GENETIC ALGORITHMS FOR MULTIRESERVOIR SYSTEM OPTIMIZATION

ABSTRACT

A Genetic Algorithm (GA) model has been developed and applied to derive optimal operational strategies for a multireservoir system in Tamil Nadu, India with the objective of minimizing the square of irrigation deficit. Release for irrigation and other demands (industrial and municipal demands), from the reservoirs are considered as the output variables. The monthly time stepped model was solved with 13 years of inflow data to derive the optimal operational strategies. The sensitivity analysis in obtaining optimal number of population, probability of crossover and number of generation showed values of 150, 0.84 and 275 respectively for the present study carried out. It is also found that the GA model has resulted in better performance when compared to the actual releases.

KEY WORDS:

• Genetic Algorithm
• Multireservoir operation
• Uniform crossover
• Modified mutation