混合智能算法改进Elman模型的生物氧化釜多传感器系统故障诊断研究

Fault diagnosis of multi-sensor system in biological oxidation reactor based on Elman model improved by hybrid intelligent algorithm

生物氧化提金是一种新型的冶金工艺,该过程中需要用到多种传感器采集工艺数据,为氧化预处理过程的调节提供数据参考。传感器如果发生故障导致传输错误的数据,将会影响反应釜内的环境调节。文中主要对生物氧化提金系统中的多传感器系统进行故障诊断研究,首先提出一种SOA-PSO混合智能算法模型;然后使用4种标准测试参数进行实验,得出该混合智能算法有效;最后采用所提算法对Elman故障诊断模型的权值阈值进行寻优,建立SOA-PSO-Elman智能故障诊断模型。与PSO-Elmam模型和AFSA-PSO-Elman模型进行实验对比,仿真结果表明,SOA-PSO-Elman智能故障诊断模型可行有效。

Biological oxidation gold extraction is a new metallurgical process,in which a variety of sensors are required to collect process data to provide data reference for the adjustment of oxidation pretreatment process.If a sensor malfunctions,resulting in incorrect data transmission,it will affect the environmental regulation inside the reactor.In this paper,the fault diagnosis of the multi-sensor system in the biological oxidation gold extraction system is mainly studied.A SOA-PSO hybrid intelligent algorithm model is proposed.The four standard test parameters were used in the experiment to indicate the effectiveness of the hybrid intelligent algorithm.The proposed algorithm is used to optimize the weight threshold of Elman fault diagnosis model,and the SOA-PSO Elman intelligent fault diagnosis model is established.In comparison with PSO-Elmam model and AFSA-PSO-Elman model,the simulation results show that the SOA-PSO-Elman intelligent fault diagnosis model is feasible and effective.