基于改进果蝇优化BP神经网络的冲击地压预测

针对煤矿开采过程中存在非线性、强耦合性等特点导致的动力灾害难以预测的问题，引入一种候选解的线性生成机制(LGMS)、混沌搜索、粒子群算法(PSO)和模拟退火算法(SA)修正果蝇算法(IFOA)，利用改进后的果蝇优化算法良好的搜索全局最优解的能力, 自适应地调整 BP 网络的权值和阈值，建立了煤岩冲击地压灾害预测模型。以唐山开滦煤矿样本数据为例进行仿真验证，结果表明其鲁棒性和测量精度明显提高，且网络具有较强的收敛性能和优化能力。

Prediction of Rock Bust Based on Improved FOA-BP Neural Network

In view of the problem that the dynamic disaster is unpredictable due to the non-linearity and strong coupling in the coal mining process, this paper introduced a Linear Generation Mechanism of candidate Solution(LGMS), Chaotic Search, Particle Swarm Optimization algorithm(PSO) and Simulated Annealing algorithm(SA) to modify Fruit fly algorithm(IFOA), and then by using the capability of searching the global optimal solution of modified FOA, the weight and threshold of BP neural network were adjusted adaptively, a prediction model of rock burst was established. Finally, taking the sample data of Kailuan Coal Mine in Tangshan as an example for simulation verification, the experimental results showed that the accuracy of robustness and measurement are obviously improved, and the network has strong convergence performance and optimization ability.