摘要
液压钻机故障特征参数与故障状态之间呈现较强的非线性,依赖线性数学模型的故障诊断方法诊断正确率不高。针对上述问题,提出了一种基于粒子群算法(PSO)优化BP神经网络的液压钻机故障诊断方法。该方法利用BP神经网络提取特征参数之间的非线性关系,实现典型故障的分类识别;利用PSO优化BP神经网络的权值和阈值,提高网络训练的收敛速度。仿真结果表明,PSO优化的BP神经网络迭代次数少,收敛速度快。该方法能够对测试样本进行有效分类,故障诊断正确率高。
Because of the strong nonlinearity between fault feature parameters and iault status of hydraulic drilling rig, the daagnostic accuracy of the fault diagnosis method which relies on linear mathematical model is not high. To solve this problem, a fault diagnosis method based on BP neural network optimized by Particle Swarm Optimization （ PSO } is proposed. With this method, the nonlinear relationship between the feature parameters is extracted by using BP neural network for realizing classification recognition of the typical faults ; and the weights and thresholds of BP neural network are optimized by adopting PSO for increasing the convergence rate of network training. The result of simulation reveals that the BP neural network optimized by PSO features smaller number of iterations and faster convergence rate; the diagnosis method can classify the testing samples effectively, and the accuracy of fault diagnosis is high.
出处
《自动化仪表》
CAS
2016年第4期42-46,共5页
Process Automation Instrumentation
关键词
液压钻机
故障诊断
BP神经网络
粒子群算法
全局优化
可靠性
数据分析
Hydraulic drilling rig Fault diagnosis BP neural network Particle swarm optimization （PSO） Global optimization Reliability Data analysis
