Generation maintenance scheduling using improved binary particle swarm optimisation considering aging failures

This study presents a coordinated deterministic and stochastic framework for maintenance scheduling (MS) of generators, using improved binary particle swarm optimisation (IBPSO). Fundamental concern of MS is to reduce the generator failures and to extend the generator lifespan, thereby increases the system reliability. The IBPSO finds an optimal schedule for the generators and overcome the drawbacks of the conventional methods. The objective of this study is to reduce the loss of load probability and minimising the annual supply reserve ratio deviation for a power system, which are considered as a measure of power system reliability. Moreover, in this study the impacts of aging failures of the generators are considered in order to calculate the unavailability of power system which is modelled using the Weibull distribution. The proposed algorithm is tested on IEEE reliability test system. Comprehensive study has also been carried out in the context of Kerala (India) power system. It can accomplish a significant levelisation in the reliability indices over the maintenance planning period and demonstrates the potential to solve the MS problem. The numerical results are obtained using the proposed method, which outperforms other compared techniques such as genetic algorithm, particle swarm optimisation, binary particle swarm optimisation methods and conventional methods.