Transcript ES Lecture 3 by TS (Energy demand) 6 Jun 2015

EE5003: Energy Systems
3: Energy Consumption and Demand
Tilak Siyambalapitiya
June 2015
1
Lecture 1: Energy Consumption and
Demand
Contents
 Energy consumption in developed countries
and developing countries
 Regional consumption patterns
 Sectoral consumption
 Per-capita consumption
 Global/Sri Lanka demand for energy
 Demand growth and forecasting
 Energy and the economy
2
Sri Lanka Primary Energy Supply: 2013
1980
Total primary energy = 4.8 million TOE
Biomass
Petroleum
Large Hydro
2010
Total primary energy = 10.8 million TOE
Biomass Petroleum Coal
Large Hydro New Renewable Energy
3
Sri Lanka’s Energy Supply:2013
Status: year 2013
Sri Lanka’s Secondary Energy Supply:2013
What is primary energy? What is secondary energy ?
Secondary energy is the final or intermediate form in which energy is delivered to end-use
customers. eg: electricity, charcoal, diesel, gasoline.
Losses occur in refining, power generation, T&D.
What was the conversion efficiency from primary to secondary energy in 2013 ?
Why have petroleum product demand reduced in 2013, when compared with 2012?
Similarly biomass ?
Status: year 2013
Where is the energy used? (2013)
Is energy use in households, efficient and productive ?
Households and commercial are the largest single consuming group, using almost
50% of the secondary energy.
Household end-use efficiency is low (use of firewood). Thus they need a larger
secondary energy input to serve the losses as well.
Status: year 2013
Do the indicators show that Sri Lanka’s
Energy Efficiency is improving at Macro-level
ENERGY INTENSITY VARIATIONS
4.5
Commercial Energy Index
4.0
GDP Index (INDEX 1984=100%)
3.5
Primary Energy Index
2.5
2.0
1.5
1.0
0.5
2013
2011
2009
2007
2005
2003
2001
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
0.0
1977
Index
3.0
Year
7
Read more on Sri Lanka’s Energy Supply and
Demand, trends and future outlook
Sri Lanka Energy Balance 2013
www.energy.gov.lk
Worldwide Primary Energy
Consumption: Americas
Consumption by fuel*
Oil
Million tonnes oil equivalent
US
Canada
Mexico
Total North America
Argentina
Brazil
Chile
Colombia
Ecuador
Peru
Trinidad & Tobago
Venezuela
Other S. & Cent. America
Total S. & Cent. America
Natural
Gas
Coal
Nuclear
Energy
Hydro
electricity
Renewables
2010
Total
850.0
102.3
87.4
1039.7
621.0
84.5
62.0
767.4
524.6
23.4
8.4
556.3
192.2
20.3
1.3
213.8
58.8
82.9
8.3
149.9
39.1
3.3
1.7
44.2
2285.7
316.7
169.1
2771.5
25.7
116.9
14.7
11.0
10.6
8.4
2.1
35.2
57.3
282.0
39.0
23.8
4.2
8.2
0.4
4.9
19.8
27.6
5.0
132.9
1.2
12.4
3.7
3.8
0.5
^
2.1
23.8
1.6
3.3
4.9
9.2
89.6
4.9
9.1
2.0
4.4
17.4
20.7
157.2
0.4
7.9
0.8
0.2
0.1
0.1
^
1.7
11.1
77.1
253.9
28.4
32.2
13.0
18.3
22.0
80.3
86.8
611.9
9
Worldwide Primary Energy Consumption: Europe
Consumption by fuel*
Oil
Million tonnes oil equivalent
Austria
Azerbaijan
Belarus
Belgium & Luxembourg
Bulgaria
Czech Republic
Denmark
Finland
France
Germany
Greece
Hungary
Republic of Ireland
Italy
Kazakhstan
Lithuania
Netherlands
Norway
Poland
Portugal
Romania
Russian Federation
Slovakia
Spain
Sweden
