 Voltage conditions are within permissible levels so that the technical losses can be
minimised.

 Improvement of reliability and security of power supply.

 Improvement in quality of supply.

 Adequate capacity for load growth for next 5 years.

 VOLTAGE REGULATION

The voltage drop is indicative of losses in the system. As Indian Electricity Rules provides some
voltage variation limits, these are taken as one of the guiding factors in formulating various
system improvement proposals. The Indian Electricity Rules sets the following limits for various
voltage levels of a system:

a) LT Lines :(-)6% to (+)6%

b) 11 KV & 33 KV :(-) 9% to (+) 6%

c) EHV :(-) 12.5% to (+) 10%

TECHNICAL LOSSES

The target levels and maximum tolerable loss levels for each voltage level on the system are
accepted in India as follows:

SN System Components Levels for Peak Power Loss

Target Level % Maximum
Tolerable %
1 Transformation to inter-mediate voltage level,
transmission system & step down to sub- 1.50 3.00
transmission voltage level
2 Sub-transmission system and step-down to
2.25 4.50
distribution voltage level
3 Distribution lines and service connections 4.00 7.00
TOTAL 7.75 14.50
The target levels as set may appear to be too low however all utilities have to set their own
target for lowering losses within a specified period and work towards achieving this goal. The
present T&D loss of APDCL is around 27% and it is targeted to reduce this loss by around 1.5%
per year.

3. Demand forecast and System Long Term planning

 The ability to forecast the long-term demand for electricity is a fundamental prerequisite
for the development of a secure and economic power system.
 The term forecast refers to determination of requirements for the future using a
systematic process of defining future loads in sufficient quantitative detail to permit
important system expansion decisions to be made.
 An accurate forecast requires adequate data, for without it, the results of the forecast may
be unreliable.
 The demand for electricity depends on a number of socio-economic factors such as
economic growth, industrial production, the new technological developments that
influence the life styles, and also on the Governmental policies.
 Prediction of future energy demand requires an intuitive and wise judgment and may call
for revising the estimate at regular intervals to take care of new policies, changes in
socio-economic trends. Thus, the projection may have to consider these changes into
account.
 Even with the best information, forecasting is by nature a vague science. It is open
to the occurrence of the unpredictable events that may vary the consumption
pattern. Over estimation of demand may lead to unnecessary investment in
transmission and generation assets. Though in an open and competitive market,
excess generation will tend to force electricity prices down, on the other hand the
excess investment in the infrastructure will impose additional costs to the
consumers.
 Under estimation of demand may lead to shortages of supply and infrastructure. In the
electricity market, energy prices would most likely to rise in this scenario, while system
security would be below standard. This may lead to curtailment of economic growth and
deny potential investors. Both extremes are undesirable for the electricity industry and to
the overall economy of the State.
 Comprehensive plans has to be prepared for a 5 year period with a perspective plan for
the next 10 years to match with the five year plans of the planning commission.
 For preparation of the plan projections or load forecasts for the electricity demand for
short time and long time basis is an essential pre requisite
 Depending on the assumptions considered for growth and accuracy of data
compiled from the stakeholders, the forecasts can be made realistic.
 More realistic the forecasts more pragmatic will be the Plan formulation as Optimization
of utilization of funds and scarce resources is a necessity
  For planning to be effective it should cover the generation sector, transmission sector and
the distribution sector. For every new generating station coming up, the transmission
system has to be studied to examine whether the existing transmission system will be
adequate to evacuate this power and if necessary it would need to be strengthened.
 At the same time, there should not be any constraint in the sub transmission or
distribution system also. Actually the generating stations are planned based on the
projected demand of the end users in the distribution system.
 The Central Electricity Authority is entrusted with projecting these forecasts in the form
of Electric Power Survey Reports.
 An expert committee is formed from representatives of the DISCOMS- private and state
sector, transmission cos, PGCIL, REC, various concerned ministry, experts from TERI
and other economists, Planning commissions etc
 A thorough study of the electrical energy statistics such as category wise annual energy
consumption, availability of electricity, no of consumers and connected load, no of
electrified villages, no of energized agriculture pump sets, new plans of large scale
energy intensive industries, captive plants being installed etc are collected from the
concerned utilities of each state and union territory for the last five years The Central
Electricity Authority in close co-ordination with all the power utilities of the different
states of India, makes a comprehensive study based on past trends and projects a load
demand state wise and the entire nation as a whole every five years and prepares the
Electric load survey reports.
 The 18th EPS report has been prepared to cater to the 12th annual plan from 2013-
2017 and long term planning upto 2022. Based on these reports the state utilities are to
make their own load flow studies and prepare their business plans for strengthening of the
transmission and distribution network as well as proposals for generating stations. Since
electricity is now governed by The Electricity Act 2003, commercial considerations come
into fore while making the project proposals.

