摘要
在煤炭机电设备故障检测过程中,由于涉及到众多的硬件设备,容易出现电压稳定性较差,间接影响故障检测信号不稳定,检测样本数据空间维数大、诊断实时性差等缺点。本文提出了采用动态模糊自学习理论和BP神经网络相结合的方法针对故障进行诊断,首先通过动态模糊自学习方法对设备故障的有效数据,使用BP网络对其进行快速分类诊断。仿真实验表明:本文算法能够有效地提高故障诊断正确率,从而提高诊断的识别与决策能力。
During the process of detecting coal and electrical equipment, as numerous hardware devices have been involved, the voltage stability is poor, indirectly influencing the instability of detected signal and the samples' data have large spatial dimension and the detection is of poor timeliness. In this paper,the combination of dynamic fuzzy self-learning theory and BP neutral network to detect the faults. First of all, effective data of the equipment fault are collected with dynamic fuzzy learning method, and then BP network is used to quickly detect and classify it. Simulation experiment indicates that algorithm in this paper can effectively improve the accuracy of fault detection so as to improve the detection's recognition and decision-making capacity.
出处
《科技通报》
北大核心
2016年第10期195-198,249,共5页
Bulletin of Science and Technology
基金
重庆市教委科学技术研究项目(KJ132202)
