Railway Switch Maintenance Signal via Compression Device

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Complex railway switches require frequent and resource-intensive maintenance, particularly lubrication, due to environmental factors like weather, leading to high costs and inefficiencies in both reactive and preventive maintenance approaches.

Innovation Solution

A method using a compression/decompression device, trained on input vectors from railway switch data, to generate a maintenance signal based on deviation values, predicting when lubrication is necessary, thereby reducing unnecessary maintenance interventions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If regular preventive maintenance is performed on railway switches, then total infrastructure failures are prevented, but personnel and financial expenditures increase and resources are tied up

Engineering Contradiction:
Improveprevention of total infrastructure failureVSAvoidresource efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent replaces traditional mechanical preventive maintenance schedules with an automated monitoring system using sensors and machine learning algorithms. The system continuously collects operational data from the railway switch and uses a trained compression/decompression device to predict maintenance needs, substituting regular manual interventions with intelligent automated prediction.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The monitoring system enables the railway switch to essentially monitor itself and predict its own maintenance needs. The automated system collects data from the switch's own operational parameters and generates maintenance signals without requiring external preventive maintenance schedules, allowing the system to self-diagnose when maintenance is actually needed.

Inventive Principle:
Principle #25Self-service

2Productivity

If reactive (corrective) maintenance is performed on railway switches, then maintenance costs are reduced, but high costs result from total infrastructure failure

Engineering Contradiction:
Improvemaintenance cost reductionVSAvoidinfrastructure availability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system performs preliminary maintenance actions by predicting future maintenance needs before actual degradation occurs. The trained compression/decompression device analyzes operational patterns and generates maintenance signals in advance, allowing maintenance to be scheduled proactively rather than reactively, preventing infrastructure failure while optimizing maintenance timing.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements continuous feedback by monitoring operational parameters, comparing them against learned patterns from training data, and generating maintenance signals when deviations indicate upcoming maintenance needs. This closed-loop feedback mechanism enables the system to adapt to actual switch behavior and predict maintenance requirements accurately.

Inventive Principle:
Principle #23Feedback

3Reliability

If frequent lubrication is performed on railway switches, then proper function under environmental variables is ensured, but resource usage and costs increase

Engineering Contradiction:
Improveproper function under weather conditionsVSAvoidlubrication material usage
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The system transitions from static, schedule-based lubrication to dynamic, condition-based lubrication. The monitoring system continuously adapts its predictions based on real-time operational data and environmental conditions, adjusting maintenance timing to match actual lubrication needs rather than following fixed intervals, thereby reducing unnecessary lubrication applications.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the parameter for determining lubrication timing from fixed time intervals to dynamic predictions based on operational parameters and environmental conditions. The trained model analyzes multiple input parameters including switch operations, weather conditions, and performance metrics to dynamically determine when lubrication is actually required.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4431363A1Method for generating a maintenance signal for the maintenance of a railway switch and monitoring device for carrying out such a method
Publication Date: 2024.09.18 SIEMENS MOBILITY GMBH
  • EP4431363A1 patent drawingFigure 1
  • EP4431363A1 patent drawingFigure 2
  • EP4431363A1 patent drawingFigure 3

AI summary

The invention relates, inter alia, to a method for generating a maintenance signal (WS) that describes a maintenance recommendation for the maintenance of a railway switch (10). According to the invention, it is provided that input measured values ​​(I1-In, M1-Mm; ) recorded for at least one observation period (ZR) in which at least one switch cycle (WUL) of the railway switch (10) has taken place are used to generate the maintenance signal.EM) an input vector (Ve) is formed, the input vector (Ve) is fed into a compression/decompression device (22), the compression/decompression device (22) outputs an output vector (Va) at its output for the input vector (Ve), wherein the compression/decompression device (22) has been trained in a training procedure using training input vectors (Vet) created for one or more railway switches without maintenance requirements, such that the deviation between the training input vectors (Vet) and the training output vectors (Vat) formed with them by the compression/decompression device (22) is minimal or falls below a predetermined deviation limit, the deviation between the output vector (Va) and its corresponding input vector (Ve) is determined by forming a deviation value (A), and the maintenance signal (WS) is generated on the basis of the deviation value (A).