IMPLEMENTATION OF GENERAL-PURPOSE BENDERS DECOMPOSITION ALGORITHM IN MATLAB AND ITS APPLICATION TO POWER SYSTEM PLANNING

For most huge systems, planning is a very important step to adapt them for future excepted changes. Planning could figure out most of potential problems and help people to make long-term decisions to get more secure and profitable. But huge system-planning problems would cost long time and lots of effort, we need some mathematic methods to simplify and solve the problems. Luckily, there comes Benders Decomposition. Benders Decomposition is a popular technique in solving certain classes of difficult problems such as stochastic programming problems, and mixed-integer nonlinear programming problems [1][2]. Then some software is needed to model basic template of Benders Decomposition algorithm with and for most engineers, MATLAB is one of the most popular software, which could solve engineering problems by matrixes. So the resultant model is solved numerically by the application of Benders Decomposition algorithm, whose implementation and development were executed using the software MATLAB. Even though the codes are for general system-planning problems, most of examples in this paper were power systems. But it does not mean it could only solve specifically, examples of power systems are popular and typical in Bender Decomposition problem solving. This paper describes the process of programming, such as basic mathematic method studying, specific models making, common models building, codes simplifying and examples testing with. At the end of it, conclusion was made about the performing of the code.M.S. in Electrical Engineering, December 201