A track circuit interference test analysis method, system and device

Abstract

The present application belongs to the field of rail transit signal technology, and specifically relates to a rail circuit interference test analysis method, system and device, aiming to solve the problems of difficult accurate positioning of mixed interference sources, inability to capture occasional faults and subjective diagnosis. The present application comprises: multi-channel synchronous acquisition and real-time storage of voltage signals and current signals of the rail circuit; real-time processing of the original signals, analysis of different frequency components through high-resolution spectrum analysis, comparison with a preset threshold to trigger an alarm and generate a real-time analysis report in combination with time domain analysis; intelligent analysis of historical data, verification and feature recognition of fault period data through establishment of a probability model of normal working conditions to generate a diagnostic analysis report. The present application can realize accurate separation and positioning of interference sources, automatically capture transient faults, and provide objective and intelligent fault diagnosis, significantly improving the efficiency and reliability of rail circuit fault troubleshooting.

Description

Technical Field

[0001] This invention belongs to the field of rail transit signaling technology, and specifically relates to a method, system and equipment for testing and analyzing track circuit interference. Background Technology

[0002] Track circuits are the foundation of railway signaling systems, and their reliability directly affects train operation safety and transportation efficiency. In modern railway operations, the operating environment of track circuits is increasingly complex, inevitably subject to various electromagnetic interferences, such as traction return current and its harmonics generated by electric locomotives, signal crosstalk between adjacent track circuits, and even radiated interference from other communication or power equipment along the line. These interference signals, superimposed on the track circuit's own operating signals (such as 25Hz, UM71, etc.), form complex composite signals, a major cause of abnormal track circuit operation and faults such as the appearance of "red light bands" (i.e., incorrect track occupancy indications).

[0003] To troubleshoot such faults, existing technologies primarily offer two types of testing methods: large automated testing benches in maintenance depots and portable testing instruments used by field maintenance personnel. However, both types of equipment have significant limitations in practical applications, directly leading to the following unresolved technical issues:

[0004] The root cause of the fault is difficult to pinpoint: Existing field testing instruments have limited capabilities in processing mixed signals. When multiple interferences coexist, they cannot accurately decompose the composite signal into individual frequency components. This means that even if maintenance personnel detect an anomaly, they cannot determine the root cause of the fault—whether it is excessive 50Hz power frequency interference, traction harmonics at a specific frequency, or interference from other unknown sources. Ultimately, they can only resort to a trial-and-error troubleshooting method, which is inefficient and cannot guarantee a complete solution.

[0005] Intermittent faults are difficult to detect: Many serious faults affecting vehicle operation are caused by transient, intermittent, and strong interference. Most existing testing instruments require personnel to be present during operation, making long-term continuous unattended monitoring impossible. This means that when a fault occurs, the testing instrument may not be connected to the system, thus missing the critical moment of the fault. This makes these "invisible" faults a persistent problem in maintenance, posing a significant threat to driving safety.

[0006] Diagnostic conclusions are subjective and experience-dependent: Due to a lack of in-depth intelligent analysis capabilities, existing equipment typically only provides raw waveforms or simple RMS data. Interpreting fault information from this data heavily relies on the individual experience and knowledge level of maintenance personnel. For inexperienced personnel, this data is insufficient to form clear diagnostic conclusions, leading to different judgments on the same problem by different people, resulting in a lack of objectivity and standardization in the fault handling process.

[0007] Therefore, there is an urgent need in this field for a new technical solution that can not only achieve long-term continuous monitoring of track circuits to capture all data, but also perform automatic and accurate analysis and intelligent diagnosis of the acquired complex mixed signals, thereby objectively and efficiently locating the root cause of the fault. Summary of the Invention

[0008] To address the aforementioned problems in the prior art, namely the difficulty in accurately locating mixed interference sources, the inability to capture intermittent faults, and the subjective nature of diagnosis, this invention provides a method, system, and equipment for testing and analyzing track circuit interference.

