A working condition recognition method based on the similarity feature of the action curve of the switch machine

Abstract

The invention relates to a working condition recognition method based on the similarity feature of the action curve of a switch machine, comprising the following steps: Step 1: Select a reference template from the historical action curve of the switch machine; Step 2: Construct a pairing matrix; Step 3: Calculate P n,m The distance d of each pair of curves in i,j , to construct the action curve distance matrix D n,m ; Step 4: through the dimensionality reduction algorithm for D n,m Carry out dimensionality reduction; step 5: draw the relative curve shape distribution diagram of the historical action curve C of the switch machine; step 6: use the clustering algorithm to analyze F n,2 :[f 1 ,f 2 ,..., f n ] for clustering; step 7: adjust the clustering parameters, and repeat step 6 until S c The shape of the inner action curve is the same; Step 8: Mark S with c representing the type of working condition c The curve inside is used to complete the identification of the working condition of the switch machine. Compared with the prior art, the present invention has the advantages of effectively solving the difficulty that the action curve of the switch machine cannot be further analyzed through machine learning and big data due to the lack of data labels, wide application, visualization of working condition change trends, and high efficiency.

Description

technical field [0001] The invention relates to the technical field of operation and maintenance of signal systems of urban rail transit and high-speed railway train operation control systems, in particular to a working condition identification method based on similarity characteristics of switch machine action curves. Background technique [0002] With the continuous extension of my country's high-speed railway and urban rail transit network, more and more railway communication and signal system equipment are deployed along the track to ensure the safety, reliability and efficiency of train operation. In this process, due to the technical upgrading of communication and signal system equipment, its work efficiency has been continuously improved, and its structure has become more sophisticated, which has also caused heavy pressure on logistics operation and maintenance. In order to further ensure driving safety and improve operation and maintenance efficiency, people urgently...

AI Technical Summary

Problems solved by technology

[0005] In view of the problems of different types of switch machine equipment in high-speed railways and urban rail transit, different suppliers, different service times, and dynamic errors in data monitoring circuits, the historical action curves of massive switch machines have complicated specifications and cannot be directly marked without professional manual marking. Applied to machine learning and big data analysis methods

Images