基于SOM-RBF算法的瓦斯涌出量动态预测模型研究

发布时间：2018-04-10 02:11

本文选题：多传感器　切入点：瓦斯涌出量　出处：《传感技术学报》2015年08期

【摘要】：针对煤矿瓦斯涌出量的多影响因素预测问题,以多传感器的瓦斯监测系统采集处理后的数据作为样本,提出了一种自组织特征映射神经网络(Self-organizing Feature Maps,SOM)与多变量的径向基函数(Radial Basis Function,RBF)结合的组合人工神经网络的模型动态预测新方法。采用先聚类、再分类建模和预测的方法,解决了由于训练样本有限和训练样本点分散所导致的预测精度降低的问题,并通过矿井监测到的各项历史数据进行试验。结果表明,与其他预测模型相比较,该模型的预测精度更高,泛化能力更强。预测平均相对误差为2.16%,均相对变动值ARV为0.005 9,均方根误差RMSE为0.131 1,有效地实现了对煤矿绝对瓦斯涌出量的动态预测,有较高的实用价值。
[Abstract]:Aiming at the problem of multi-influencing factors prediction of coal mine gas emission, the data collected and processed by the multi-sensor gas monitoring system is taken as the sample.A new method of model dynamic prediction based on self-organizing feature mapping neural network (Som) and multivariable radial basis function (RBF) combined artificial neural network (Ann) is presented in this paper.The method of first clustering, then classifying modeling and forecasting is used to solve the problem of low prediction accuracy caused by limited training samples and scattered training samples, and the experiments are carried out through various historical data monitored by mines.The results show that compared with other prediction models, the prediction accuracy of the model is higher and the generalization ability is stronger.The average relative error of prediction is 2.16, the average relative variation value ARV is 0.005, and the root mean square error RMSE is 0.131 1. The dynamic prediction of absolute gas emission in coal mines is realized effectively, which is of high practical value.
【作者单位】：辽宁工程技术大学电气与控制工程学院;国网辽宁省电力有限公司辽阳供电公司;
【基金】：国家自然科学基金项目(51274118) 辽宁省教育厅基金项目(L2012119) 辽宁省科技攻关项目(2011229011)
【分类号】：TD712.5;TP183

【参考文献】