[0009] In a first aspect, this invention proposes a method for testing and analyzing track circuit interference, comprising:

[0010] The voltage and current signals of the track circuit are acquired synchronously through multiple channels to obtain the original time-domain waveform signals;

[0011] The original time-domain waveform signal is stored in real time to form historical data;

[0012] The original time-domain waveform signal is processed in real time. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze the different frequency components it contains, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time value and rate of change; comparing the analyzed frequency components, real-time value and rate of change with a preset alarm threshold, and triggering an alarm when the threshold is exceeded; capturing and saving waveform data before and after the alarm time and generating a real-time analysis report containing fault information.

[0013] The historical data is processed, including: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results and generating a diagnostic analysis report containing the fault causes.

[0014] Output the real-time analysis report or the diagnostic analysis report.

[0015] Furthermore, the multi-channel synchronous acquisition of voltage and current signals of the track circuit includes:

[0016] Voltage signals are acquired synchronously through multiple electrically isolated voltage acquisition channels, and current signals are acquired synchronously through multiple electrically isolated current acquisition channels.

[0017] The signals acquired from each acquisition channel are converted from analog to digital to generate digitized raw time-domain waveform signals, wherein the frequency range of the acquisition covers from the normal operating signal frequency of the track circuit to the frequency of potential interference signals.

[0018] Furthermore, the real-time storage of the original time-domain waveform signal includes:

[0019] A time identifier is added to the original time-domain waveform signal, and the signal is buffered in the form of data units with the time identifier.

[0020] The cached data units are continuously written to a non-volatile storage medium to form a continuous historical data sequence with a preset time length. When the storage capacity reaches the preset upper limit, the earliest stored historical data is overwritten according to the time order.

[0021] Furthermore, the step of performing frequency domain analysis on the original time-domain waveform signal to analyze its different frequency components includes:

[0022] Determine the specific frequency band to be analyzed and its center frequency; based on the center frequency and the original sampling frequency, digitally shift the original time-domain waveform signal to shift the spectrum of the specific frequency band to the baseband to obtain the frequency-shifted signal;

[0023] The frequency-shifted signal is subjected to an anti-aliasing low-pass filter, and the cutoff frequency of the low-pass filter is determined according to a predetermined frequency resolution amplification factor and the original sampling frequency.

[0024] The low-pass filtered signal is resampled at a reduced sampling rate, and the sampling interval of the resampling is equal to the frequency resolution amplification factor.

[0025] Perform a fast Fourier transform on the resampled signal and calculate its spectrum to obtain a spectrum with improved frequency resolution within the specific frequency band.

[0026] Based on the spectrum, signal components of different frequencies and their amplitude information are separated and analyzed from the original time-domain waveform signal.

[0027] Furthermore, the preset alarm thresholds include upper and lower amplitude thresholds set independently according to the type of different frequency signal components being parsed, a real-time signal value change rate threshold set based on historical data or theoretical values, and a frequency offset range threshold set for a specific frequency component.

[0028] The comparison includes: comparing the amplitude of the analyzed specific frequency component with its corresponding upper amplitude threshold and lower amplitude threshold, comparing the calculated real-time signal value change rate with the change rate threshold, and / or comparing the analyzed signal frequency with its corresponding frequency offset range threshold, wherein any comparison result exceeding the threshold will trigger an alarm.

[0029] Furthermore, the establishment of the probabilistic model for normal operating conditions includes:

[0030] Data from multiple periods during which no faults occurred were selected from the historical data as training data;

[0031] Extract time-domain and frequency-domain features from the composite signal corresponding to the training data;

[0032] The extracted features are mathematically transformed to conform to a Gaussian distribution.

[0033] Based on the transformed feature data, a Gaussian mixture model is used to model the probability distribution of composite signal features under normal operating conditions. The Gaussian mixture model is used to calculate the probability that the input features belong to normal operating conditions.

[0034] Furthermore, the step of inputting historical data of the fault occurrence time and the time period before and after it into the probability model for verification and probability calculation, and the step of identifying fault characteristics based on the calculation results, includes:

[0035] The historical data of the fault period is input into the Gaussian mixture model to calculate the probability value that it belongs to the normal operating condition;

[0036] When the probability value is lower than the preset fault determination threshold, the fault period data is determined to be abnormal.

