Международный журнал роевого интеллекта и эволюционных вычислений

Абстрактный

Реальное снижение потерь мощности за счет улучшенного роя светлячков, когнитивного развития, черной дыры и усовершенствованных алгоритмов летучих мышей

Канагасабай Ленин

This paper proposes an Improved Glow worm Swarm Optimization (IGSO) algorithm, Cognitive Development Optimization (CDO) algorithm, Black Hole Algorithm (BHA) and Bat Algorithm with Combination of Numerous Schemes (BACS) to solve the optimal reactive power problem. Glow worm swarm optimization (GSO) algorithm is a new algorithm which stimulated from the light emission behavior of glow worms to attract prey. GSO algorithm has limitation in global search, short fall in accuracy computation and often falls into local optimum. In order to prevail over the above said shortcomings of GSO, this work presents improved GSO algorithm, to solve the problem. Glow worm Swarm Optimization Algorithm incorporated with the parallel hybrid mutation which unites the uniform distribution mutation with the Gaussian distribution mutation. In proposed (IGSO) algorithm dynamic moving step length is implemented to each individual. When the position unchanged in any generation then Normal distribution variation to the glow worm is applied. Then in this paper Cognitive Development Optimization (CDO) algorithm utilized for solving reactive power problem. Piaget’s Theory on Cognitive Development, which has; maturation, social interaction, balancing; all these three processes are utilized throughout the new learning phase and improving constantly the cognitive infrastructure. Then this work presents Black Hole Algorithm (BHA) for solving optimal reactive power problem. Evolution of the population is through push the candidates in the itinerary of the most exceptional candidate in iterations and black hole which exchange with those in the exploration space. Tremendous candidate amongst all the candidates in iterations is selected as a black hole and left over candidates structured as the standard stars. Black hole formation is not capricious but it is form as genuine candidates of the created population. To improve the exploration and exploitation stars gravity information has been utilized. Gravitational forces between the stars are definite and progression of stars towards the black hole is accustomed during the incursion in solution space. Then in this paper Bat Algorithm with Combination of Numerous Schemes (BACS) is proposed to solve optimal reactive power problem. Bat algorithm is mimicked from the actions of the Bat; mainly time delays are used for emission to reflection and employ it for navigation. The global convergence capability of the algorithm becomes weaker when the progress of operator increases and when the exploration operator augments, then the convergence accurateness will be inadequate. Consequently, in this paper, numerous schemes have been selected to solve the problem and it work as autonomous selection strategy. In the proposed algorithm different individuals prefer different strategy to modernize the position with reference to the quality of fitness. Proposed Improved Glow worm Swarm Optimization (IGSO) algorithm, Cognitive Development Optimization (CDO) algorithm, Black Hole Algorithm (BHA) and Bat Algorithm with Combination of Numerous Schemes (BACS) has been tested in standard IEEE 14, 30,300 bus test system and simulation results show the projected algorithm reduced the real power loss considerably.