基于二次补偿策略的VMD-LSTM瓦斯浓度预测方法
VMD-LSTM Gas Concentration Prediction Method Based on Secondary Compensation Strategy
孔金浩1
作者信息
摘要
瓦斯浓度过大可能会造成窒息、爆炸等安全事故,严重威胁着煤矿企业的安全生产.为了更好地监测瓦斯浓度,及时预防瓦斯事故,提出了基于二次分解策略的VMD-LSTM瓦斯浓度预测方法.该方法首先根据PSO算法确定惩罚因子与模态数的最佳参数组合,然后利用VMD分解瓦斯浓度数据进行LSTM的初次浓度预测,最后基于二次补偿策略,对初次结果补偿预测得到更高精度结果.试验表明,该方法有效提高了瓦斯预测精度,具有较好的现实推广应用价值.
Abstract
Excessive gas concentration may lead to safety accidents such as suffocation and explosion,posing a serious threat to the safe production of coal mining enterprises.In order to better monitor gas con-centration and prevent gas accidents in time,a VMD-LSTM gas concentration prediction method based on secondary decomposition strategy is proposed.This method firstly determines the optimal combination of pen-alty factor and modal number using the PSO algorithm.Then,VMD is utilized to decompose the gas concen-tration data for the initial concentration prediction of LSTM.Finally,based on a secondary compensation strategy,the initial results are compensated to obtain higher prediction accuracy.Experiments show that this method effectively improves prediction accuracy of gas,and has good practical application value.
关键词
长短时记忆网络/浓度预测/瓦斯/变分模态分解Key words
LSTM/concentration prediction/gas/variational mode decompositionk引用本文复制引用
出版年
2024
NETL
NSTL
万方数据