[0037] Using the data from the fault period and the calculated probability values, the parameters of the Gaussian mixture model are adjusted through a parameter optimization algorithm to update the probability model;

[0038] Based on the updated probability model, the characteristic probability of the fault period data is recalculated, and at least one key feature that causes the probability value to decrease is identified. The key feature corresponds to a specific signal frequency component or time domain parameter.

[0039] Furthermore, the method also includes:

[0040] The system receives and responds to external control commands, including configuration commands for setting the signal type and range of each acquisition channel, control commands for starting or stopping the signal acquisition process, and status query commands for querying the current working status.

[0041] In a second aspect, the present invention provides a track circuit interference testing and analysis system, based on the track circuit interference testing and analysis method of the first aspect, the system comprising:

[0042] The data acquisition module is configured to synchronously acquire voltage and current signals from the track circuit through multiple channels to obtain the original time-domain waveform signal.

[0043] The data storage module is configured to store the original time-domain waveform signal in real time to form historical data;

[0044] The real-time analysis report generation module is configured to perform real-time processing on the original time-domain waveform signal. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze its different frequency components, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time signal value and rate of change; comparing the analyzed frequency components, real-time value, and rate of change with a preset alarm threshold, triggering an alarm when the threshold is exceeded, capturing and saving waveform data before and after the alarm time, and generating a real-time analysis report containing fault information.

[0045] The diagnostic analysis report generation module is configured to process the historical data, the processing including: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results and generating a diagnostic analysis report containing the fault cause.

[0046] The output module is configured to output the real-time analysis report or the diagnostic analysis report.

[0047] In a third aspect, the present invention provides an electronic device comprising:

[0048] At least one processor;

[0049] A memory communicatively connected to at least one of the processors; wherein,

[0050] The memory stores instructions that can be executed by the processor to implement a track circuit interference test and analysis method.

[0051] The beneficial effects of this invention are:

[0052] This invention employs high-resolution spectrum analysis algorithms such as ZOOM-FFT to clearly decompose the acquired composite signal into multiple independent frequency components, including track circuit operating signals, power frequency interference, harmonics, and unknown interference, and quantifies their amplitudes. This allows maintenance personnel to directly and objectively identify the root cause of the fault without guesswork, fundamentally eliminating trial-and-error troubleshooting methods and significantly improving troubleshooting efficiency.

[0053] The acquisition unit of this invention supports continuous recording of signal data over long periods, ensuring that any signal anomaly, regardless of its duration, can be completely recorded. This provides complete and reliable first-hand data for subsequent fault analysis and reproduction, filling a gap in the field of instantaneous fault monitoring and greatly enhancing the ability to detect highly concealed safety hazards.

[0054] This invention introduces an intelligent diagnostic model (such as a Gaussian Model) based on unsupervised learning. This model autonomously constructs a health status benchmark for track circuits by learning from massive amounts of normal data. When a fault occurs, the system can automatically analyze the deviation characteristics between the fault data and the model, thereby outputting objective, quantitative diagnostic conclusions and possible causes. This transforms the fault diagnosis process from subjective judgment to scientific analysis based on a data model, ensuring the consistency and reliability of diagnostic results, and is particularly helpful in improving the fault handling capabilities of inexperienced personnel. Attached Figure Description

[0055] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0056] Figure 1 This is a flowchart illustrating a track circuit interference testing and analysis method according to the present invention;

[0057] Figure 2 This is a data acquisition sub-system architecture diagram of a track circuit interference testing and analysis method according to the present invention. Detailed Implementation

[0058] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0059] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0060] The first embodiment of the present invention provides a method for testing and analyzing track circuit interference, comprising:

[0061] Step S10: Multi-channel synchronous acquisition of voltage and current signals of the track circuit to obtain the original time-domain waveform signal;

[0062] Step S20: The original time-domain waveform signal is stored in real time to form historical data;

[0063] Step S30: Perform real-time processing on the original time-domain waveform signal. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze the different frequency components it contains, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time value and rate of change; comparing the analyzed frequency components, real-time value and rate of change with a preset alarm threshold, and triggering an alarm when the threshold is exceeded; capturing and saving waveform data before and after the alarm time and generating a real-time analysis report containing fault information.

