瓦斯突出因素与风险预测方法对比
Comparison of Gas Outburst Factors and Risk Prediction Methods

瓦斯安全是煤矿安全生产的核心议题，直接关系到矿工生命安全和煤炭工业的可持续发展。随着开采深度增加，高瓦斯矿井比例上升，瓦斯压力增大，灾害风险加剧，瓦斯爆炸、煤与瓦斯突出等事故频发，成为制约煤炭行业发展的关键瓶颈。进行瓦斯突出风险预测有很大必要。本文主要讨论有关预防瓦斯突出方法：反向传播神经网络，改进自适应动态甲虫优化–反向传播神经网络，支持向量机，随机森林，卷积神经网络等，比对它们的优势和局限性，为瓦斯突出风险进行防范。
Gas safety constitutes the cornerstone of safe coal production, directly affecting the lives of miners and the sustainable development of the coal industry. With increasing mining depth, the proportion of high-gas mines continues to rise, leading to elevated gas pressures and intensified disaster risks. Consequently, gas explosions and coal-gas outbursts occur frequently, constituting a critical bottleneck that constrains the development of the coal sector. Therefore, accurate prediction of coal-gas outburst risk is of great necessity. This paper systematically compares state-of-the-art prediction methods, including Back-Propagation neural networks (BP), Modified Self-Adaptive Dynamic Beetle Optimization-Back-Propagation neural networks (MSADBO-BP), Support Vector Machines (SVM), Random Forest (RF), and Convolutional Neural Networks (CNN). The advantages and limitations of these approaches are analyzed to provide a theoretical basis for the prevention and control of coal-gas outburst risks.