论文部分内容阅读
采煤机截割部摇臂齿轮箱承担着综采工作面截割部动力传动的重任,其故障与否直接影响采煤机正常工作。而传统的故障诊断方法-BP神经网络采用基于梯度下降的算法,存在容易陷入局部极小值、收敛速度慢等不足,这些不足严重影响了BP网络的应用。然而粒子群算法(PSO)有很好的全局收敛特性。因此,为了提高网络的性能,采用粒子群算法来优化BP神经网络,将改进的PSO引入神经网络的拓扑结构,用PSO的迭代代替BP中的梯度修正。结果表明:提出的改进方案可以有效地优化神经网络,提高其在采煤机齿轮箱故障诊断中的应用价值。
The shearer gear arm of cutting section of shearer takes on the responsibility of power transmission of the cutting section of fully mechanized coal mining face, whose failure directly affects the normal operation of shearer. However, the traditional fault diagnosis method-BP neural network uses the algorithm based on gradient descent, which is easy to fall into the local minimum and the convergence speed is slow. These problems seriously affect the application of BP network. However, Particle Swarm Optimization (PSO) has good global convergence. Therefore, in order to improve the performance of the network, particle swarm optimization is used to optimize BP neural network, the improved PSO is introduced into the topological structure of neural network, and the PSO iteration is used to replace gradient correction in BP. The results show that the proposed scheme can effectively optimize the neural network and improve its application value in the fault diagnosis of shearer gearbox.