[0064] Step S40: Process the historical data. The processing includes: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results and generating a diagnostic analysis report containing the fault causes.

[0065] Step S50: Output the real-time analysis report or the diagnostic analysis report.

[0066] To more clearly illustrate the track circuit interference testing and analysis method of the present invention, the following is in conjunction with... Figure 1 The steps in the embodiments of the present invention are described in detail below, including steps S10-S40.

[0067] Step S10: Multi-channel synchronous acquisition of voltage and current signals of the track circuit to obtain the original time-domain waveform signal;

[0068] In this embodiment, the multi-channel synchronous acquisition of voltage and current signals of the track circuit includes:

[0069] Step S11: Voltage signals are synchronously acquired through multiple electrically isolated voltage acquisition channels, and current signals are synchronously acquired through multiple electrically isolated current acquisition channels.

[0070] Step S12: Perform analog-to-digital conversion on the signals obtained from each acquisition channel to generate digitized original time-domain waveform signals, wherein the frequency range of the acquisition covers from the normal operating signal frequency of the track circuit to the frequency of potential interference signals.

[0071] like Figure 2 As shown in this embodiment, the multi-channel synchronous acquisition is implemented through a dedicated acquisition unit. This acquisition unit adopts a modular hardware design, with its core being a data acquisition module containing eight independent analog-to-digital conversion channels and a digital signal processor. Specifically, channels 1 to 4 are designed as voltage acquisition channels, and channels 5 to 8 are designed as current acquisition channels. Each channel can acquire and process up to 16,834 data points per second and employs an electrical isolation design with an isolation strength of no less than 5000V to ensure that the signals of each channel are independent and safe.

[0072] The data acquisition unit integrates signal sensing, processing, storage, and communication. In the signal acquisition and conditioning section, the voltage channel uses a high-impedance differential input circuit, allowing direct connection to the track circuit voltage test point; the current channel is equipped with a through-type switching current transformer, supporting non-invasive measurement of rail return current or traction current. All channel signals are amplified and filtered before being sent to the core processing module for high-fidelity analog-to-digital conversion and real-time processing.

[0073] The data processing and storage section is centered around a digital signal processor, responsible for adding precise time stamps and encapsulating buffers to the acquired data. It also utilizes a built-in large-capacity non-volatile memory to achieve long-term continuous waveform recording, supporting cyclic overwrite storage. The unit also integrates Wi-Fi and USB communication interfaces, supporting wireless data transmission and remote command interaction. The power management module features a built-in rechargeable lithium battery, status indicators, and lightning protection grounding. Combined with an industrial-grade protective casing, it ensures stable and reliable operation of the equipment in the complex environment of railway sites.

[0074] To achieve safe and reliable field measurements, especially in complex electromagnetic environments and railway signaling sites where high potential differences may exist, all channels employ a rigorous electrical isolation design with an isolation strength of 5000V. This isolation effectively prevents signal crosstalk, measurement errors, and even equipment damage caused by common ground or potential differences between different measurement points, ensuring the safety and data independence of each channel and back-end processing unit when measuring high-current signals such as traction return current or voltage signals with common-mode voltage. The analog signals acquired by each acquisition channel are converted to digital signals by a high-performance analog-to-digital converter, generating high-fidelity digital raw time-domain waveform signals.

[0075] The acquired frequency band covers 0 Hz to 8000 Hz. This wideband design is specifically designed to cover the frequencies of normal operating signals from track circuits to all potential interference signals. Specifically, it can fully acquire the basic operating signals of 25 Hz phase-sensitive track circuits, 50 Hz power frequency traction current and its harmonic interference, as well as the center and sideband signals of widely used UM71, WG-21, and ZPW-2000 series track circuits, with standard carrier frequencies including 1700 Hz, 2000 Hz, 2300 Hz, and 2600 Hz. Furthermore, this range also reserves the capability to acquire other possible high-frequency interference. Current signal acquisition is typically achieved using a through-hole current transformer, which can be directly clipped onto the rail lead-in or return line without disconnecting the measured conductor, facilitating rapid on-site deployment and measurement. Through the above method, the acquisition unit can simultaneously and synchronously acquire multiple raw voltage and current waveforms from key monitoring points of the track circuit, providing a comprehensive, synchronous, and high-fidelity data foundation for subsequent integrated analysis of interference components and fault location. This invention overcomes the shortcomings of existing testing equipment that cannot simultaneously process multiple mixed signals or suffer from insufficient synchronization, laying a solid data acquisition foundation for accurate analysis of fault phenomena under the coupling of multiple interference sources.

