基于GM(1,N)_GRNN组合模型的瓦斯涌出量预测研究

为解决瓦斯涌出量影响因素众多、难以准确预测的问题,本文利用多变量灰色系统易于处理不规则数据,GRNN神经网络模型训练速度快、人为干预因素少等优势,建立起1阶N变量灰色模型与GRNN神经网络嵌入型组合模型GM(1,N)_GRNN。用该模型对某煤矿回采工作面的瓦斯涌出量进行了预测,并与GM(1,N)模型、GRNN两种模型单独预测的结果做了对比,发现组合模型预测结果的平均误差仅3.7%,明显优于两种模型单独预测的平均误差。因此,对煤矿安全生产有重要指导意义。

Prediction of Gas Emission Based on GM (1, N) _GRNN Combined Model

In order to solve the problem that there are many influencing factors of gas emission, it is difficult to accurately predict the problem. In this paper, we use multivariable gray system to deal with irregular data. The GRNN neural network model has the advantages of fast training speed and low human intervention. The model of the 1 order N variable grey model and the GRNN neural network embedded model GM (1, N) _GRNN were established. The model is used to predict the gas emission from a coal mining face and compared with the GM (1, N) model and GRNN model alone. It is found that the average error of the combined model is only 3.7%, which is better than the average error of the two models. Therefore, the safety of coal production has important guiding significance.