Research On Prediction Of Gas Emission Based On Improved Local Linear Embedding And Chaotic Bee Colony Algorithm

Coal mine gas accident is one of the most important threats in the safe mining process of coal mines in China.Therefore,it is very important to predict the gas emission from the coal mining face and to innovate the technology.Various factors affect the gas emission,and they have certain complexity,nonlinearity and feature correlation.In order to obtain better prediction results than traditional prediction methods,based on the analysis of existing gas emission predictions at home and abroad,combined with machine learning and intelligent algorithms,local linear embedding theory(LLE)and artificial bee colony(ABC)The algorithm and the nuclear extreme learning machine(KELM)are applied to the prediction of gas emission,and propose improved methods to establish new prediction models.Firstly,the improved local linear embedding algorithm is used to reduce the characteristic attributes of the gas emission factors,and the redundant information is eliminated to ensure the dimensionality reduction.The reduced and normalized samples are divided into training sets and test sets,and the KELM model is trained with the training set.Because the penalty coefficient and kernel parameters of KELM model need to be optimized in this process,this paper proposes a method of introducing local chaos search(Chaos Searching)strategy to improve ABC algorithm,and use this algorithm to optimize KELM model parameters and establish improvement based on LLE algorithm and CSABC-KELM coal mine gas emission prediction model.The prediction model was tested and analyzed using test set data,and compared with the unprocessed CSABC-KELM,KELM,ABC-ELM,ELM,and BP neural network prediction models.The results show that the improved LLE algorithm and the CSABC-KELM prediction model have higher prediction accuracy,less error,and enhanced generalization ability than other traditional prediction models.It can effectively predict the gas emission quantity and improve safety factor the coal mining process.