Hydrothermal scheduling flexibility enhancement with pumped-storage units

This paper presents an efficient MILP algorithm to solve hydrothermal scheduling problem with enhanced flexibility via multiple pumped-storage units. A comprehensive Hydraulic model for pumped-storage and river system constraints is developed. The performance improvement of the proposed algorithm falls in two-fold. First, a piecewise approximation approach is applied throughout all nonlinear functions of the model, representing fully linear HTS model. Second, multi-threadng technology which accelerates the solution process of Branch and Cut method. Simulation results show that the developed linear MIP-HTS algorithm exhibits high performance in terms of objective cost and execution time. Further results demonstrated that implementation of multi-thread parallel programming can further improve the computation time as compared to the serial executions. This, however, allowed one to procure better solution by taking tighter optimality gap that results in more accurate solution.