CONDENSER MAINTENANCE COST OPTIMISATION USING GENETIC ALGORITHMS

摘要

Power plant optimisation is one of the most dynamic issues in process and system engineering. This paper discusses the exploitation of genetic algorithm optimisation for improving condenser maintenance scheduling in a thermal power plant. In collaboration with a local utility, an innovative software application has been developed to optimise the cost of scheduling regular condenser maintenance. Typically maintenance is based on a simplistic linear costing formula. However, by utilising the information available from the resident distributed control system (DCS) a complex cost function was developed which more accurately embraces the performance related, contractual and direct maintenance issues. A binary genetic algorithm optimisation scheme was then applied to determine the optimum number and scheduling pattern of maintenance activity.