IBA-LSSVM Wind Power Prediction Based On Variational Modal Decomposition

In recent years,in order to alleviate energy shortage and reduce environmental pollution,China’s new energy has developed rapidly,the application scale has been expanding,the cost has been continuously reduced,the consumption contradiction has been significantly alleviated,and the role of clean substitution has become increasingly significant.Among them,wind energy is a clean and environment-friendly energy with great utilization value,which is widely used in wind power generation.The future power system is bound to provide higher penetration clean energy for sustainable global economic growth.However,the continuous access of a large number of clean energy poses unprecedented challenges to the power system.Therefore,short-term prediction of wind farm power can enable the power dispatching department to timely adjust the dispatching plan according to the change of wind power in advance,ensure power quality and reduce the operation cost of power system,which is an effective way to reduce the adverse impact of wind power on the power grid and improve the installed proportion of wind power in the power grid.Aiming at the randomness of wind power generation,and in order to reduce the rejection rate and improve the accuracy of wind power prediction,a short-term wind power prediction method based on variational modal decomposition(VMD)and improved least squares support vector machine(LSSVM)is proposed in this paper.Firstly,the original wind power sequence with nonlinearity and randomness is decomposed into a series of stable modal components by VMD.According to the analysis results of the approximate entropy of each sub mode,the sub sequence is reorganized and divided into trend component,detail component and random component,so as to reduce the complexity and instability of the original data;Then the improved inertia weight,adaptive frequency and mutation mechanism are used to improve the bat algorithm,and the parameters of the least squares vector machine are optimized to find the optimal parameters.The IBA-LSSVM wind power prediction model is established to predict the three reconstructed sub modes respectively,and the prediction results of the sub modes are superimposed to obtain the final predicted value of power generation.Finally,the proposed prediction method is used to predict the power of a wind power plant in Ningxia.The error analysis of the prediction results proves the effectiveness of the model.Through the comparison of different prediction methods,it is verified that the proposed model has higher prediction accuracy.Therefore,the application of this prediction method to the actual wind power prediction has academic significance and engineering value.There are27 figures,6 tables and 68 references.