Switzerland
Turkey
Turkmenistan
Ukraine
United Kingdom
Uzbekistan
Other Europe & Eurasia
Total Europe & Eurasia
13.0
3.3
6.6
35.0
4.2
9.2
8.7
10.4
83.4
115.1
18.5
6.7
7.6
73.1
12.5
2.7
49.8
10.7
26.3
12.6
9.1
147.6
3.7
74.5
14.5
11.4
28.7
5.6
11.6
73.7
5.0
28.3
922.9
Natural
Gas
9.1
5.9
17.7
17.4
2.3
8.4
4.5
3.5
42.2
73.2
3.3
9.8
4.8
68.5
22.7
2.8
39.2
3.7
12.9
4.5
12.0
372.7
5.1
31.0
1.4
3.0
35.1
20.4
46.9
84.5
41.0
14.1
1023.5
Coal
2.0
^
^
4.9
6.6
16.0
3.8
4.6
12.1
76.5
8.5
2.6
1.4
13.7
36.1
0.2
7.9
0.5
54.0
3.4
6.2
93.8
2.7
8.3
2.0
0.1
34.4
36.4
31.2
1.3
15.7
486.8
Nuclear
Energy
10.9
3.5
6.3
5.2
96.9
31.8
3.6
0.9
2.6
38.5
3.3
13.9
13.2
6.0
20.2
14.1
1.8
272.8
Hydro
electricity
7.8
0.8
^
0.1
1.3
0.8
^
3.2
14.3
4.3
1.7
^
0.1
11.2
1.5
0.3
^
26.7
0.8
3.8
4.6
38.1
1.3
9.6
15.1
8.2
11.7
2.9
0.8
2.5
22.3
195.9
Renewables
1.4
^
^
1.5
0.2
0.6
2.5
2.2
3.4
18.6
0.6
0.7
0.7
5.6
0.1
2.2
0.3
1.9
2.8
0.1
0.1
0.1
12.4
4.3
0.3
1.0
^
4.9
1.2
69.6
2010
Total
33.3
10.0
24.4
69.8
18.0
41.3
19.5
29.1
252.4
319.5
32.5
23.4
14.6
172.0
72.8
6.1
100.1
41.8
95.8
27.1
34.5
690.9
16.2
149.7
50.7
29.0
110.9
26.0
118.0
209.1
49.8
83.4
2971.5
10
Worldwide Primary Energy Consumption: ME and Africa
Consumption by fuel*
Oil
Natural
Gas
Coal
Nuclear
Energy
Hydro
electricity
Renewables
2010
Total
Million tonnes oil equivalent
Iran
Israel
Kuwait
Qatar
Saudi Arabia
United Arab Emirates
Other Middle East
Total Middle East
86.0
11.2
17.7
7.4
125.5
32.3
80.2
360.2
123.2
4.8
12.9
18.4
75.5
54.5
39.6
329.0
1.1
7.7
8.8
-
2.2
0.9
3.0
0.1
^
^
^
0.1
212.5
23.7
30.6
25.7
201.0
86.8
120.7
701.1
Algeria
Egypt
South Africa
Other Africa
Total Africa
14.9
36.3
25.3
79.0
155.5
26.0
40.6
3.4
24.4
94.5
0.3
0.7
88.7
5.7
95.3
3.1
3.1
^
3.2
0.3
19.6
23.2
0.3
0.1
0.7
1.1
41.1
81.0
120.9
129.5
372.6
11
Worldwide Primary Energy Consumption: ME and Africa
Consumption by fuel*
Oil
Million tonnes oil equivalent
Australia
Bangladesh
China
China Hong Kong SAR
India
Indonesia
Japan
Malaysia
New Zealand
Pakistan
Philippines
Singapore
South Korea
Taiwan
Thailand
Vietnam
Other Asia Pacific
Total Asia Pacific
42.6
4.8
428.6
16.1
155.5
59.6
201.6
25.3
6.9
20.5
13.1
62.2
105.6
46.2
50.2
15.6
13.5
1267.8
Natural
Gas
27.3
18.0
98.1
3.4
55.7
36.3
85.1
32.2
3.7
35.5
2.8
7.6
38.6
12.7
40.6
8.4
4.8
510.8
Coal
43.4
0.5
1713.5
6.3
277.6
39.4
123.7
3.4
1.0
4.6
7.7
76.0
40.3
14.8
13.7
18.9
2384.7
Nuclear
Energy
16.7
5.2
66.2
0.6
33.4
9.4
131.6
Hydro
electricity
3.4
0.3
163.1
25.2
2.6
19.3
2.1
5.5
6.4
1.8
0.8
0.9
1.2
6.3
7.4
246.4
Renewables
1.5
12.1
^
5.0
2.1
5.1
^
1.8
2.3
0.5
1.0
1.1
^
32.6
2010
Total
118.2
23.6
2432.2
25.9
524.2
140.0
500.9
62.9
18.9
67.6
27.6
69.8
255.0
110.5
107.9
44.0
44.6
4573.8
12
Worldwide Primary Energy Consumption:
Summary
Consumption by fuel*
Oil
Million tonnes oil equivalent
Total World
of which: OECD
Non-OECD
European Union
Former Soviet Union
4028
2114
1914
662
202
Natural
Gas
2858
1398
1461
443
537
Coal Nuclear Hydro RenewEnergy electricity ables
3556
1104
2452
270
169
626
521
105
208
59
776
309
466
83
56
159
123
36
67
0
2010
Total
12002
5568
6434
1733
1023
13
ENERGY AND THE ECONOMY

