Comprehensive Evaluation Of Grid Connected Operation Performance Of Wind Farm Based On GWO-LSTM

With the development of wind power technology,people have put forward higher requirements for the quality of power supply,and it is particularly important to scientifically and reasonably evaluate the performance of wind farms connected to the grid.At present,the comprehensive evaluation of wind farms connected to the grid is often limited to a certain aspect of the indicators,resulting in one-sided and inaccurate evaluation results,and the weight calculation of the evaluation method is relatively cumbersome,and the time sequence of the indicators is not considered.In response to the above problems,this article constructs an indicator system from multiple aspects and uses intelligent algorithms to optimize the evaluation method.The specific content is as follows:1.To sort out the theories of the current wind farm operation performance evaluation methods at home and abroad,and on the basis of in-depth analysis and discussion,sum up the shortcomings of the current power grid operation performance evaluation,draw on the experience of experts from relevant power departments and refer to domestic and foreign literature,Established a wind power grid-connected operation performance index system considering safety,reliability and economy,and then elaborated the definition and calculation method of each index,and eliminated the more relevant indexes by using the correlation coefficient method.2.Analyze and compare multiple comprehensive evaluation methods,construct a comprehensive quantitative evaluation method based on the improved TOPSIS combined weight method,use the modified partial derivative to improve the analytic hierarchy process to obtain the subjective weight,and use the multiplier composite normalization method to convert the subjective weight Combined with the objective weights obtained by the CRITIC-entropy weight method to obtain the combined weights,the TOPSIS method is improved,the Euclidean distance between the ideal solutions is weighted,and the improved TOPSIS method is used for comprehensive quantitative evaluation.An example of the development of a power supply station in a suburban area in Jiangsu verifies the effectiveness and feasibility of the above method.3.Analyze the comprehensive evaluation method of wind farm grid-connected operation performance,consider the time sequence of evaluation,and use the gray wolf optimization algorithm to optimize the hyperparameters of the long and short-term memory neural network,including the number of hidden layer neurons,the number of iterations,Learning rate,establish a comprehensive evaluation model of GWO-LSTM neural network.The accuracy of the LSTM neural network model is 0.68,and the accuracy of the GWO-LSTM neural network model is 0.96.The optimized LSTM network has higher recognition accuracy.Finally,the GWO-LSTM neural network model is applied to a power supply station in a suburb of Jiangsu Evaluation analysis.