基于数字孪生的行车多因素隐性级联故障诊断

To realize the intelligent transformation and upgrading of the state operation and maintenance of the crane, an implicit cascade fault diagnosis and analysis method was proposed based on digital twin technology and data-driven method for the multi-agency linkage and multi-factor coupling of the crane. With data-driven as the core, a data-driven digital twin model of the crane was constructed, and the composition and interaction behavior of the digital twin system of the crane were elaborated. The explicit and implicit faults were classified and defined, and the SDAE-MCSVM-FBN method was designed to solve multi-factor implicit cascade faults. A prototype system of data-driven digital twin model of the crane was constructed. Taking the train operation process in the workshop of a large state-owned enterprise steel plant as an example, compared with the manual spot check in the traditional operation and maintenance mode, before and after the application of the method in this paper, the proportion interval of time for fault maintenance and equipment downtime was 24.5 % 4 - 32.8 % and 20.5 % ~ 32.4 % respectively. The validity and feasibility of the proposed method for the diagnosis of hidden cascading faults were verified.