Reduction of frequency fluctuation for wind farm connected power systems by an adaptive artificial neural network controlled energy capacitor system

Abstract

Frequency fluctuations are a major concern for transmission system operators and power grid companies from the beginning of power system operation due to their adverse effects on modern computer-controlled industrial systems. Because of the huge integration of wind power into the power grid, frequency fluctuations are becoming a serious problem, where randomly varying wind power causes the grid frequency fluctuations of the power system. Therefore, in this paper, the minimisation of the frequency fluctuation of a power system, including a wind farm, is proposed using an energy capacitor system (ECS). A scaled-down, multi-machine power system model from Hokkaido prefecture, Japan, is considered for the analysis. A novel adaptive artificial neural network (ANN) controller is considered for controlling the DC-bus connected ECS. The control objective is to standardise the line power of the wind farm, taking into consideration the frequency deviation. The effects of wind power penetration levels, as well as load variations, are also analysed. The proposed control method is verified by simulation analysis, which is performed with PSCAD/EMTDC using real wind speed data. The adaptive ANN-controlled ECS was found to be an effective means of diminishing the frequency fluctuation of multi-machine power systems with connected wind farms.