Ship fault real-time diagnosis method based on CNN and RNN fusion

Abstract

The invention relates to a fault diagnosis method based on CNN and RNN fusion. The method can be used for ship real-time fault diagnosis. The method comprises the following steps: firstly, establishing a correlation matrix between ship characteristic parameters by utilizing historical ship data; the correlation matrix can represent the relationship between attributes; then adding multi-sensor original data in a sliding window acquisition window for correlation processing, taking the processed multivariable time sequence as the input of a convolutional neural network, automatically extracting fault features by the convolutional neural network, and finally, recombining the extracted feature vectors as the input of a recurrent neural network for fault classification. Experimental results showthat the method does not need to manually extract data features, and is high in fault diagnosis accuracy and short in response time. The advancement of the method is mainly reflected in that the method has a good effect in ship fault diagnosis and classification, is quick in time response, and can meet the use requirements of ship fault real-time diagnosis.

Description

technical field [0001] The invention belongs to the technical field of ship systems, in particular to a real-time fault diagnosis method for multiple sensor raw data. Background technique [0002] As the main tool of water transportation, the safety and stability of ship operation play a vital role in the water transportation industry. Due to the large number of ship equipment and the complex operating environment, the probability of equipment failure is greatly increased. If the equipment fails to be found in time, it will bring great danger to the navigation of the ship. Fault diagnosis is of great significance to the health management of ships by detecting and identifying faults. Effective fault detection can improve the safety of ships, so it is necessary for real-time online fault diagnosis of ship equipment. Contents of the invention [0003] Purpose of the invention: The purpose of the present invention is to overcome the problem that the existing technology cannot...

AI Technical Summary

Problems solved by technology

[0003] Purpose of the invention: The purpose of the present invention is to overcome the problem that the existing technology cannot meet the real-time diagnosis of multiple devices, and propose a real-time diagnosis method for ship faults based on CNN and RNN fusion with reasonable design, high accuracy and short response time

Images