Hybrid heuristic optimization approach for optimal Distributed Generation placement and sizing

Abstract

This paper presents a hybrid algorithm that combines Particle Swarm Optimization (PSO) and Nonlinear Optimal Power Flow (OPF) in the optimal sitting and sizing of Distributed Generation (DG). The objective function considered is to minimize the power losses in distribution systems. The proposed approach makes use of a sensitivity index based on derivatives to identify the best candidate buses for sitting the DG. This index is considered in the PSO initial population aiming at reducing the search space, though, improving the convergence of the method. The OPF is then used to determine the optimal size of each DG. Results are compared with previous papers presenting different methodologies for the 69-bus radial distribution system to validate the proposed approach.

Power electronics in smart grid distribution power systems: a review

“…The components that make up a typical HEV include a battery pack, motor controller, motor/generator, internal combustion engine, and transmission and driveline components. The primary PEs in an HEV include a DC-AC motor controller which provides three-phase power to a PM motor Srivastava et al, 2010;Bayoumi, 2005;Bayoumi and Soliman, 2007;Soliman et al, 2009a;Saber and Venayagamoorthy, 2010;Zhang et al, 2011b;Dias et al, 2012;Bayoumi et al, 2011;Bayoumi, 2010a).…”

Aplicação De Algoritmos Genéticos Para O Planejamento De Geração Distribuída Em Sistemas De Distribuição

“…Adicionalmente, outros aspectos têm contribuído para este aumento, como a desregulamentação dos sistemas de potência, avanços tecnológicos associados ao desenvolvimento de pequenas unidades de geração eficientes e a necessidade de redução de perdas técnicas (DIAS et al, 2012) As vantagens da GD dependem de sua correta localização no SDE distribuição (DIAS et al, 2012), (KAYAL;CHANDA, 2013). A escolha de pontos inadequados para a conexão de GD pode reduzir suas potenciais vantagens, anulá-las ou até mesmo afetar negativamente a operação do sistema.…”

“…Quando o planejamento de geração distribuída considera a possibilidade de investimento em múltiplas fontes, a alocação ótima torna-se um problema de otimização inteira mista, não linear e não convexo (DIAS et al, 2012 De acordo com a referência (KAYAL; CHANDA, 2013), a escolha dos objetivos relacionados com o planejamento de GD impacta na localização dos recursos e as premissas relativas a este assunto ainda são alvo de estudo e investigação.…”

Optimal Placement of Distributed Generation Units in a Distribution System with Uncertain Topologies using Monte Carlo Simulation

“…The methodologies proposed in the literature can be stratified by objective function, algorithm applied and additional control variables utilized. In relation to algorithms, the most common applied are: Exhaustive Search [3], Genetic [4][5][6][7][8][9][10], Particle Swarm Optimization [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26], Mixed-Integer Linear Programming [27,28], Artificial Neural Networks [29], Kalman Filter [30], Evolutionary [31], Non-dominated Sorting Genetic Algorithm II [32], Tabu Search [33], Multi-Objective Nonlinear Programming [34], Chaotic Artificial Bee Colony [35] and Fuzzy Approach [36]. There are also new algorithms proposed by the authors: Chaotic Local Search and Modified Honey Bee Mating Optimization [37], Modified Discrete Particle Swarm Optimization [38], Modified Teaching-Learning Based Optimization [39], Plant Growth Simulation [40], Imperialist Competitive Algorithm [41] and Improved Multi-Objective Harmony Search [42].…”