WCSE

Real-Time Ship Fault Diagnosis Algorithm Based on the Fusion of CNN and RNN

Abstract— As the main tool of water transportation, the safety and stability of ship operation play a crucial role in the water transportation industry. However, the existing fault diagnosis methods of ship equipment cannot meet the real-time diagnosis of multiple equipment, to avoid the subjectivity of manual feature extraction and meet the real-time diagnosis of ship faults, designs a real-time fault diagnosis method for ships based on the fusion of convolutional neural network and recurrent neural network in this paper. This method directly inputs the raw data from multiple sensors in a certain window into the network after correlation processing, automatically extract features by convolutional neural network and complete fault diagnosis by recurrent neural network. Experimental results show that this method does not require manual extraction of data features, and its fault diagnosis accuracy is high and the response time is short. Therefore, the method proposed in this paper can be applied to the real time fault diagnosis of ships.

Index Terms—real-time fault diagnosis of ship faults, convolutional neural network (CNN), recurrent neural network (RNN)

Youyu Zeng, Qiang Xie
Nanjing University of Aeronautics and Astronautics, CHINA

[Download]

Cite: Youyu Zeng, Qiang Xie , " Real-Time Ship Fault Diagnosis Algorithm Based on the Fusion of CNN and RNN " Proceedings of 2020 the 10th International Workshop on Computer Science and Engineering (WCSE 2020), pp. 70-75, Shanghai, China, 19-21 June, 2020.