Combined Prediction of Gas Emergence Based on Modalihj Decomposition and Temporal Convolutional Networks
In order to effectively analyze and process coal mine gas emission data,which is crucial for maintaining safety in mining opera-tions,complete ensemble empirical mode decomposition with adaptive noise is proposed for decomposing the gas emission volume sequence.Once each component,obtained through decomposition,is analyzed,a temporal convolutional networks model is constructed,the purpose of which is to establish prediction models for each component.The Logistic chaos mapping,which is employed to generate a golden jackal population,plays an essential role in the optimization process.When the Cauchy-Gaussian mutation operator is introduced to update the golden jackal positions and select the optimal location,it enhances the search capability of the algorithm and avoids local optima that could hinder accurate predictions.Consequently,the prediction output values of each component are superimposed to obtain the final gas emission volume prediction.Test results indicate that the CEEMDAN-IGJO-TCN combined forecasting method not only reduces the com-plexity of the prediction but also improves its accuracy,making it a valuable tool in monitoring gas emissions in coal mines.
万方数据
维普
