Reservoir Operation Optimization Using Simulated Annealing Algorithm and Application of Asymptote Method in its Validation

Reservoir optimum operation is one of the optimization issues in water resources engineering which has been evaluated by varios known optimization methods, yet. In this study firstly reservoir simulation was done based on standard operation policy (SOP). Then simulated annealing algorithm (SA) has been used as a numerical optimization method with inteligent random structure for optimal reservoir operation. Because of SA algorithm sensivity in input paratmeters, therefore sensivity analysis were done and appropriate combination of that parameters was selected. In order to validation of SOP-SA combined algorithm, asymptote method was applied. Results indicated capability of algorithm on global optimal finding. The SOP-SA combined algorithm was run with objective of annual deficit minimization with constraint and limitations such as reservoir storage and release volume for Doroodzan dam. After algorithm run, reservoir storage and dam release in operation period were studied. Results of Research shown that SOP-SA combined algorithm reduced deficit intensity in drought priod by distribution of deficit in different mounths. In addition, the number of months with a very low monthly supply probability percentage (less than 50 percent); 22 months in the SOP algorithm, were reduced to 10 months.