Behavior based UUV (unmanned underwater vehicle) propulsion operating system exception identification method

Abstract

The invention relates to a behavior based UUV (unmanned underwater vehicle) propulsion operating system exception identification method which includes, through an observer, estimating status information of a UUV, such as location, velocity and angle in real time and comparing the status information with status information acquired by a UUV sensor in real time to acquire residual information of the status information; setting top limit threshold, bottom limit threshold and time window of the residual information, rejecting residual signals of false alarm and judging whether behavior of the UUV is abnormal or not primarily; matching specific two-dimensional space identification behaviors from an identification behavior library and determining self exception point locations and danger levels; performing matching to different exception modes of an UUV propulsion operating system according to residual characteristics of system status data acquired in real time to realize accurate positioning of the exception points of the propulsion operating system; and finally determining danger levels of the exceptions. Propulsion operating exception points are positioned accurately and exception danger levels are determined according to the specific identification behaviors in two-dimensional motion space, so that exception identification difficulty is reduced and exception identification accuracy is improved.

Description

technical field [0001] The invention relates to a behavior-based UUV propulsion control system abnormal identification method. Background technique [0002] With the rapid development of ocean development, people have put forward higher requirements for the intelligent operation level of UUV operations, which makes the safety and reliability of UUV very important in research and application. As a complex nonlinear system, UUV will not only fail to complete the expected tasks, but may also cause major economic losses once an abnormality occurs. Therefore, early identification of abnormal lines for UUV systems is of great significance. Among the various abnormalities of UUV, the failure rate of the propulsion control system is usually high, and the abnormality of the propulsion control system may cause irreparable consequences, so it is particularly important to identify the abnormality of the propulsion control system. [0003] At present, domestic scholars are researching ...

AI Technical Summary

Problems solved by technology

[0003] At present, domestic scholars are researching the fault diagnosis of UUV thrusters. Wang Yujia used the technology of combining fuzzy logic and neural network in the article "Research on Fault Fusion Diagnosis Method of Underwater Robot Thruster" in "Journal of Bullets, Arrows and Guidance" in 2008. To monitor the working status of the propulsion system, Zhu Daqi from Shanghai Maritime University et al. aimed at the fault identification of UUV actuators and sensors in the article "Fault Diagnosis and Fault-tolerant Control of Unmanned Underwater Robot Propulsion System" in the "Journal of System Simulation" in 2010. Problem Constructed a fault identification model based on a recurrent neural network, but due to the need for a large number of samples for training, the ability to identify new faults is limited, and the present invention starts from the behavior point of view, as long as the behavior of the UUV is abnormal, the fault can be found

Images