Integrating Large Scale Wind Farms in Fuzzy Mid Term Unit Commitment Using PSO

This paper presents a new approach for unit commitment (UC); where a large scale wind power exists and the wind speed has a fuzzy characteristic; by using particle swarm optimization method (PSO). In this approach, the system reserve requirements, the requirement of having a load balance, and the wind power availability constraints are realized. A proper modeling of these constraints is an important issue in power system scheduling. Since these constraints are "fuzzy" in nature, any crisp treatment of them in this problem may lead to over conservative solutions. In this paper, a fuzzy optimization-based method is developed to solve power system UC problem with a fuzzy objective function and its constraints. This fuzzy mid term UC problem is, at first, converted to a crisp formulation and then is solved by PSO. This method is applied to unit commitment of a 12-unit test system and the results of the particle swarm optimization method are compared with the results of the conventional numerical methods such as mixed integer nonlinear programming (MINLP). Numerical tests results show that near optimal schedules are obtained, by application of this method. Also this method provides a balance between the costs and the constraints satisfaction.