Stochastic optimazation models for planning and operation of a multipurpose water reservoir

We consider the capacity determination problem of a hydro reservoir. The reservoir is to be used primarily for hydropower generation; however, commitments on release targets for irrigation as well as mitigation of downstream flood hazards are also sec-ondary objectives. This thesis is concerned with studying the complex interaction among various system reliabilities (power, flood, irrigation, etc.) and to provide de-cision makers a planning tool for further investigation. The main tools are stochastic programming models that recognize the randomness in the streamflow. A chance con-strained programming model and a stochastic programming model with recourse are formulated and solved. The models developed incorporate a special target-priority policy according to given system reliabilities. Optimized values are then used in a simulation model to investigate the system behavior. Detailed computational results are provided and analyzed.