论文部分内容阅读
煤与瓦斯突出是主要矿难形式之一,危害极大。依据现场监测瓦斯突出的相关数据,对瓦斯突出的危险程度进行预测,提前做好防范措施,可大大降低事故危害。文中提出支持向量机算法与改进粒子群算法相结合的瓦斯突出危险程度预测模型:通过对容易陷入局部最优的粒子群进行改进,并应用改进粒子群算法求解影响支持向量机分类预测性能的最佳参数,然后把最佳参数应用于擅长模式识别的支持向量机算法,进行瓦斯突出样本数据的训练,构建瓦斯预测模型;最后,使用瓦斯预测模型对新的瓦斯突出数据进行预测。实验结果表明,采用该方法进行瓦斯突出预测的准确率,比纯支持向量机算法提高4.6%。
Coal and gas outburst is one of the major forms of mine disaster, which is extremely harmful. According to the on-site monitoring of gas outburst related data, predict the degree of gas outburst danger, do precautionary measures in advance, can greatly reduce the accident hazard. In this paper, a prediction model of gas outburst danger degree based on support vector machine and improved particle swarm optimization is proposed. By improving the particle swarm easily falling into the local optimum and applying improved particle swarm optimization to solve the problem of predicting the performance of SVM, Then the optimal parameters are applied to the SVM algorithm which is good at pattern recognition to train the gas outburst sample data to construct the gas prediction model. Finally, the gas prediction model is used to predict the new gas outburst data. The experimental results show that the accuracy of this method for gas outburst prediction is 4.6% higher than that of pure support vector machine.