Reliability Evaluation Method Of Distribution System With Wind Turbine Generator

Wind energy is an alternative to fossil fuels.It is plentiful,clean and produces no air pollution emissions,and wind power technology has been vigorously developed in recent years.In order to alleviate energy crisis and improve power supply's reliability,an increasing number of wind turbine generators(WTG)are accessing the distribution system.But,compared with the traditional distribution system,the radial power supply structure of the distribution system is changed when WTG access to distribution system.Besides,the reliability of the distribution system becomes a little unstable because of the non-linear,non-stationary characteristics of wind power generation.Therefore,it is of great theoretical and practical meaning to consider the characteristics of WTG output and study a suitable reliability evaluation method for distribution systems with WTG.A wind power forecasting method based on wavelet transform and convolution neural network is proposed.Firstly,the original wind power time series is transformed into several stationary time series with different frequencies by wavelet transform,considering the non-linear and non-stationary effects in the WTG output.Then,the deep convolution neural network model is forecasted for each time series,thereby improving the accuracy of the forecast.The paper adopts the WT-CNN forecasting method tolastly establish a model of the output of WTG.The model is obtained by predicting the output power of WTG for one year,and it is applied to the reliability evaluation of distribution systems with WTG,so that the evaluation results are more in line with the actual situation.A reliability evaluation method of WTG power distribution system based on Monte Carlo thought is proposed.According to the reliability parameters of the distribution system components,the operation of the distribution system is simulated by generating random number.When the component failure occurs in the distribution system,the operation state of the distribution system is analyzed by the node failure analysis method.When the distribution system appears island operation state,achieve division of the island based on a reasonable strategy.Finally,the reliability index of the distribution system is calculated by statistical method.The methods proposed in this paper are simulated and analyzed based on the MATLAB pratform.Firstly,the reliability evaluation of distribution system with different distributed generation is simulated and the reliability index is analyzed.Secondly,under the different power prediction algorithms,the reliability evaluation of distribution system with WTG is simulated and the reliability index is analyzed.The results show that the WTG is connected to the distribution system can enhance the reliability index,thereby improving the reliability of the distribution system,but the improvement effect is not as good as that of the distribution system with conventional two-state distributed power.Besides,in this paper,the proposed wind power forecasting method based on WT-CNN has excellent forecasting performance,and it is better than SVM and BP algorithm.Lastly,under the different power prediction method,the proposed reliability evaluation method of distribution system with WTG can obtain more realistic reliability index,and indicating feasibility and effectiveness of the proposed method.