基于EMD-DBO-BiLSTM的风电外送线路载流量预测方法OA

The ampacity of overhead transmission lines for wind power delivery is intricately linked with surrounding meteorological factors.Precisely forecasting this ampacity holds paramount significance in augmenting the transmission capability of wind farms.However,there are two major problems with existing load capacity prediction methods:insufficient extraction of data features of meteorological elements on transmission lines,and poor robustness of a single prediction model.In response,the empirical mode decomposition(EMD)algorithm is introduced to dissect the time series of meteorological elements into various components with differing frequencies,thereby uncovering latent correlation patterns within the data.Following this,the bidirectional long short-term memory(BiLSTM)neural network is deployed to individually predict each component,supplemented by the dung beetle optimizer(DBO)algorithm for fine-tuning BiLSTM hyperparameters,thus enhancing the stability of line carrying capacity predictions.Based on the above research,a wind power transmission line carrying capacity prediction method based on EMD-DBO-BiLSTM is proposed.Case analysis shows that compared to the four traditional methods,the average absolute error of the proposed method has decreased by 22.74%,9.30%,7.08%and 7.76%respectively.The analysis results verify the effectiveness of the method.