Railway Point Switch Fault Diagnosis Using Wavelet Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional condition monitoring systems for railway point switches are not reliable in detecting all fault types, particularly those exhibiting oscillations, and require manual operator assessment, which is inefficient and prone to false alarms.
Innovation Solution
A diagnostic system that extracts characteristic features from waveforms using wavelet approximation and applies them to supervised machine learning algorithms, specifically Random Forest, to automatically classify operating behavior and generate signals indicating normal operation or fault types, reducing the need for manual investigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional condition monitoring monitors average current drawn from an electric motor, then the system can detect some faults, but it cannot reliably detect all fault types particularly those exhibiting oscillations about a set value
Solution Approach 1:
The patent extracts characteristic features from the current waveform data, specifically focusing on oscillation patterns about the set value. By extracting these specific features rather than monitoring the entire waveform or only average values, the system can reliably detect oscillation-type faults while maintaining manageable complexity through targeted feature analysis
Solution Approach 2:
The patent transforms the one-dimensional average current monitoring into multi-dimensional waveform analysis by examining the shape, oscillations, and characteristic features of the current waveform over time. This dimensional expansion enables detection of fault patterns that were invisible to simple average current monitoring
2Measurement precision
If manual operator assessment is used to diagnose faults, then the system can investigate waveform details, but it requires operator knowledge and experience and is inefficient
Solution Approach 1:
The system performs self-diagnosis by automatically extracting characteristic features from waveforms and comparing them against known fault patterns. This eliminates the need for manual operator assessment while maintaining diagnostic precision, as the system serves itself by autonomously identifying and classifying fault conditions based on extracted waveform features
Solution Approach 2:
The patent replaces the mechanical process of manual waveform assessment by operators with an automated computational system that extracts and analyzes waveform features algorithmically. This substitution maintains the precision of detailed waveform analysis while dramatically improving diagnosis efficiency and eliminating dependency on operator expertise
3Ease of operation
If simple average current monitoring is used, then the system is easy to operate, but it generates false alarms and requires threshold value comparisons that miss oscillation faults
Solution Approach 1:
The patent changes the monitoring parameter from simple average current values to characteristic waveform features including oscillation patterns, shape characteristics, and temporal variations. This parameter transformation maintains ease of operation through automated feature extraction while significantly improving fault detection accuracy by capturing the nuanced patterns that indicate actual faults versus normal variations
Data Source
Figure 1
Figure 2a~2b
Figure 3~4
AI summary
According to embodiments of the present disclosure, there is provided a diagnostic system for diagnosing faults in a railway point switch. The diagnostic system comprises: an input arranged to receive a waveform associated with operating the railway point switch during an event; a feature extraction module arranged to extract characteristic features representing a shape of the waveform; and a classification module arranged to apply the extracted characteristic features to logic rules for classifying the features according to an operating behaviour of the railway point switch during the event, classify the operating behaviour of the railway point switch based on the application of the extracted characteristic features to the logic rules, and generate a signal indicating the classified operating behaviour.