4. FORECASTING METHODS

For forecasting the demand for electricity, many reliable statistical methods have been
developed which are very much effective and efficient in projecting the future demand.
The statistical models adopted are illustrated as below:

a. Trend analysis: Trending methods are widely used as a tool for forecasting which works with
historical data, extrapolating past load growth patterns into future. Trending techniques involve
fitting trend curves to basic historical data adjusted to reflect the growth trend itself.

b. Multivariable regression Technique (Econometric Method) : The econometric method

determines energy demand by considering the influence of independent variables, such as
population, income, economic growth, cost, industrial & commercial activity and also other
socio economic variables. Econometric models are estimate equations that relate electricity
demand to external factors.
c. Partial end use method: The end-use method determines energy demand through total kWh
use from all of the electrical appliances used in the households. The forecast is “built from the
sum of end using activities”. End-use models must include kWh consumption data by type of
equipment or process. This method is most readily applied to the residential sector, because
the data required for this forecast include: forecast year, number of residential customers,
residential housing stock or commercial buildings, industrial process data, major appliances,
and kWh use per appliance. In the basic form, this model is simple accounting procedure that
enumerates the end uses and adds the electricity use for each end use of its components. In
view of practical difficulties in assessing the end usage of equipments, the method of partial
end use model is being increasingly used. This method is being followed by CEA in its
forecasting methodology in the EPS. In this methodology the specific consumption of each of
the category is being assessed based on the past sample and correction is made to account for
changing scenario

5. Example of Trend Analysis

Mathematical cum statistical method of load forecasting

In this method it is assumed that the demand for power increases in an exponential manner
over a limited period of years. This assumption seems to be very close to the actual conditions
for short term load growth.
If, D= maximum demand in MW or KW , Yo is base year and Y= the year in which load is to be
predicted then ,
D= exp [a+b(Y-Yo)], where a and b are constants
a=1/n[ΣVi] and b=[ΣVi*Yi]/ [ΣYi*Yi]
and V1=lnD1, V2=lnD2 ……Vn=lnDn and D1, D2 are the
demands for consecutive years.
Consider the following the following example:
The maximum demand of one state is available from 2001 to 2008 . to find the maximum
demand in 2013.

Year Max D’ =D/1000 Vi=ln D’ Yi Yi*Yi Vi*Yi

Demand D
in MW
2001 6166 6.166 1.8109 -3 9 -5.4327
2002 7020 7.020 1.9488 -2 4 -3.8976
2003 9074 9.074 2.2054 -1 1 -2.2054
2004 10634 10.634 2.3736 0 0 0
2005 12541 12.541 2.5283 1 1 2.5283
2006 14106 14.106 2.6334 2 4 5.2668
2007 16135 16.135 2.7682 3 9 8.3046
Sum 16.2939 0 28 4.6051

.let D’=D/1000, let us take base year as 2004 , so that ΣYi =0

a =[ΣVi]/7 =16.2939/7=2.3277
b=[ΣVi*Yi]/[Yi*Yi]=4.6051/28= 0.1644

Max demand in 2013= 1000exp[a+b(Ynew-Ybase)]

