基于SSA优化BP神经网络的冲击地压数据预测研究

Research on Rock Burst Data Based on SSA Optimized BP Neural Network

为解决目前冲击地压实验数据误差较大,冗余信息较多的问题,结合神经网络和麻雀搜索算法各自的特点,提出基于麻雀搜索算法(SSA)优化BP神经网络的数据预测模型。结合三轴声发射试验的试验数据,运用MATLAB等软件进行综合仿真模拟实验,并将SSA-BP算法的预测结果与BP神经网络的模拟预测结果进行仿真比较。仿真结果表明,引用SSA算法优化BP神经网络避免了BP权值平衡的缺点,不仅提高了学习的速度,而且具有更高的计算精度。

In order to solve the problems of large error and redundant information of rock burst experimental data,Combined with the characteristics of neural network and sparrow search algorithm,A data prediction model based on sparrow search algorithm(SSA)optimized BP neural network is proposed.Combined with the experimental data of triaxial acoustic emission test,MATLAB software was used for simulation experiments,The prediction results of SSA-BP algorithm and BP neural network are compared.Simulation results show that,Using SSA algorithm to optimize BP neural network avoids the disadvantage of BP weight balance,not only improves the learning speed,but also has higher calculation accuracy.