A fault identification method of a switch machine based on clustering integration

Abstract

The invention discloses a fault identification method of a switch machine based on clustering integration, which comprises the following steps: step 1, preprocessing power data of a switch machine: processing power data of the switch machine by using DTW dynamic time adjustment; step 2, generating a base clustering member by collocating a data feature set and a base clustering algorithm; obtaining3 base clustering members by adopting a hierarchical condensed hierarchical clustering algorithm based on hierarchical clustering and a density-based DBSCAN, DPCA as a base clustering algorithm; step3, constructing the data mining framework of switch machine fault based on a clustering ensemble algorithm. A W-VOTE clustering integration algorithm based on weighted votes is adopted to realize analysis, mining and fault diagnosis of switch machine fault data. Without prior knowledge, the method has the characteristic of more comprehensive fault diagnosis, higher fault identification accuracy and visual fault identification.

Description

technical field [0001] The invention belongs to the technical field of rail transit operation control, and relates to a switch machine fault identification method based on cluster integration. Background technique [0002] The switch machine is an important component of high-speed rail signal equipment, and the normal operation of the switch machine ensures the safety of high-speed rail. Electric switch machines are widely used in my country's high-speed railways, and the working place is located outdoors, which is a signal device with a high probability of failure. [0003] In the prior art, for the fault diagnosis and maintenance of the switch machine, the diagnostic measures used on site include: [0004] (1) Manual periodic inspection, relying on the work experience of on-site maintenance personnel to conduct regular inspections, has limited efficiency in judging and handling switch machine failures. The manual periodic inspection method has high labor intensity, low f...

AI Technical Summary

Problems solved by technology

[0004] (1) Manual periodic inspection, relying on the work experience of on-site maintenance personnel to conduct regular inspections, the efficiency of judging and handling switch machine failures is limited

The manual periodic inspection method is labor-intensive, the fault identification efficiency is low, and the fault processing delay is long, which cannot meet the high-speed development requirements of today's railways.

[0005] (2) Traditional fault diagnosis methods, such as model-based, state-based, expert system-based, Bayesian networks, support vector machines and other fault diagnosis methods, all of which require rich and complete prior knowledge to support

Traditional fault diagnosis methods largely rely on prior knowledge. Because the turnouts are widely distributed along the railway line, the external environment and working conditions vary greatly, resulting in various fault modes and causes, and it is difficult to diagnose all faults comprehensively.

[0006] (3) The existing monitoring system-based switch machine fault diagnosis method has a low degree of intelligence, misdiagnosis, missed diagnosis, low diagnostic efficiency, and requires the intervention of on-site personnel in practical applications. Experience has a greater impact on the diagnosis

Images