反演在化工过程故障诊断中的探索

Exploration of Inversion in Fault Diagnosis of Chemical Process

为了提高化工装置的安全稳定运行能力,通过某种手段识别化工生产中的故障,发现潜在的设备故障和预判设备劣化趋势,将装置的安全隐患于萌芽中消除,避免企业安全损失。文章通过详细分析反演系统的原理、控制方法和化工过程故障诊断,结合化工过程故障诊断的现状,从混合故障、BP神经网络和深度神经网络三方面分析了故障诊断方法,最后结合某一化工过程进行故障反演,促进故障反演在化工过程中的应用。

In order to improve the safe and stable operation ability of chemical equipment,faults in chemical production can be identified by some means.By identifying potential equipment failures and predicting equipment deterioration trends,the safety hazards of the device can be eliminated in the bud to avoid safety losses of the enterprise.In this paper,the principle,control method and fault diagnosis of chemical process of inversion system are analyzed in detail.Combined with the current situation of chemical process fault diagnosis,the fault diagnosis methods are analyzed from three aspects of mixed fault,BP neural network and deep neural network.Finally,fault inversion is carried out based on a chemical process to promote the application of fault inversion in chemical process.