Generation Scheduling Approach for Hydrothermal Power Systems = 硕士学位论文 水火电机组联合发电优化调度方法

Abstract

In power systems operation, a system consisting of both hydro and thermal power plants is more complex than the scheduling operation of an all thermal generation due to a number of constraints of water available in a given period for hydro generation. The hydro power plants having a negligible operating cost compared to the thermal power plants; the objective of generation scheduling approach of hydrothermal power plant is to minimize the thermal operating cost under the constraints of water available to operate the hydro plant during a given period of operation. A variety of optimization techniques; conventional and artificial intelligence techniques has been applied to solve the hydrothermal scheduling problems, however these techniques consist of a certain drawbacks like drastic growth of computational and dimensional requirement, complex algorithm and convergence characteristics. The work done in this thesis named “Generation Scheduling Approach of Hydrothermal Power Systems” (GSAHTPS) develops a solution of short term hydrothermal scheduling problems; the solution approach is based on mixed integer linear programming (MILP). The IBM ILOG Cplex Optimization Studio is used to build linear programming codes; the Optimization Programming Language (OPL) in Cplex taken as modeling language for combinatorial optimization that may simplify the optimization problems substantially is used, this OPL provides support for modeling linear and integer programs and also provide access to state-of-the-art linear programming algorithm. The generating power characteristics of both hydro and thermal power plants being non-linear, the piecewise linear approximations are used to solve this scheduling problems using IBM ILOG Cplex Optimization Studio. Three methods for piecewise linear approximation that are one dimension, triangle and rectangle method are developed in this thesis due to the non-linear output power characteristics of hydro and thermal power plants. In this research we mainly focus on the economic dispatch and unit commitment problems. To assess the efficiency and the powerful performance of the proposed method; a typical case study of forty six thermal power plants from the standard of IEEE118 bus system is investigated with four additional hydropower plants connected in cascade. The numerical results show that the solution techniques are computationally efficiency, simple and suitable for decision support of short term hydrothermal scheduling problems.