Contribution to the long-term generation scheduling of the Cameroonian electricity production system

Abstract

This paper presents the new long-term generation scheduling model that has been developed to improve the generation scheduling of the electric system of Cameroon. Based on the Dantzig-Wolfe Decomposition Technique (DWDT), the proposed solution approach fully exploits the particular graph structure of the hydraulic system to develop a more adapted model. The decomposition procedure of the DWDT is first applied to separate the problem formulated as a stochastic linear program into two subproblems: a hydraulic subproblem, modelled as a network flow with gains problem in a graph and a thermal subproblem. Then, an algorithm of resolution in which the subproblems are solved alternately (each by an adapted method) in a finite iterative process, is developed from the coordination procedure of the DWDT. An application of the solution approach, which can also take into account the non-linear aspects of the production system, on the electricity production system of Cameroon, permits to get in better execution time, known results.