基于PSO-BP神经网络算法矿井瓦斯涌出量回归预测应用  

Application of Mine Gas Emission Regression Prediction Based on PSO-BP Neural Network Algorithm

作　　者：刘大可 张浩强 郭翔 LIU Dake;ZHANG Haoqiang;GUO Xiang

机构地区：[1]中色国矿帕鲁特有限责任公司,北京100029

出　　处：《中国矿山工程》2024年第3期38-43,共6页China Mine Engineering

基　　金：中国有色集团青年科学基金资助项目(2023KJZX003)。

摘　　要：本文针对矿井瓦斯涌出量预测问题,建立了PSO-BP神经网络算法模型,收集了山西某煤矿2017年至2023年期间的20组样本数据,将其中的15组作为训练集,对剩余5组的样本数据进行瓦斯涌出量回归预测,并最终对比了PSO-BP神经网络算法与BP神经网络算法的平均绝对误差、均方误差、均方根误差、平均绝对百分比误差和预测准确率等评价指标。结果表明,基于PSO-BP神经网络算法的瓦斯涌出量预测模型具有更高的准确性,能够满足矿山实际需求,具有较好的实用性和创新性,为其他矿井在瓦斯涌出量预测方面提供了一定的借鉴意义。With the development of Deep Learning in the field of Artificial Intelligence,BP neural network algorithms are also widely used in research work in various industries,and gas emission prediction is a multi-dimensional,nonlinear,small-sample prediction problem.Aiming at the problem of mine gas emission prediction,this paper establishes a PSO-BP neural network algorithm model,collects 20 sets of sample data from some coal mine in Shanxi Province from 2017 to 2023,and uses 15 sets of them as training sets.Five sets of sample data were used for regression prediction of gas emission,and finally compared the mean absolute error,mean square error,root mean square error,mean absolute percentage error,and prediction accuracy of the PSO-BP neural network algorithm and the BP neural network algorithm.The results show that the gas emission prediction model based on the PSO-BP neural network algorithm has higher accuracy,can meet the actual needs of mine safety production,has better practicability and innovation,and provides a new model for other mines in gas emission.It provides some reference value for other mines in the direction of gas emission prediction.

关 键 词：瓦斯涌出量预测 粒子群优化算法 反向传播神经网络 回归预测 评价指标 

分 类 号：TD722[矿业工程—矿井通风与安全]