= 1000exp[2.3277+0.1644(2013-2004)]
=1000exp 3.8073
= 45028MW
6. Transmission and Distribution Systems Planning.
 Once the load forecast study is completed, addition of generating stations is finalized, the
planning for transmission and distribution system has to be carried out.
 First the existing transmission system database has to be prepared based on data from
extensive field work for assessment of loading of lines and sub stations, losses, voltage
profile fault level. A load flow study is conducted first on the base case with the existing
demand. Next the result of the load forecast in the next five years is applied on the base
case and a load flow study is conducted. This study will give a clear picture of the
voltage profile , loading on the lines and substations etc.
 Based on an analysis of the results the new sub station and lines required can be
determined and the load flow study is again conducted . This is done several times until a
system emerges where the stipulated values as per grid code and the transmission
planning criteria of the CEA is met.

Finally some critical studies are conducted to examine the condition of overloading or
changes in voltage profile when a breakdown occurs in a line or interconnecting transformer
 when regular shutdown for maintenance is being done on one line/transformer. These
criticality studies are very important in system study, to avoid Transmission system failure.
Following are checked for inadequacy of the back up transmission system.

 Overloading of transformers.
 Overloading of lines.
 Requirement of reactive compensation.

7. Checking of adequacy of DISTRIBUTION SYSTEM

Sizing of the transformer capacity of a sub station.
.
 Suppose the 33/11 kV existing 2x5MVA,sub station handles a peak load = 3MW say

 A new industries will be connected to this SS with load = 1.5MW, say

 Additional load due to normal load growth in the next five years as projected from load growth
of the previous five years of the different categories of consumers

 say average load growth per year=6%,

Load in 1st year=P1

Load in 2nd year=P2=P1+P1x0.06=P1(1+.06)
Load in 3rd Year = P3= P2+P2x0.06=P2(1+.06)= P1(1+0.06)2
……………………………..
 Load in 5th year = P5=P1(1+0.06)5-1so load at the end of five years = 3.x1.06^4 =3.79MW

 Total load=3.79+1.5=5.29MW considering economical loading of 70% and PF of 0.85, capacity

required will be= 5.29/.85=6.22, 6.22/.7= 8.88MVA

 Loading on existing transformer will be =8.88/2x5=88% ie it will be overloaded.

 so if one 5mva is replaced by a 10mva, loading will be= 8.88/15=59%. And if both are replaced by
10MVA, then transformers will have loading of 44% only. In the event of outage of the 10MVA
transformer, the 5 MVA transformer will not be able to cater to entire load.

 So, S/S Transformer capacity is selected as 2x10MVA considering 100% redundancy to take care
of outage of transformer.

Voltage regulation of 33kv Feeder

1) Peak load of 33kV feeder =10MW,

2) Conductor used=ACSR Raccoon, Length=20kM

3) Resistance/Km, for ACSR Raccoon= 0.432ohm, at 600 C

4) Reactance/kM, for ACSR Raccoon= 0.3891ohm,

5) (for triangular configuration of conductor :horizontal distance-1525mm and vertical 1325mm )

6) Current =I=10000/1.732x33x0.85=205 Amp at 0.85pf

7) Voltage drop =Vd= v3xI(Rcos Φ + Xsin Φ) L

8) =v3x205x(0.432x.85+0.3891x0.526)x20=4060Volts

9) %Voltage Regulation =voltage dropx100/(V-Vd) =4060x100/(33000-4060) =14.03%.

10) If peak load increases to 15 MW then Vd=1.732x308x0.5718x20= 6094Volts

11) Voltage Regulation=6094x100/(33000-6094)=22.64%

12) So the feeder becomes overloaded

13) We may consider higher capacity conductor by using ACSR Wolf conductor in place ACSR
Raccoon

14) Voltage drop=1.732x308x(0.85x0.1871+0.526x.3866)x20 = 3865Volts

15) Voltage Regulation= 3865x100/29135= 13.26%, so the reconductoring will not be effective.

(here resistance of ACSR Wolf= 0.1871 ohm and Reactance of ACSR Wolf= 0.3866ohm)