摘要
针对铝电解槽故障特征种类繁多,难以快速准确的实现故障类型诊断,设计了一种基于最大-最小蚂蚁系统(MMAS)优化的极限学习机(ELM)故障诊断方法;介绍了电解槽常见的故障类型及其对槽电压的影响,对采集到的故障情况下的槽电压信号进行降噪处理,根据对降噪后故障信号的局域均值分解(LMD)结果得到故障特征;采用ELM算法辨识故障类型,针对ELM算法存在的参数问题,采用MMAS对ELM隐含层参数寻优;结果表明,MMAS优化的ELM既保证了较快的训练速度,同时获得了更高的故障测试正确率。
In view of problem that reduction cell had variety of fault characteristic species and it was difficult to achieve fast and accurate diagnosis of fault types. An extreme learning machine fault diagnosis method based on Max--Min Ant System was designed. Described the cell common type of fault and its effect on cell voltage. All the cell voltage signals when fault occurs were denoised. Get the fault characteris- tic according to the LMD decomposition result of fault signals after denoise. ELM algorithm was adopted to identify the fault types. In view of problem that existent parameters of ELM algorithm. MMAS was employed to search for the optimal hidden layer weights and thresholds. The experimental results showed that MMAS to Optimize the ELM had both high training speed and test accuracy.
出处
《计算机测量与控制》
2015年第10期3326-3329,共4页
Computer Measurement &Control
作者简介
孙伟（1963-），男，江苏徐州人，教授，博士生导师，主要从事复杂工业过程控制方面的研究。
