论文部分内容阅读
为评价冲击矿压危险程度,提出一种基于粒子群算法和BP神经网络(PSO-BP)的冲击危险评估方法。利用已有冲击矿压数据,通过BP网络建立回归模型,并采用PSO算法对模型的连接权重和阀值进行优化,克服了BP网络收敛速度慢、易陷入局部最优的缺点。选取冲击矿压的10种主要影响因素,利用典型冲击矿井的20组工程数据建立PSO-BP评估模型,并将该模型与标准BP模型进行对比分析,结果表明PSO-BP模型较标准BP模型的评估准确率提高15%。最后,通过某矿冲击危险评估的工程实例验证了该方法的可行性和普适性。
To evaluate the risk of rock burst, a method based on PSO and BP neural network (PSO-BP) is proposed to evaluate the impact hazard. The existing rockburst data are used to establish the regression model through BP network and the PSO algorithm is used to optimize the connection weights and thresholds of the model, which overcomes the shortcomings of slow convergence speed and easy falling into local optimum. The 10 main influencing factors of rockburst are selected. The PSO-BP assessment model is established by using 20 sets of engineering data from a typical impact mine and the model is compared with the standard BP model. The results show that the PSO-BP model is better than the standard BP model Evaluation accuracy increased by 15%. Finally, the feasibility and universality of this method are verified by engineering examples of the risk assessment of a mine impact.