ENERGY IS A CATALYST FOR ECONOMIC
GROWTH

A VITAL INPUT TO MANY PRODUCTS AND
SERVICES

WIDE DISPARITIES EXIST IN PER CAPITA
CONSUMPTION
14
USEFUL DEFINITIONS
GROSS DOMESTIC PRODUCT AND GROSS
NATIONAL PRODUCT
The summation of value added by all the sectors of
an economy is known as the Gross Domestic
Product (GDP).
The estimate of GDP is limited to activities
occurring within the country.
The net income from abroad is added to GDP to
calculate Gross National Product (GNP).
15
GDP AND GNP
Expressed in two currency terms
When measured in the currency of the year, GDP is
given in current terms.
When measured in the currency of a given reference
year, GDP is given in constant terms.
The data in constant terms, which do not have the
effect of inflation, is useful to estimate the real growth
of the economy or its sectors.
16
ENERGY INTENSITY IN THE ECONOMY
Energy input
Energy Intensity 
GDP
ALSO KNOWN AS THE SPECIFIC ENERGY
CONSUMPTION OF THE ECONOMY
MEASURED IN A COMMON UNIT SUCH AS TONS
OF OIL EQUIVALENT (TOE) PER MILLION RUPEES.
17
Sri Lanka Energy and the Economy
GDP in 1982 costs
(LKR million)
Total Primary Energy
Use (thousand toe)
Energy intensity
(toe/million LKR of
1982 costs)
1990
1995
2000
2005
2010
129,244
167,953
214,422
259,885
353,205
5,914
6,984
8,802
9,844
11,034
45.8
41.6
41.0
37.9
31.2
18
Energy-Economy Elasticities
E  k .( GDP ) 1 .( X )  2 .( Y )  3
E
 ( GDP )
 k .  1 .( GDP ) (1 1) .( X ) 2 .( Y ) 3
E
E
 1 .
 ( GDP )
( GDP )
1  Income Elasticity of Demand
Ln ( E )  Ln ( k )  1 Ln ( GDP)
19
TYPICAL VALUES OF
INCOME (GDP) ELASTICITY OF ENERGY
DEMAND
 Developing
countries usually have high GDP
elasticities of energy demand.
 Developed
countries are more likely show GDP
elasticities of about 1.0.
A
strong emphasis on energy conservation and
a move towards less energy intensive industries,
may even cause the elasticity to fall below 1.0.
20
CONCERNS OF POLICY/DECISION MAKERS

What is the optimum level of petroleum prices
with respect to other alternatives, such as wood
fuel and electricity ?

What is the technical and practical feasibility of
substituting petroleum products with other forms
of energy ?

What is the price-elasticity of demand for each
petroleum product ?

What are the cross-price elasticities between
petroleum products ?

