A detection and separation method for micro-faults in industrial process and its monitoring system

Abstract

The invention discloses a method for detecting and separating minor faults in an industrial process and a monitoring system thereof. The method of the present invention comprises the steps of: collecting sensor data under normal operating conditions of an industrial process as training data, and establishing a principal component analysis model of the training data; given a suitable sliding time window width, calculating each The improved reconstruction contribution value of the variable; determine the control limit of the improved reconstruction contribution of each variable; collect the sensor data under real-time working conditions as the test data; calculate the improved reconstruction contribution of each variable in the test data, and compare with the above Corresponding control limits are compared, and fault analysis is performed on the test data; if the result of the fault analysis is that a fault occurs, the variable with the largest contribution to improvement and reconstruction is determined as the fault variable to realize fault separation. Compared with the prior art, the method of the invention does not need a mathematical model of the industrial process, and can simultaneously realize the detection and separation of micro-faults in the industrial process.

Description

technical field [0001] The invention belongs to the field of industrial process monitoring and fault diagnosis, and in particular relates to a method for detecting and separating minor faults in industrial process and a monitoring system thereof. Background technique [0002] Modern industrial processes are large in scale and complex in structure. Once the process is abnormal, it may cause huge economic losses and even endanger personal safety. Process monitoring and fault diagnosis technology can effectively improve system reliability, equipment maintainability and reduce accident risk, and has become one of the research hotspots in the field of process control. In addition, more serious faults are usually evolved from small faults, and many major catastrophic accidents in history were also caused by small anomalies in the system that were not detected and resolved in time. Therefore, the harm of minor faults cannot be ignored, timely detection and separation of minor faul...

AI Technical Summary

Problems solved by technology

However, the traditional principal component analysis method has poor detection performance for small faults, resulting in a high rate of fault false positives (low detection rate)

In addition, the traditional reconstruction contribution graph method is also prone to mislocalization of fault variables when dealing with the problem of micro-fault separation.

The existing micro-fault diagnosis technology is mainly to improve the traditional fault detection algorithm, so that it is sensitive to micro-faults and obtains better detection performance, but rarely involves fault separation, and some algorithms have high computational complexity. conducive to practical application

Images