论文部分内容阅读
针对煤矿开采过程中非线性,强耦合性等特点所致的动力灾害难以预测的问题,选用一种新的线性生成机制(LGMS)改进果蝇算法(FOA)优化广义回归神经网络(GRNN)的方法,建立了冲击地压煤岩灾害预测模型。采用LGMS改进FOA,避免FOA优化GRNN时陷入局部最优,增强了其搜索全局最优解的能力,提高了GRNN的收敛性与和预测精度。选取冲击地压前的电磁辐射、声发射、红外辐射3个主要指标,根据3种指标的单项危险指数求得综合危险指数,构建冲击地压动力灾害预测的LGMSFOAGRNN模型。研究表明,所构建的LGMSFOA-GRNN模型具有很好的预测能力和泛化能力。
Aiming at the unpredictable dynamic disasters caused by the nonlinear and strong coupling characteristics in the coal mining process, a new linear generation mechanism (LGMS) is proposed to improve the performance of the Drosophila algorithm (FOA) to optimize the GRNN Method, a prediction model of rockburst in coal and rockburst has been established. Using LGMS to improve FOA and avoid FOA optimization GRNN into local optimum, enhance its ability to search global optimal solution, improve the convergence and prediction accuracy of GRNN. Three major indexes of electromagnetic radiation, acoustic emission and infrared radiation before rockburst are selected. The comprehensive danger index is calculated according to the single-hazard index of three kinds of indicators, and the LGMSFOAGRNN model for predicting the dynamic impact of the rockburst is established. The research shows that the constructed LGMSFOA-GRNN model has good predictive ability and generalization ability.