A novel multi-objective PSO for electrical distribution system planning incorporating distributed generation

Abstract

This paper presents a novel particle swarm optimization (PSO) based multi-objective planning approach for electrical distribution systems incorporating distributed generation (DG). The proposed strategy can be used for planning of both radial and meshed networks incorporating DG. The DG plays an important role in the distribution system planning due to its increasing use motivated by reduction of power loss, voltage profile improvement, meeting future load demand, and optimizing the use of non-conventional energy sources etc. The overall approach consists of two multi-objective planning stages. In the first stage, a contingency-based multiobjective planning is used to optimize the number of feeders and their routes, and the number and location of the sectionalizing switches. In the second stage, the optimum siting and sizing of the DG units is determined for the networks obtained in the first stage by another multi-objective optimization. The multiple objectives of the first planning stage are: (i) minimization of the total installation and operational cost, and (ii) maximization of network reliability. The reliability of the distribution network is evaluated by a reliability index, i.e., contingency-load-loss index (CLLI), defined as the ratio of the average non-delivered load due to failure of all branches, taken one at a time, to the total load. The objectives for the second stage optimization are the DG penetration level and the total power loss. A set of non-dominated solutions/networks is obtained by simultaneous minimization of the conflicting objectives (at each stage) using the Pareto-optimality principle based trade-off analysis. A novel multi-objective PSO (MOPSO) is proposed for solving these optimization problems using a technique for selection and assignment of leaders/guides for efficient search of the non-dominated solutions. The selection of the leaders makes use of the available non-dominated and dominated solutions. The proposed planning algorithm is tested for the static and expansion planning of typical 100-node and 21-node distribution systems, respectively. The computer simulation results are critically 292 S. Ganguly et al.evaluated. The performance of the algorithm is compared with that of the popular Strength Pareto Evolutionary Algorithm-2 (SPEA2)-based PSO and few other existing MOPSO techniques by means of statistical tests to highlight the efficacy of the proposed scheme.

Smart Distribution Networks: A Review of Modern Distribution Concepts from a Planning Perspective

“…The noticeable numerical techniques addressed in the literature (Table 3) include; ε-constraint method [121,135,161]; second order cone programming [124]; lexicographic method [135]; MCS [122,139,157]; OPF [158]; integer programming [85]; graph theory [85,150]; penalty factor (funchion) method [130,140]; CCP [145,157]; compromise method [153]; ICSP [160]; linear programming (LP) [164] and non-linear programming (NLP) [152,160]. Moreover, goal programming (GOP), exhaustive search, sequential quadratic programming (SQP) and dynamic programming methods have sucessfully employed in conventional DN related planning problems [26].…”

“…The prominent meta-heuristic techniques addressed in the literature (Table 3) include: genetic algorithm (GA) and associated evolutionary algorithms with various variants as in [122,128,131,137,139,143,147,152,155,157,162,165,166]. Also, PSO with various variants have addressed SDN planning problems as in [125,127,130,132,138,158,163]. Furthermore, an improved variant of the teaching learning algorithm (ITLBO) [133] and bacterial foraging algorithm (MBFO) [134] have been employed for short term MO planning (scheduling) problems, respectively.…”

Multi-Objective Planning Techniques in Distribution Networks: A Composite Review

“…• Benefits: Easy to code, efficient computation time and better convergence than GA. [61,65], [67]]; CRU (C) [76,88]; NTR (D) [92,93].…”

“…Planning components associated to each PT in reviewed work are designated with symbols A, B, C, and D, respectively, also shown in an overarching diagram as in Figure 1. The color coding allocated to each PT in the arrangement order includes green (A) for DGP ; orange (B) for VPQ [53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69][70]; blue (C) for CRU [71][72][73][74][75][76][77][78][79][80][81][82][83][84][85][86][87][88][89] and purple (D) for NTR [90][91][92][93][94][95][96][97][98][99][100]. It is important to note that date o...…”

A risk-based simulation and multi-objective optimization framework for the integration of distributed renewable generation and storage

“…[1-3, 7, 8, 11, 13, 14, 16-27]. The second type of operational objective functions mainly revolves around indexes such as the contingency load loss index (CLLI) [23], expected value of non-distributed energy cost (ECOST) , system average interruption duration index (SAIDI), system average interruption frequency index (SAIFI) [7,16,28], expected energy not supplied (EENS) [28,29], among others. Regarding the third type of objective functions, technical performance indicators include energy losses [1,30] and total voltage deviation (TVD) [18].…”