Method for evaluating health state of CRH380 type walking part system

Abstract

The invention discloses a method for evaluating the health state of a CRH380 type walking part system based on a confidence rule base, and particularly relates to the field of reliability of large complex systems. The evaluation method comprises the following steps: collecting measurement data of a CRH380 type walking part system in different health states as training data, giving out initial parameters of a library by a domain expert, and establishing a corresponding training model; quantitatively evaluating monitoring and environmental uncertainty existing in the process according to the measurement data; obtaining corrected attribute reliability based on an evidence discount method; quantifying cognitive uncertainty according to prior information, and constructing an expert reliability model; acquiring measurement data of the CRH380 type walking part system under the real-time working condition as test data; and fusing the quantified three-layer uncertainty factors into a confidence rule base, fitting the three-layer uncertainty factors with a real health state threshold of the system, and analyzing an evaluation result of test data. A mathematical model of a complex system does not need to be understood, and the health state of large complex equipment can be effectively evaluated.

Description

technical field [0001] The invention belongs to the field of reliability, and specifically relates to a health state method for a running part system of a high-speed train. Background technique [0002] Among the methods for assessing the health status of high-speed train running parts, it can be divided into: (i) data-driven methods; (ii) model-based methods; (iii) knowledge-based methods. A data-driven approach to understand the evolution of the health state from the time series of the observation system and provide informative results for the health state assessment of high-speed train running sections. However, the process of solving is not transparent, that is, there is no physical explanation for the changes of the system; it is necessary to know the appropriate failure threshold or health level, which may be difficult to achieve in some systems, especially when the understanding of the system mechanism is limited. Under these circumstances, the assessment of the heal...

AI Technical Summary

Problems solved by technology

[0004] (1) Monitoring uncertainty affected by the performance of the sensor itself

For example, the degradation of monitoring performance due to microstructural differences between homologous sensors, or due to factors such as unforeseen loads in the future;

[0005] (2) Environmental uncertainty affected by the operating environment and external conditions;

[0006] (3) Cognitive uncertainty caused by the evaluation process modeling of the running part of the high-speed train and the imperfection of parameter knowledge;

Images