[0076] Step S20 involves storing the original time-domain waveform signal in real time to form historical data, specifically including:

[0077] Step S21: Add a time identifier to the original time-domain waveform signal and buffer it in the form of a data unit with the time identifier;

[0078] Step S22 involves continuously writing the cached data units into a non-volatile storage medium to form a continuous historical data sequence with a preset time length. When the storage capacity reaches the preset upper limit, the earliest stored historical data is overwritten according to the time order.

[0079] In this embodiment, the real-time storage function is performed by the storage module within the acquisition unit. Specifically, after generating the original time-domain waveform signal, the digital signal processor immediately appends a high-precision timestamp to each data frame. This timestamp originates from the device's internal real-time clock, ensuring that all acquired data has an accurate timestamp. These timestamped data are organized into fixed data units and are first temporarily stored in the digital signal processor or a dedicated buffer. Subsequently, the storage control logic continuously and sequentially writes these buffered data units into a non-volatile storage medium, such as a high-capacity FLASH memory, thereby forming a continuous historical data sequence with a strict temporal relationship.

[0080] This storage module is designed to hold at least 72 hours of continuously collected data to meet the needs of long-term monitoring. When the storage medium's capacity reaches a preset upper limit, the system automatically activates a cyclic overwrite mechanism, that is, according to chronological order, the latest data overwrites the earliest stored historical data, thereby achieving long-term, uninterrupted automatic recording without manual intervention to replace the storage medium. This invention ensures that the device can operate continuously without human intervention, completely capturing all data within the critical time window before and after a fault occurs. It provides a complete and reliable data archive for in-depth fault backtracking and intelligent analysis based on historical data, effectively solving the problem that existing equipment cannot continuously record for long periods.

[0081] Step S30: Perform real-time processing on the original time-domain waveform signal. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze the different frequency components it contains, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time value and rate of change; comparing the analyzed frequency components, real-time value and rate of change with a preset alarm threshold, and triggering an alarm when the threshold is exceeded; capturing and saving waveform data before and after the alarm time and generating a real-time analysis report containing fault information.

[0082] The step of performing frequency domain analysis on the original time-domain waveform signal to analyze its different frequency components includes:

[0083] Step S31: Determine the specific frequency band to be analyzed and its center frequency; based on the center frequency and the original sampling frequency, digitally shift the original time-domain waveform signal to shift the spectrum of the specific frequency band to the baseband to obtain the frequency-shifted signal;

[0084] Step S32: Perform anti-aliasing low-pass filtering on the frequency-shifted signal. The cutoff frequency of the low-pass filter is determined according to the predetermined frequency resolution amplification factor and the original sampling frequency.

[0085] Step S33: The low-pass filtered signal is resampled at a reduced sampling rate, and the sampling interval of the resampling is equal to the frequency resolution magnification factor.

[0086] Step S34: Perform a fast Fourier transform on the resampled signal and calculate its spectrum to obtain a spectrum with improved frequency resolution within the specific frequency band.

[0087] Step S35: Based on the spectrum, separate and analyze the signal components of different frequencies and their amplitude information from the original time-domain waveform signal.

[0088] In this embodiment, the preset alarm thresholds include upper and lower amplitude thresholds set independently according to the type of different frequency signal components being parsed, a real-time signal value change rate threshold set based on historical data or theoretical values, and a frequency offset range threshold set for a specific frequency component.

