Mixed fault information extraction and matching technique for chemical production

Abstract

The invention discloses a technical method for extracting and matching mixed fault information in chemical production, and relates to the chemical technology. The technical method for extracting and matching diagnosis mixed fault information comprises the following steps: using measured variable, control variable and operating variable in a technical process as mixed resources of process information, applying and counting analysis of related information to extract principal element information so that a parameter of the fault information in each node of a closed loop of a control system is dominant and the characteristic of the fault is magnified; and on this basis, applying regular dynamic time to carry out characteristic match of the fault information, and carrying out mode match of the related information based on the optimal path of Euclidean distance. The method uses the measured variable, the control variable and the operating variable as the mixed information source of a principal element, so that the information resources are more comprehensive and accurate, the capacities of the normal information and the fault information are increased, and performances of control and diagnosis are improved. The method has the advantages of advanced technique, theoretical principle existence, and strong actual application and operability.

Description

technical field [0001] The invention relates to monitoring and fault diagnosis of chemical production technology process, in particular to a technical method for process control, monitoring and fault diagnosis based on on-site sampling data of a process control system. Background technique [0002] The process control, monitoring and fault diagnosis of the chemical production process are important guarantees for the smooth, safe and reliable operation of the chemical production process. The traditional process control is to first establish a mathematical model of the process according to the process reaction mechanism, and then apply the control algorithm to control the process. Due to the complexity and variability of the chemical process and the uncertainty of the model, it is difficult to guarantee the accuracy of the control. With the development of computer technology, network technology and communication technology, methods based on statistical process analysis for pro...

AI Technical Summary

Problems solved by technology

The data features are extracted in the sample space of a large number of variables in the process, so that only the measurement variables are processed, and a large number of control variables and operating variable information are not used, resulting in a large amount of information loss, which affects the accuracy and reliability of data processing. , which also affects the performance of process control and fault diagnosis

Images