基于鲸鱼优化算法的神经网络在煤矿瓦斯事故风险预测中的研究

Research on neural network in risk prediction of coal mine gas accident based on whale optimization algorithm

为了解决使用传统方法难以预测煤矿瓦斯事故的问题,充分利用神经网络处理非线性问题的优越性,使用BP神经网络搭建以顶底板稳定性、风量供需比、瓦斯体积分数、煤尘爆炸指数和绝对瓦斯涌出量为输入,瓦斯事故风险为输出的神经网络。结合鲸鱼优化算法的全局搜索优势,有效避免了神经网络容易陷入局部最优问题,最后使用该模型对收集的100组数据进行训练和测试,结果验证了该模型的可行性,为煤矿行业提供了一种高效的瓦斯事故风险预测方法。

In order to solve the problem that it was difficult to predict gas accidents in coal mines by traditional methods,BP neural network was used to build a neural network with top layer stability,air supply and demand ratio,gas volume fraction,coal dust explosion index and absolute gas emission amount as inputs and gas accident risk as outputs.Then,combined with the global search advantage of the whale optimization algorithm,it effectively avoided the problem that the neural network was easy to fall into local optimum.Finally,the model was used to train and test the collected 100 sets of data.The results verified the feasibility of the model and provided an efficient gas accident risk prediction method for the coal mine industry.