矿业科学学报2025，Vol.10Issue(5)：912-922,11.DOI:10.19606/j.cnki.jmst.2025104
基于改进DBO优化CNN的煤与瓦斯突出风险预测
Coal and gas outburst risk prediction based on improved DBO optimized CNN
摘要
Abstract
The gradual increase in coal-mine excavation depth leads to the significant rise in the in situ stress in deep surrounding rock and escalating risks of gas desorption and accumulation,causing a high-er likelihood of coal-gas outbursts.In this light,the present study develops a deep-learning-based pre-dictive model for coal-gas outbursts.First,the collected data were preprocessed using the Local Outlier Factor(LOF)and Multiple Imputation by Chained Equations(MICE),and employed Kendall's rank correlation coefficient to select those factors exhibiting strong correlation as the predictive indicators for gas outbursts.Next,a convolutional neural network(CNN)architecture was constructed,and opti-mized its hyperparameters via an enhanced dung beetle optimization algorithm(MSADBO).This algo-rithm incorporates an improved sine-based dynamic search-step adjustment,an adaptive Gaussian-Cauchy hybrid mutation to bolster global and local search capabilities,and a Bernoulli chaotic-map strategy to increase population diversity.Finally,comparative models were established;accuracy and other evaluation metrics were compared across models,and the safety of the predictions was analyzed via confusion matrices.Results demonstrate that the MSADBO-CNN model achieved an accuracy of 98.7％on the training set and 91.67％on both the validation and test sets,thereby attaining the highest predictive precision while also exhibiting superior robustness,generalization ability,and opera-tional safety.关键词
煤与瓦斯突出/数据预处理/预测指标/参数优选/卷积神经网络Key words
coal and gas outburst/data preprocessing/predictive indicators/parameter optimization/convolutional neural network分类
矿山工程引用本文复制引用
杜锋,李康楠,王凯,戴林超,赵明昊,王超杰,蒋立翔,王亮..基于改进DBO优化CNN的煤与瓦斯突出风险预测[J].矿业科学学报,2025,10(5):912-922,11.基金项目
国家自然科学基金(52374249,52130409) （52374249,52130409）
中央高校基本科研业务费(BBJ2024019) （BBJ2024019）
重庆市自科基金面上项目(CSTB2022NSCQ-MSX1080) （CSTB2022NSCQ-MSX1080）
河南省自然科学基金(232300420331) （232300420331）
