文献类型：期刊文献

英文题名：Based on Symplectic Geometry Decomposition Multimodal Symmetric Dot Pattern Marine Diesel Engines Fault Diagnosis Method

作者：Jia, Baozhu[1]; Liang, Guanlong[1]; Huang, Zhende[1]; Niu, Jinzhang[1]; Song, Xuewei[1]; Liao, Zhiqiang[1]

机构：[1] Naval Architecture and Shipping College, Guangdong Ocean University, Zhanjiang, China

年份：2024

起止页码：1

外文期刊名：2024 International Conference on Intelligent Ships and Electromechanical System, ICISES 2024

收录：EI(收录号：20253719172731)

语种：英文

外文关键词：Diesel engines - Electric fault currents - Geometry - Image processing - Marine engines - Vibration analysis

外文摘要：Aiming at the problems of complex structure of marine diesel engines, strong background noise in collected signals, and limited fault information in single-dimensional signals, this paper proposes a fault diagnosis method for marine diesel engines based on Symplectic Geometry Decomposition Multimodal Symmetric Dot Pattern (SMSDP). This method monitors the operating status of marine diesel engines through vibration signals and utilizes SMSDP for multi-scale analysis to enhance the signal multimodal features, thereby achieving enhanced fault features. Subsequently, an adaptive similarity judgment method is employed for diagnosing marine diesel engine faults. The effectiveness of the proposed method is validated by marine diesel engine data, The results show that the diagnosis accuracy of the proposed method is 100%. demonstrating the feasibility and effectiveness of the proposed method. Furthermore, by comparing with different signal processing methods, the proposed method has the highest diagnosis accuracy, which verified the proposed method has a superior performance to diagnose the marine diesel engines fault. ? 2024 IEEE.