How does the income of the users affect the
demand for these products ?
21
Price Elasticity of Demand
Self-price Elasticity
Sensitivity of the demand for an energy product to its
own price
Pi Priceperunit ofenergyproduct
i
Ei  Demandfortheenergyproduct
i
iiP self - priceelasticityofdemand for
iiP
ii
 0
 E i

 Ei
 
 P i

 Pi

i,



i
i
 or E  ii . P
P

Ei
Pi



i i
 1
0. 5 

ii
Highly price elastic
 0
Price inelastic
22
Price Elasticity of Demand (Contd.)
Cross-price Elasticity
Sensitivity of the demand for an energy product to to the
price of another energy product.
Pj  Price per unit of energy product j
E i  Demand for the energy product i
ijP  cross  price elasticity of demand on i by j ,
ijP
 E i

 Ei
 
 P j

 Pj




i
j
 or E  ij . P
P

Ei
Pj



23
EXAMPLE
Self- and cross- price elasticities
of an energy market
Gasoline
Gasoline
Diesel
Kerosene
Electricity
LP Gas
-1.4
+0.8
+0.3
+0.1
+0.2
Diesel
+0.3
-0.9
+0.5
+0.2
0.0
Kerosene
+0.1
+0.1
-0.2
0.0
+0.1
Electri- LP Gas
city
0.0
+0.2
+0.1
-0.7
+0.2
+0.1
0.0
+0.1
+0.1
- 0.5
i  row , j  column ,
ij  Change of demand for i ,
attributed to a change in price of j
24
Income Elasticity of Demand
Sensitivity of the demand for an energy
product to the income of its users (or the
country).
Y  Income of energy consumer
E k  Demand for the energy product k
kY  Income elasticity of demand for k ,
 E k 

 Ek 

E k


k
k Y
Y 
or


Y.
k

Y
E
Y




 Y 
Usually ,
kY  1
25
Income Elasticity of Demand (Contd.)
Defining E k as the per capita demand
for energy product k , and Y as the per capita income ,
Income elasticity of demand,
 Ek 


 Ek 

Yk  
 Y 


 Y 


Energy Product
Electricity
Kerosene
LPG
Fuelwood
IED
0.1
- 0.2
0.2
- 0.4
26
ENERGY DEMAND FORECASTING
The future is uncertain,
and all forecasts, therefore,
will be finally proved to be
inaccurate !!!
27
THEN WHY PREPARE FORECASTS ?
PLANS REQUIRE THEM !!!
 To develop capacity expansion plans. eg:- Power stations,
transmission lines, oil refineries, fuelwood supplies.
 To develop financial plans for supply institutions. eg:Investment plans, revenue forecasts, pricing studies
 To develop Institutional Plans
 For various other studies
Management
eg:- Manpower
eg:- Energy
28
FORECASTING
TIME-FRAME
 Long Term (15-30 years) for planning oil
exploration, power stations or firewood supplies
 Medium Term (5-15 years) for facilities planning.
eg. oil refinery, power transmission planning,
pricing studies.
 Short Term (1-5 years) for distribution planning,
cashflow analysis, budgeting
 Very Short Term (from the next hour up to 1
year) for operations planning of energy facilities.
29
COVERAGE OF VARIOUS
FORECASTS
Forecast
Window
Very short
Short
Time-frame
Likely Application
Next hour to
1 year
Operations planning
1 to 5 years
Operations planning ,
financial planning,
eg. Economic load
dispatch in a power
system, optimum refinery
operations
budgeting, decisions on
short-term investments
Medium
5 to 15 years
Firm decisions on major
new investments
eg. new refinery, new
baseload thermal plant
Long
15 to 30 years
Expansion planning,
siting decisions
30
FORECASTING TECHNIQUES

Time Trend Analysis

Time Series Analysis

Econometric Methods

End-use Methods
Combinations of two or more of the above
techniques, are also common. These are
known as hybrid techniques.
eg:- Econometric end-use technique
31
TIME-TREND ANALYSIS
Year Sales (GWh)
1990
2395
1991
2567
1992
2766
1993
3207
1994
3482
1995
3818
1996
3552
1997
3971
1998
4460
1999
4754
2000
5188
Year Sales (GWh)
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
5178
5454
6160
6599
7201
7766
8169
8350
8372
9191
The 20-year compound growth rate is,
1
 9191 20
 1  0.0695o r 6.95%


