List of non-admissible solutions (reveals loads not energized; transformers capacity and heat capacity of all the branches not satisfied; bus voltage limits ...
INESC Coimbra - Jornadas 2009 - Innovative ideas toward the Electrical Grid of the Future -
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems 29
28 28
30 29
36
37 36
27
31 30
32 31
38 37
33 32
39 38
34 33
40 39
35 34
41 40
42 41
43 42
44
45
43
44
Presented by:
46 45
35
52
51 51
69
71
50
1
2
SE
1
3 2
4
5 4
3
6 5
7 6
8 7
9 8
10
11 10
9
12 11
13 12
14
15
16
17
15
14
13
16
18 17
19 18
20 19
21
22 21
20
23 22
24 23
25 24
26 25
27 26
● Romeu M. Vitorino, (Ph.D. Student)
70 46
52
47
48 47
49 48
67
65
68
50
69 68
73
49
66 66
53
54 53
55 54
72
Supervisors:
67
● Luís P. Neves, (Ph.D.) 56
55
57 56
58 57
59 58
60 59
61 60
62 61
63 62
64 63
65 64
● Humberto M. Jorge, (Ph.D.)
Introduction Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
Proposal of a new method to improve reliability and also efficiency
(minimization of active power losses) in radial distribution systems through a process of network reconfiguration; To evaluate reliability is used the Monte Carlo simulation (MC) and an
historical data of the network (level of reliability and the severity of potential contingencies in each branch); Two Perspectives of optimization:
• 1st (no investment) – using only the switches presented in the network; • 2nd (with investment) – possibility to place a limited number of tieswitches (only in certain branches), defined by a decision agent.
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
2
Genetic Algorithm Approach (IGA) Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
– Main Features of the IGA (Improved Genetic Algorithm) ● List of non-admissible solutions (reveals loads not energized; transformers capacity and heat capacity of all the branches not satisfied; bus voltage limits violation); ● Two-termination criterion (Number of generations and a convergence threshold); ● Adaptive Crossover and Mutation probabilities according to the genetic diversity in the population; ● Suitable coding and decoding technique (obtains a small chromosome length and can assure the radiality of the network); ● Elitist selection (the best individual at the early generation (k) is maintained in the next generation (k+1)). Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
3
Distribution System Reliability Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
– Branch Reliability A fundamental element of the network to assure continuity of service.
To define the reliability level of each network branch it was considered the following: Level 1 2 3 4
Failure Very unlikely Unlikely Likely Very likely
Historical information about component failures
To better characterize the possible contingencies, were defined
degrees of severity (based on historical data) each with different average interruption durations (“Dav”);
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
4
Distribution System Reliability Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
– Monte Carlo Simulation method The method considered N trials and, in each trial, an annual number of
contingencies, in locations according to the reliability level of each branch;
The interruption duration of each contingency is variable and follows a
normal density curve with a standard deviation (σ) and average (μ). 1 0.9
f ( x; μ , σ ) = e
− ( x − μ )2 2σ
2
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
4
6
8
10
12
14
16
18
20
Estimation of the annual reliability indexes in the considered network
configuration. In this study, the total energy not distributed (TEND) is used as the reliability index to minimize.
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
5
Evaluation of the solutions Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
– Multi-objective fitness assignment A fitness function we want to minimize is used to evaluate the
performance of the solutions and considers two objectives:
fitness = (α1WLoss + α 2TEND ) × 100 1st part – represents the annual active energy loss (WLoss); 2nd part – represents the annual total energy not distributed (TEND),
obtained through the MC simulation.
Due to the difference between the values of WLoss and TEND it was
used a normalization method (parameters α1 and α2) also capable to reflect the importance of both objectives to the decision agent.
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
6
Case Study – MV network (12.66 kV) Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
Characteristics: Restricted number of branches for tie-switch placement !
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
11
Conclusions Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
Possibility of obtaining network solutions that allow the simultaneous
optimization of losses and reliability considering different weights; Good performance of IGA (convergence speed and stability increased)
due to the introduction of new features (dynamic “pc” and “pm”, …); Two-perspective approach was used, giving to the decision agent the
possibility to compare the benefits achieved with both solutions (with and without investment); Through network reconfiguration it is possible to make the distribution
system more efficient (in terms of load balancing and voltage stability) considering only the reduction of the electrical losses.
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
12
Perspectives of future developments Jornadas INESCC 2009 - Innovative ideas toward the Electrical Grid of the Future -
Inclusion of new objectives in the field of reliability (minimization of
number and duration of the interruptions (NI,DI) according to the geographic zone of each consumer); Annual energy loss estimation considering the daily load diagram of
each consumer (LV or MV); Consider different network configurations throughout the year (Winter,
Summer and Half-Season); Use of Multi-objective optimization techniques (MOO) allowing to
capture multiple solutions of the Pareto front (“Decision Maker” will decide after search).
Network Reconfiguration to Improve Reliability and Efficiency in Distribution Systems
