基于CSABC-ELM的采煤工作面瓦斯涌出量预测

为了提高煤矿工作面瓦斯涌出量的预测精度,研究一种将极端学习机(ELM)与利用混沌搜索策略改进的人工蜂群(CSABC)算法相结合的预测方法。改进后的人工蜂群算法有效解决了ABC算法易陷入局部最优、后期收敛慢等缺陷,利用CSABC优化ELM的输入层和隐含层参数,避免了随机产生ELM参数所造成的误差,建立基于CSABC-ELM的瓦斯涌出量预测模型。利用实际煤矿监测数据对该模型进行试验分析,并与ABC-ELM,ELM和BP神经网络的预测结果进行比较。结果表明,CSABC-ELM预测误差更小,精度更高,泛化性能也更强,能有效地对煤矿瓦斯涌出量进行预测。

Gas Emission Prediction Based on CSABC-ELM in Coalface

In order to improve the prediction accuracy of gas emission in coalface,a prediction method combining extreme learning machine(ELM)with artificial bee colony(CSABC)algorithm improved by chaos search strategy is studied.The improved artificial bee colony algorithm effectively solves the defect that ABC algorithm is easy to fall into local optimum and slows down later.The input layer and hidden layer parameters of ELM are optimized by using CSABC to avoid the error caused by randomly generated ELM parameters,A gas emission prediction model based on CSABC-ELM was established.The actual coal mine monitoring data are used to test the model and compared with the prediction results of ABC-ELM,ELM and BP neural network.The results show that the prediction error of CSABC-ELM is smaller,the precision is higher and the generalization performance is stronger.Therefore,it can effectively predict the gas emission of coal mine.