A stochastic MILP model for long-term hydrothermal scheduling considering water resource management

A multistage scenario tree based stochastic model is proposed for long-term hydrothermal scheduling (LHTS) in this paper to hedge against the uncertainties of natural inflows, water demand, grid load and wind power generation. With scenarios reduction, a 3-stage, 81-scenario stochastic tree is established based on stochastic weather condition and net load. In addition, detailed formulations of hydrothermal system and water resource management such as water supply/recession procedure, distributed water usage allocation policy and etc. are also included in the basic nonlinear model. Then the nonlinear functions in the formulation such as thermal generating costs function, hydro power production function, water recession function and reservoir evaporation function are replaced by their piecewise linear equivalents and the stochastic mixed integer linear programming (MILP) formulation is solved by commercial solver CPLEX. The numerical results show that the proposed stochastic MILP model for LHTS is efficient.