Application of Natural Computational Algorithms in Optimal Enhancement of Reliability Parameters for Electrical Distribution System

Reliability improvement of distribution system components is gaining more attention due to the structural changes of a power system which provides multiple choices to the consumers for selecting the utility. The optimal enhancement strategy is based on the trade-off between the investment required for improving the reliability parameters of the system components from the present level and the reduction in interruption power which is in terms of cost. These costs are modeled in terms average failure rate and average interruption duration of system sections. Adequacy of the power supply at loads points is ensured by imposing the constraints on upper bounds on load points and sections reliability parameters of the system. Consumers concern on supply reliability is fulfilled by imposing bounds on customer oriented reliability indices. This makes the optimal design problem as optimization problem of non-linear objective with linear and non-linear constraints. This paper applies the population based natural computational algorithms such as genetic algorithm, particle swarm optimization, differential evolution and firefly algorithm for solving the optimal reliability enhancement model of the sample test system. Results obtained are compared with the results of existing literature which uses polynomial time algorithm.