Closed-loop fault modeling and active diagnosis method under under-data

Abstract

The invention discloses a closed-loop fault modeling and active diagnosis method under under-data. The method comprises the following steps: step 1 of a closed-loop fault modeling problem for petrochemical equipment under the under-data; step 2 of a second-best solution problem for the excitation input; and step 3 of an active diagnosis problem for faults of a closed-loop system. According to theclosed-loop modeling under the under-data, the method provided by the invention performs dynamic analysis based on the physics principle for the faults of which the mechanism can be speculated, performs approximation on the dynamic operation based on a mathematical approximation method for the faults of which the action mechanism is difficult to be speculated, and finally determines the parametersof the fault model by the specific actual system fault data in different operating environments. In this process, the data compatibilization technology is used to reasonably construct the output signal of the system, so that the abnormality in the system can be more effectively reflected.

Description

technical field [0001] The invention relates to the technical field of electronic products, in particular to a closed-loop fault modeling and active diagnosis method under data shortage. Background technique [0002] Petrochemical equipment generally adopts a closed-loop control mode. On the one hand, the closed-loop control improves the stability and robustness of the system. On the other hand, when abnormal conditions occur in the system, the output changes are reduced, resulting in an obvious cover-up effect on some faults. If there are nonlinear and uncertain links in the system, the relationship between the open and closed loop residual signals will become extremely complicated, making it difficult for the existing methods based on analytical models to obtain ideal results. Another difficulty brought about by closed-loop control is the propagation of abnormal conditions in the system. Faults in certain parts may affect the performance of multiple parts of the system alo...

AI Technical Summary

Problems solved by technology

[0007] Step 1. Aiming at the problem of closed-loop fault modeling of petrochemical equipment under data shortage: According to the operating environment of different loads and objects of petrochemical equipment, adopt the method of combining mechanism and data to realize fault classification and fault modeling in complex environments, and analyze petrochemical The main influencing factors of the equipment operating environment, complete the division of the possibility of failure and the environmental dependence, and classify the failures through the extraction of fault characteristics; on this basis, according to the different fault characteristics that can be caused by different environments Establish a more practical and targeted mathematical description method;

Images