Grey-forecasting-model-based method for predicting weak fault degrees of autonomous underwater vehicle propeller

Abstract

The invention provides a grey-forecasting-model-based method for predicting weak fault degrees of an autonomous underwater vehicle propeller. In a grey background value construction stage, the proximate integral of an accumulative generation sequence is calculated to serve as a grey background value; in a winterization equation solving stage, through a difference value between a predicted original-sequence value and an actual original-sequence value, a point with the minimum predictive residual is determined, and an original sequence value corresponding to the point with the minimum predictive residual serves as an initial value of a winterization equation solution; in a predicted sequence construction stage, on the basis of a residual sequence between a predicted sequence and an original sequence, secondary prediction is conducted, and on the basis of a residual sequence obtained by the secondary prediction, the predicted sequence of the original sequence is amended to make a prediction result adjustable. By the adoption of the grey-forecasting-model-based method for predicting the weak fault degrees of the autonomous underwater vehicle propeller, the problem is solved that prediction errors are relatively large when the weak fault degrees of the AUV propeller are predicted, and a grey background value construction method, winterization equation solving method and predicted sequence construction method of a traditional grey method are improved separately.

Description

technical field [0001] The invention relates to the technical field of fault diagnosis of autonomous underwater robots, in particular to a method for predicting the weak fault degree of propellers of autonomous underwater robots based on a gray prediction model. Background technique [0002] Autonomous underwater vehicles (AUVs) work unmanned and without cables in complex and changeable marine environments. Safety is an important feature of AUVs. Condition monitoring and fault diagnosis are the basis and key technologies to ensure the safety of AUVs. AUV is composed of multiple parts, among which the thruster is the key component and the heaviest load component. Researching the thruster fault diagnosis technology is of great significance to improve the safety of AUV. Many scholars have achieved good research results in AUV thruster fault diagnosis technology, but most of them focus on the hard faults of the propeller and the faults with a large output loss, and seldom study ...

AI Technical Summary

Problems solved by technology

These reasons lead to a large prediction error when using the traditional gray GM(1,1) method to predict the degree of propeller weak faults

[0006] Aiming at the problem that the traditional gray GM (1,1) method is used to predict the weak fault degree of AUV thrusters, the prediction error is relatively large. The pre

Images