矿井瓦斯涌出量预测方法

Prediction of mine gas emission

为解决煤矿瓦斯涌出量预测不准确的问题,提出基于多种方法优化融合的瓦斯涌出量预测方法,建立瓦斯涌出量预测模型.采用适用于瓦斯涌出量系统特点的加权策略函数对最小二乘支持向量机进行改进,利用免疫遗传算法对加权最小二乘支持向量机进行核参数和正则化参数寻优.应用状态转移概率修正预测误差残值,使瓦斯涌出量预测模型的预测精度得到提高.研究结果表明:矿井瓦斯涌出量预测模型具有较好的快速性和准确性,具有广泛的应用前景.

In order to solve the accuracy problem of coal mine gas emission prediction, several methods of gas emission prediction, based on optimizing and fusion, were proposed. And a gas emission prediction model was established. The application of weighted function method in gas emission system characteristics is to improve the least squares support vector machine. Optimizing kernel parameter and regularization parameter of weighted least squares support vector machine by using immune genetic algorithm. Prediction accuracy of the gas emission prediction model is improved, because of the application of the state transition probability, which corrects prediction error residuals. Experimental results show that the model of mine gas emission prediction has the widespread application prospect, with its rapidity and accuracy.