Rail Network Monitoring with Distributed Acoustic Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for monitoring rail networks lack effective means to distinguish between train-related and environmental acoustic signals, making it difficult to accurately assess the condition of both the train and the rail infrastructure in real-time.
Innovation Solution
The method involves using distributed fibre optic sensing, specifically fibre optic distributed acoustic sensing (DAS), to identify a characteristic train signal by aligning measurement signals from multiple channels based on the train's speed, allowing for the separation of train-related signals from environmental signals, and then analyzing these signals to derive insights into the train's condition and the rail track's health.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed fibre optic sensing is used to monitor rail networks, then real-time monitoring capability and coverage scope are improved, but the ability to distinguish between train-related and environmental acoustic signals deteriorates
Solution Approach 1:
The patent divides the continuous acoustic signal into discrete events by identifying characteristic train signals separate from environmental background signals. This segmentation allows the system to process and analyze train-related acoustic responses independently, resolving the contradiction between comprehensive real-time monitoring and accurate signal discrimination.
Solution Approach 2:
The patent extracts characteristic train signals from the mixed acoustic environment by comparing acoustic responses across multiple sensors and time windows. This extraction process isolates the train-related signals from environmental noise, enabling precise measurement while maintaining real-time monitoring capability.
2Measurement precision
If multiple sensing channels are used to improve signal discrimination, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent combines data from multiple sensing channels through correlation analysis and time-windowed processing to improve signal discrimination. By merging information across channels in a coordinated manner, the system achieves high measurement precision while managing complexity through unified processing algorithms.
Solution Approach 2:
The patent creates a multi-functional processing system that handles both environmental monitoring and train detection using the same sensor infrastructure and processing algorithms. This universal approach allows multiple sensing channels to serve dual purposes, improving precision without proportionally increasing overall system complexity.
Data Source
AI summary
This application describes methods and apparatus for monitoring of rail networks using fibre optic distributed acoustic sensing (DAS), especially for condition monitoring. One method involves taking (902) a first data set corresponding to measurement signals from a plurality of channels of at least one fibre optic distributed acoustic sensor (100) having a sensing fibre (101) deployed to monitor at least part of the path of the rail network (201). The first data set corresponds to measurement signals acquired as a train (202) passed along a first monitored section of the rail network. The method involves identifying (903) a speed of the train through the first monitored section and dividing (904) the first data set into a plurality of time windows. Each time window contains a different subset of the first data set, with the measurement signal for each successive channel in a time window being delayed with respect to the previous channel by a time related to the identified train speed. For each time window, any appropriate time shift is identified (905) and applied (906) to the measurement signals for a channel so as to substantially align the measurement signals of the channels within the time window. The data from the time windows is then combined (907) after any time shifts have been applied to form an aligned first data set; and a characteristic train signal is derived (908) from the aligned first data set. The characteristic signal may be removed from the aligned first data set (1007) to leave remainder data. The characteristic trains signal and/or remainder data may be analysed for condition monitoring.