2395


32
TIME-TREND ANALYSIS (CONTD.)
LINEAR TREND MODEL
S  a. T  b
S  349 .2  T  693,136
ADJUSTED R2  0.975
EXPONENTIAL TREND MODEL
S  e ( k .T  l )
S  e 0.0672T 125.9 
ADJUSTED R2  0.987
Compoundgrowthrate  6.95% per year
Linear trend  349.2 GWh/year
Exponential Trend = 6.72% per year
33
TIME-TREND ANALYSIS (CONTD.)
10,000
Actual
Linear fit
Exponential Fit
9,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
-
1990
Electricity Sales (GWh)
8,000
34
TIME-SERIES ANALYSIS
Development of a function or functions to describe
the quantity to be forecast, by means of its own
values in the past.
DT  a1 DT 1  a2 DT 2 .......... an DT n
DT is the demand in time T . a1 , a2 ,..... an are constants .
Time series analysis addresses issues related to a
shorter time-step. This is because cyclic
patterns are more evident on time steps shorter than
1 year.
35
TIME-SERIES ANALYSIS-
MOVING AVERAGE AND EXPONENTIAL
SMOOTHING
MOVING AVERAGE
FT
1

N
T 1
A
i
i (T  N )
EXPONENTIAL SMOOTHING
FT  F( T 1)  s.[ A( T 1)  F( T 1) ]
Daily
Average
demand
(kl)
550
530
510
490
470
450
430
410
Actual
390
Exponential smoothing applied
370
Moving average of 5 values
350
36
0
12
24
36
48
60
72
TIME-SERIES ANALYSIS-
AN AUTOREGRESSIVE MODEL
 DT 1
Ln( DT )  a.Ln( DT 12 )  b.Ln
 D
T 13



c

Daily Average
demand (kl)
550
530
510
490
470
450
430
410
390
370
350
Actual
Autoregressive model
0
12
24
36
48
60
72
37
ECONOMETRIC MODELS
Analysis of historic correlation between energy demand and
other economic variables and use such relationships to
project future demand.
A typical econometric equation
Demand    1( GDP ) 
 2 ( Population )   ( Energy price )
3
EXOGENOUS VARIABLES (Independent of the demand)
GDP, Population and Energy price
ENDOGENOUS VARIABLES (Determined within the model)
38
ECONOMETRIC MODELSDRIVING VARIABLES

Macroeconomic Variables:- Economic activity, population,
value added in industry, number of households, vehicles
population, degree of rural electrification etc.

Energy Prices

Seasonality:- Weather data (temperature, relative humidity,
wind speed, rainfall) and other factors giving rise to cyclic
performance
A complete econometric model to forecast energy demand will
consist of a number of equations, with at least one equation
for each energy product. There have to be an adequate number
of equations to forecast endogenous variables.
39
Eg: DEMAND FOR
TRANSPORT FUELS
E  a . G  b. P  c.( T  1982 )  d
F  transport fuels sold ( thousand MT )
G = Gross domestic product in constant terms
( million Rs )
P = population ( million )
T = time ( year )
E  148.76  0.0044  G
{ R 2  0.94,
t - statistics were 3.399 and 11.965, respectively }
R-SQUARED and t-statistics
are acceptable.
40
DEMAND FOR
TRANSPORT FUELS
A DIFFERENT MODEL
E   . G 1 . P 2 .( T  1982 )  3
LINEAR FORM OF THE ABOVE
Ln( E )  Ln(  )  1 . Ln( G ) 
 2 . Ln( P )   3 . Ln( T  1982 )
RESULTS OF REGRESSION ANALYSIS
Ln ( E )  -2.418  0.764  Ln ( G )
{R 2  0.942, t  statistics  3.32 and 12.24, respectively }
ie E  0. 089  G 0.764
R-SQUARED and t-statistics are acceptable.
(National) Income elasticity of demand for transport fuels is
0.764.
41
OTHER MODELLING AND FORECASTING
TECHNIQUES
End-use Methods
The demand forecast moves to the lowest
possible level of disaggregation, often to the
level of the device that converts energy to its
useful service.
42
OTHER MODELLING AND FORECASTING
TECHNIQUES
Informed Opinion (Judgemental)
Eg:- A group of knowledgeable persons may jointly discuss and
understand all the underlying issues, study contributing factors
such as historic price-elasticity of demand and world population
growth, to make a judgemental forecast of the demand for crude
oil in year 2020.
43