Unit commitment problem of thermal generation units for short term operational planning using simple genetic algorithm

Unit Commitment problem plays a major role in power system since the improved UC schedules may save the electric utilities millions of dollars per year in production cost. The objective of the optimal commitment is to determine the on/off states of the units in the system to meet the load requirement and spinning reserve requirement at each time period such that the overall cost of generation is minimum, while satisfying various constraints. Several research have been done in this field for the past three decades. With the development of modern power system, it is not practical to use the classical approaches to solve large scale UC problem. Due to the limitations of the mathematical programming methods. Al based techniques are used GA is an adaptive search method for optimal or near optimal commitment order. GA can get a sub-optimal solution that is very close to the global optimal solution and can meet the demand of engineering application. A GA implementation using the standard reproduction. cross over and mutation has been used to get optimal solution. The approach of UC using GA consists of repeating the process of economic dispatch and minimizing the objective function for various unit combinations and over a population of feasible solution. UC solution based on GA has been programmed/ implemented in C++. The performance of GA approach is initially tested as a case study for 3-generator system and for a load pattern of 24 hours. This paper applies the GA to the UC scheduling problem and illustrates details of the performance of genetic algorithm. The aim of this paper is to propose the suitability of a new approach to the solution of the UC problem. (author)