Abstract
Background: The increasing concern of global climate change, the promotion of renewable energy sources, primarily wind generation, reduces the power generation from conventional plants that lead to the reduction in pollutant emission. The exploitation of wind power generation is rising throughout the world. The objective of Unit Commitment (UC) is to identify the optimal generation scheme of the committed units such that the overall generation cost is reduced, when subjected to a variety of constraints at each time interval. The optimum generation planning in electrical power system is difficult, since UC Problem has many variables and system and unit constraints of thermal generating units. Nowadays, it is essential to include reliability analysis of the power system in operation strategy of the generating units. Here, the generator failure and malfunction are considered in UC problem formulation.
Methods: This paper presents a meta-heuristic algorithm based approach to determine the thermal generation schedule with consideration of wind energy system. A novel evolutionary algorithm known as Grey Wolf Optimization (GWO) algorithm is applied to solve the UC problem.
Results: The potential of the GWO algorithm is validated by the standard test systems. Besides, the ramp rate limits are also incorporated in the mathematical problem formulation. In order to validate the applicability of the GWO, the standard test system is used for demonstration.
Conclusion: The GWO algorithm is applied for the first time to solve wind integrated thermal UC problem considering generator forced outage rates. The simulation results reveal that the GWO algorithm has the capability of obtaining economical resolutions with good solution quality. The implementation of algorithm for solving the chosen problem is simple and robust which indicates that the GWO is a promising alternative for solving wind integrated thermal UC problems.
Keywords: Grey wolf optimization, ramp rate limit, reliability analysis, unit commitment, wind power generation, UC problem.
Graphical Abstract