System and method for fault prediction and on-line evaluation of health state of marine diesel generator

Abstract

The invention provides a system and method for fault prediction and on-line evaluation of health state of marine diesel generator, which includes a shell, a displayer, an alertor, a power supply and adata processing device, wherein the data processing device include a deep learning integration module, a full life cycle history database, and a fault multimodal expert system database. The deep learning integration module utilizes an integrated learning method, and includes multiple supervised and unsupervised deep learning network models. The fault multimodal expert system database includes allfault feature databases and corresponding fault category labels, and also comprises a memory, a GPU, a CPU, an external device control interface, a data acquisition module and a wireless signal transceiving device, and realizes automatic monitoring and evaluation of an equipment. The system has the advantages of more accurate and convenient fault prediction, state on-line monitoring and the health evaluation.

Description

technical field [0001] The invention belongs to the technical field of fault prediction and health assessment of marine diesel generators, and in particular relates to a system and method for fault prediction and health status online evaluation of marine diesel generators. Background technique [0002] With the rapid improvement of modern technology and advanced manufacturing technology, ships are developing in the direction of large-scale, intelligent, automated, integrated and multi-functional. As one of the important power sources of large ships, marine diesel generators are important for ensuring high efficiency and stability of ships. Sailing has an irreplaceable role. Marine diesel generators operate continuously for a long time under complex and changeable sea conditions, with heavy workloads, variable loads, frequent switching between paralleling and electrolysis, and are prone to various failures due to the influence of salt-alkali corrosion and high temperature. L...

AI Technical Summary

Problems solved by technology

[0003] Before the present invention, there were relatively few products or methods on the market for fault prediction and repair and maintenance of marine diesel generators, and the traditional "after-event maintenance", "planned maintenance" and "scheduled maintenance" for land equipment were still used more. "Maintenance" method, but this method is increasingly unsuitable for the needs of modern shipping, because in the event of a sudden failure at sea, the crew cannot be given enough time to repair at sea, and external rescue cannot be in place in time, and this kind of ship It is impossible for large-scale equipment with long endurance to reverse flight when encountering problems, so the traditional methods are often very inefficient and not intelligent. In the past, regular maintenance and replacement of parts based on experience were used to estimate the life of parts based on experience. The method is easy to cause waste and misjudgment, and brings safety hazards, so it cannot meet the needs of crew members for intelligent fault prediction and health assessment

Images