Design and Implementation of a Virtual Maintenance Training System for a Certain Equipment Based on Digital Twin Technology

Aiming at the issues of complex structure, numerous components, difficult fault diagnosis, and strong maintenance dependency of marine equipment, a virtual maintenance training system for a certain equipment based on digital twin technology is proposed to enhance the maintenance capabilities of crew members. The system is designed based on the “five-dimensional” model of digital twins and utilizes Unity3D technology to achieve equipment modeling and realistic training simulation. In the service dimension of the model, on one hand, the system records real-time user operation data from the PC end and employs the FP-growth algorithm to mine user training data to achieve customized assessment, thereby promoting training through evaluation. On the other hand, it uses intuitionistic fuzzy Petri nets for forward and backward reasoning of equipment faults and introduces information entropy to adjust the fuzziness of equipment component faults, thereby improving the accuracy of fault reasoning and diagnosis and achieving precise fault location. Through solving and analyzing the practical cases, the system can effectively address the maintenance capability shortcomings of trainees in multiple dimensions.