Area
|
Year
|
2004
|
2019
|
2010
|
2011
|
2012
|
2013
|
2014
|
2015
|
2016
|
2017
|
2019
|
2021
|
X^3
|
X^2
|
X
|
R^2
|
2021
LF
|
A
|
MW
|
18
|
26
|
29
|
31
|
33
|
35
|
39
|
46
|
49
|
53
|
58
|
66
|
0.3
|
-4
|
32
|
99.6%
|
62%
| | |
LF
|
42%
|
46%
|
46%
|
48%
|
50%
|
52%
|
51%
|
52%
|
55%
|
56%
|
59%
|
62%
|
|
|
|
|
|
B
|
MW
|
58
|
96
|
117
|
127
|
134
|
161
|
190
|
232
|
271
|
312
|
386
|
454
|
1.7
|
-21
|
187
|
99.5%
|
57%
| | |
LF
|
52%
|
61%
|
62%
|
60%
|
61%
|
58%
|
59%
|
60%
|
59%
|
58%
|
57%
|
57%
|
|
|
|
|
|
C
|
MW
|
18
|
40
|
42
|
47
|
51
|
55
|
107
|
182
|
212
|
239
|
291
|
358
|
0.3
|
7.3
|
-25
|
99.0%
|
45%
| | |
LF
|
55%
|
44%
|
47%
|
47%
|
46%
|
49%
|
42%
|
41%
|
40%
|
43%
|
44%
|
45%
|
|
|
|
|
|
D
|
MW
|
228
|
293
|
302
|
357
|
394
|
429
|
470
|
518
|
575
|
623
|
704
|
793
|
2.5
|
-32
|
333
|
99.5%
|
68%
| | |
LF
|
53%
|
62%
|
66%
|
58%
|
59%
|
59%
|
61%
|
63%
|
65%
|
66%
|
68%
|
68%
|
|
|
|
|
|
E
|
MW
|
20
|
31
|
35
|
38
|
40
|
43
|
45
|
50
|
54
|
56
|
64
|
71
|
0.3
|
-5
|
45
|
98.7%
|
57%
| | |
LF
|
54%
|
58%
|
55%
|
55%
|
57%
|
57%
|
57%
|
58%
|
59%
|
59%
|
59%
|
57%
|
|
|
|
|
|
F
|
MW
|
15
|
25
|
28
|
31
|
33
|
35
|
37
|
43
|
47
|
50
|
57
|
68
|
0.3
|
-5
|
41
|
99.4%
|
63%
| | |
LF
|
56%
|
57%
|
57%
|
55%
|
58%
|
61%
|
62%
|
63%
|
63%
|
64%
|
64%
|
63%
|
|
|
|
|
|
G, H
|
MW
|
35
|
61
|
68
|
74
|
79
|
85
|
91
|
97
|
104
|
129
|
156
|
175
|
1
|
-17
|
128
|
99.1%
|
56%
| | |
LF
|
52%
|
59%
|
52%
|
57%
|
57%
|
59%
|
59%
|
60%
|
61%
|
56%
|
55%
|
56%
|
|
|
|
|
|
P1a
|
MW
|
53
|
56
|
68
|
68
|
69
|
71
|
73
|
74
|
77
|
78
|
82
|
85
|
0.4
|
-10
|
85
|
97.3%
|
73%
| | |
LF
|
57%
|
74%
|
67%
|
69%
|
69%
|
70%
|
70%
|
72%
|
73%
|
73%
|
73%
|
73%
|
|
|
|
|
|
P2
|
MW
|
50
|
70
|
70
|
70
|
70
|
70
|
70
|
90
|
100
|
120
|
140
|
170
|
1.1
|
-18
|
120
|
97.2%
|
44%
| | |
LF
|
21%
|
39%
|
43%
|
46%
|
46%
|
46%
|
46%
|
46%
|
46%
|
45%
|
44%
|
44%
|
|
|
|
|
|
P3
|
MW
|
3
|
28
|
28
|
32
|
36
|
40
|
43
|
47
|
51
|
55
|
55
|
55
|
y = 87.024ln(x) - 6.6143
| | | |
43%
| | |
LF
|
0%
|
20%
|
39%
|
40%
|
40%
|
40%
|
40%
|
42%
|
43%
|
43%
|
43%
|
43%
|
|
|
|
|
|
P4
|
MW
|
0
|
0
|
0
|
0
|
0
|
0
|
0
|
10
|
20
|
30
|
50
|
50
|
0.1
|
25
|
-7
|
97.3%
|
32%
| | |
LF
|
|
|
|
|
|
|
|
27%
|
29%
|
30%
|
30%
|
32%
|
|
|
|
|
|
Region
|
MW
|
465
|
666
|
735
|
786
|
855
|
935
|
1064
|
1269
|
1426
|
1592
|
1868
|
2145
|
8.2
|
-109
|
991
|
99.4%
|
63%
| | |
LF
|
53%
|
62%
|
63%
|
63%
|
62%
|
63%
|
63%
|
63%
|
63%
|
63%
|
63%
|
63%
|
|
|
|
|
|
The aim is to predict the peak load, as well as, the energy demand of the
regional utility for 10 years from the current year; with a time step of 1 year.
Best fit model Load = a x^3 + b x^2 + c x + d (x = year)
Subarea Loads
Urban
Residential
Commercial
Public
Small industrial
Distribution losses
Rural
• Residential
• Agricultural
• Others (small industrial, public, etc.)
Large customers
(> 1 MW)
Rural loads
- The residential part may be estimated based on the estimated number of homes and the estimated power consumption of each home.
- The agricultural part is determined based on the estimated number of wells, their average depths and their average water flows.
- The remaining part of the rural types of the loads should also be estimated., sometimes, a fixed percentage (say 25%) may be considered.
- Total Demand (TD) is the sum of: “TD = SD + LC + IET + FD + IL + SL + AD”
- The Supplied Demand (SD),
- Load Curtailment (LC), The loads interrupted based on load shedding scheme.
- Import/Export Transactions (IET), availability to be confirmed for future model.
- Frequency Drop term (FD), the system operator has intentionally dropped the frequency to compensate, somewhat, the generation deficiency
- Interrupted Loads (IL), The loads interrupted based on some types of contracts.
- System Losses (SL), that may be affected by network expansion activities, and
- Auxiliary Demand (AD) of the power plants,
Using the prediction Model - Using a standard software and based on historical data, we need, initially, find out the driving parameters for the load. For instance, GDP, population, per capita demand and average electricity price may be four main driving parameters.
- However, other parameters may also be tried and checked. If not considered, we have, implicitly, assumed that they are either non-driving parameters or there are some types of correlations between them and those already observed.
- Various scenarios may be checked. For instance, one scenario may be considered as the load being dependent on GDP and population, only. Other combinations may be tried as new scenarios. Various fitting procedures and models may also be checked. These are, typically, available in commercial software.
- New scenarios may be generated with weighted driving parameters. For instance, a driving parameter may also be given a higher weighting in comparison with another.
- A scenario may also be generated by a combination of already generated scenarios, weighted based on their respective accuracies which are already checked.
- We should use a procedure for checking the method accuracy. If the historical data is available for the last 15 years, we may use the results of the first 10 years for producing the model. Thereafter, its prediction behavior may be checked for the next 5 years, using actual data. Once done and approved, the best model may be used to forecast the loads of the coming years.
Share with your friends: |
