Pump scheduling optimization for electrical energy consumption and reliability level of water supply system

Abstract

Pump scheduling and optimization are very essential and important to the operation management of the water supply system in order to reduce high operational cost. Scope of this study is limited to only the water supply part and distribution system. This study will consider fixed pumps and use approximated model in order to reduce its complexity. The inputs for the optimization process include the average of data demand profile, the pump technical characteristic, the electric tariff and reservoir parameters. Adaptive Weight Genetic Algorithm (AWGA) is used as an optimization search engine. The process started with initialization, fitness evaluation, selection, crossover and mutation. Elitism is also included during the process to ensure that some of the best individuals are retained in each generation. Six tests have been conducted and the optimal result that produces the least electric energy cost with percentage difference index (PDI) is 37.97%. The tolerance for six tests is about ± 3%