Deep Fault Diagnosis of Marine Diesel Generator Sets Based on Multi-Level Decision Fusion Model

Abstract

The effective fault diagnosis of the diesel generator sets can enable maintenance personnel to locate the fault location timely and accurately, and find the cause of the fault, which plays a vital role in ensuring the stable and reliable operation of the diesel generator sets. In order to fuse multiple evidence bodies and solve the problems such as uncertainty in fault diagnosis caused by single information, this paper combines the Deep Belief Network with D-S evidence theory to build a multi-level decision fusion model to realize the deep diagnosis of diesel generator sets fault, which can not only locate the fault location but also get the fault cause. The calculation results show that the model is effective in improving the accuracy of fault diagnosis for diesel generator sets, and improves the reliability of fault diagnosis.