Submersible micro-fault diagnosis method, device, computer equipment and storage medium

Abstract

The invention provides a submersible micro-fault diagnosis method, device, computer equipment and storage medium. The method includes: building a chance-constrained model of the submersible based on the diagnostic model of the submersible, wherein the diagnostic model includes a micro-fault model and a normal model ; Transform the chance constraint model into an expectation constraint model; based on the expectation constraint model, obtain the functional relationship of the Bhattacharyachian distance between the output of the minor fault model and the output of the normal model, and maximize the Bhattacharyachian distance in the functional relationship, and the calculation is assisted signal; based on the Bayesian decision-making of the minimum error rate, the auxiliary signal is input into the submersible to obtain the fault diagnosis result; the active diagnosis of the small fault of the submersible is realized. In addition, the auxiliary signal does not violate the linear constraints of the actuator, and its injection The chance constraints and transformed expectation constraints of the system are not violated, so the auxiliary signal is physically achievable and its injection does not damage the submersible, thus ensuring a high fault diagnosis rate.

Description

technical field [0001] The present invention relates to the technical field of deep-sea submersibles, in particular to a method, device, computer equipment and storage medium for diagnosing micro-faults of submersibles. Background technique [0002] As an important marine scientific research platform, the safety and reliability of deep-sea manned submersibles determine the personal safety of submersibles and the smooth development of marine scientific research operations, as well as my country's deep-sea exploration and development capabilities. Since deep-sea manned submersibles work in a deep-sea environment with high pressure, high salt, and strong uncertainty, minor system failures are likely to evolve into significant failures, directly threatening the safe operation of submersibles in deep-sea environments and the safety of diving personnel. Personal safety. The complex and changeable deep-sea environment is difficult to meet the conditions for timely fault diagnosis ...

AI Technical Summary

Problems solved by technology

The complex and changeable deep-sea environment is difficult to meet the conditions of timely diagnosis and fault-tolerant control of faults. Therefore, fault diagnosis and safety maintenance of deep-sea manned submersibles before launching are very important, which will directly affect the safety of submersibles and diving personnel and the completion quality of deep-sea exploration and development operations

[0003] At present, the micro-fault diagnosis methods of deep-sea manned submersibles are all passive diagnosis methods, that is, only using the original input and output data of the submersible to generate residual signals or corresponding statistics indicating faults, and on this basis, the possible faults are analyzed. Analysis; fault diagnosis is not effective when the data contains fault characteristics that are not obvious

However, due to the characteristics of low amplitude of minor faults and easy to be covered by system noise, the system representation of faults is not much different from that of the system when the system is normal, so it is difficult for passive fault diagnosis methods to diagnose minor faults of submersibles in time.

Images