A decoupled solution of hydro-thermal optimal power flow problem by means of interior point method and network programming

This paper presents a new decoupled model together with a very efficient coordination algorithm to solve a hydrothermal optimal power flow (HTOPF) problem over a certain time horizon. Based on the Lagrange relaxation at the level of the KKT (Karush-Kuhn-Tucker) conditions of the primal problem, the HTOPF is decomposed into thermal plant subproblems formulated as OPF and hydroplant subproblems. To solve efficiently the thermal OPF subproblems, the warm-starting scheme has been incorporated into interior point quadratic programming (IPQP). As to the hydroplant subproblems, a united network flow model is presented in which a fixed head plant is treated as a special case of a variable head plant. The hydroplant subproblem can be formulated as a minimum-cost maximum-flow problem for which unit cost functions of hydroplants are defined exactly. A proposed variant of the partitioning shortest path algorithm has brought about a great speed up in the computation of the subproblems. The validity of the proposed method has been examined by solving the IEEE test systems and a Chinese power system consisting of 13 thermal plants and 12 hydro power plants; the last system is a large size problem such that it has 107712 primal and dual variables. Simulation results obtained are quite convincing.