基于PCMRA-SVM的采煤工作面瓦斯涌出量预测

Prediction for Gas Emission Quantity of Working Face Based on PCMRA-SVM

为了准确、快速地预测采煤工作面瓦斯涌出量,针对瓦斯涌出系统的特点,提出了一种基于PCMRA-SVM的瓦斯涌出量预测模型。以钱营矿的25组瓦斯涌出量观测数据进行仿真实验,并与BPNN、SVM、CIGOA-ENN方法的预测结果进行对比。结果表明:PCMRA-SVM模型的最大、最小和平均相对误差分别为4.06%、0.02%和1.73%,均优于CIGOA-ENN、SVM、BPNN,验证了所提出模型的有效性、可靠性及准确性。

In order to forecast the gas emission quantity in the coal mining face accurately and rapidly,a prediction model of gas emission quantity based on PCMRA-SVM was proposed. The simulation experiment was carried out on 25 groups of gas emission data from the Qianying coal mine,and the forecast results were compared with the results of BPNN,SVM and CI- GOA-ENN method. The research results show that the maximum, minimum and average relative errors of PCMRA-SVM model were 4. 06% ,0. 02% and 1.73% ,which were better than CIGOA-ENN, SVM and BPNN,which verify the validity,relia-bility and accuracy of the proposed model.