A practical mixed integer linear programming based approach for unit commitment

摘要

This paper presents a practical mixed integer linear programming (MILP) based approach for unit commitment (UC), which is suitable for both traditional and deregulated environments. In general, the UC problem is complicated large-scale, combinatorial, and non-convex in nature, it is very difficult to solve by conventional approaches to achieve both solution accuracy and efficiency. With the recent development of solution techniques, there is a trend to tackle the UC problem by using MILP approaches. In this paper the authors propose a detailed procedure to formulate the UC problem in MILP manners. The problems are then solved via a state-of-the-art optimization package. The usefulness of the proposed solution technique is illustrated by testing the problem with actual system data. The solution obtained not only gives the unit on/off states and MW schedules, but also provides marginal price information associated with system constraints such as load demand requirement to assist strategic bidding in the power market.