论文部分内容阅读
复杂工业流程存在多稳态工况切换及切换的过渡过程,导致传统PCA(principal component analysis)故障检测方法易于误报故障。本研究提出了基于稳态因子的过渡过程判别方法及基于相似因子的工况自适应匹配方法,将其融入PCA构建了新的故障检测方法,且将该方法用于电站锅炉补给水处理流程的故障检测。以该流程的全工况运行数据对算法进行了验证,结果表明:此方法能有效消除过渡过程的影响,并能通过工况匹配提高故障检测性能且减少故障误报,可以实现水处理流程的全工况故障检测。
Complex industrial processes have the transition process of multi-steady state switching and switching, leading to the easy fault detection of the traditional principal component analysis (PCA) fault detection method. In this study, a method of transitional process identification based on steady-state factor and a condition-based adaptive matching method based on similarity factor were proposed. A new fault detection method was built into PCA, and the method was applied to the process of power plant boiler feedwater treatment Fault detection. The algorithm is validated by the data of the full condition of this process. The results show that this method can effectively eliminate the influence of the transition process, and can improve the fault detection performance through matching of working conditions and reduce the false alarm, so that the water treatment process Full condition fault detection.