[0089] The comparison includes: comparing the amplitude of the analyzed specific frequency component with its corresponding upper amplitude threshold and lower amplitude threshold, comparing the calculated real-time signal value change rate with the change rate threshold, and / or comparing the analyzed signal frequency with its corresponding frequency offset range threshold, wherein any comparison result exceeding the threshold will trigger an alarm.

[0090] In this embodiment, the real-time processing is performed by the analysis host. The analysis host receives the real-time data stream of the raw time-domain waveform signal from the acquisition unit via a wireless network or USB connection. The real-time processing executes two core tasks in parallel:

[0091] High-resolution frequency domain analysis and continuous time domain monitoring. In frequency domain analysis, to accurately separate different frequency components from mixed signals, such as the 25Hz track circuit signal, the 50Hz traction current, and the ZPW-2000 series signal, and to accurately obtain their amplitudes, the system employs the ZOOM-FFT algorithm. Specifically, firstly, based on the specific target frequency band to be analyzed, such as the need for detailed observation of 50Hz power frequency interference and its nearby harmonics, the center frequency of that frequency band is determined. Next, according to the formula... For the original time-domain waveform signal Digital frequency shifting is performed to shift the center frequency of the target frequency band to near zero frequency (baseband), thereby obtaining the frequency-shifted signal. ,in This is the original sampling frequency. To prevent spectral aliasing during subsequent downsampling, the frequency-shifted signal needs to be... Perform anti-aliasing low-pass filtering.

[0092] This device preferably uses an Infinite Impulse Response (IIR) filter to implement this low-pass filtering because it offers higher computational efficiency while achieving the same filtering performance, better meeting the requirements of real-time processing. It is important to emphasize that this low-pass filtering is a digital processing step within the ZOOM-FFT algorithm for the frequency-shifted signal. Its cutoff frequency is dynamically determined based on the target narrowband being analyzed, aiming to ensure the spectral quality after downsampling. This processing does not affect the system's front-end's ability to fully acquire the original wideband signal. The system can be configured by setting different center frequencies. High-resolution analysis was then performed on each operating frequency band of the track circuit (e.g., 1700Hz, 2000Hz, etc.). The specific implementation of this IIR filter is defined by the following difference equation: .in, This refers to the filter order or the length of the time-domain data involved. and The filter coefficients are pre-determined based on filter performance (such as cutoff frequency and ripple). This is the current output of the filter. For past output values, These are past input values. By recursively calculating this equation, low-pass filtering can be performed efficiently, with its cutoff frequency strictly set according to Shannon's theorem, not exceeding [a certain value]. Where D is the predetermined frequency resolution amplification factor, i.e., the downsampling factor. The filtered signal is then obtained. Then, the filtered signal... Resampling is performed according to the magnification factor D, that is, one sample is drawn from every D sampling points, thereby reducing the effective sampling rate to [missing value]. To facilitate the use of a standard number of points (e.g., N1 points) in subsequent Fast Fourier Transform (FFT) algorithms, it is usually necessary to pad the extracted data sequence with points, for example, by padding the end of the sequence with zeros, to restore or adjust the data length to an easily computed N1-point value. Then, the Fast Fourier Transform is performed on this padded N1-point sequence. The FFT calculation itself can be implemented using an optimized butterfly algorithm, and its calculation process can be described as follows: .in, and These represent the even and odd index positions of the padded sequence, respectively. and is the rotation factor.

[0093] Through this series of operations, the frequency resolution within the target frequency band can be increased from the original without increasing the number of FFT operation points. Significantly improved to This allows for extremely high spectral analysis accuracy within the narrow band of interest, enabling the accurate separation and quantization of amplitude and phase information of individual frequency components from composite signals based on this high-resolution spectrum. Simultaneously, in time-domain analysis, the system calculates the instantaneous and true RMS values ​​of each channel signal in real time, and calculates the rate of change of the signal based on the time series. The analysis host has a pre-configured library of flexibly configurable alarm thresholds. These thresholds include upper and lower amplitude limits independently set according to the type of different frequency signal components analyzed, a real-time rate of change threshold for the signal set based on historical normal data statistical analysis or theoretical values, and allowable frequency offset range thresholds set for specific frequency components such as track circuit operating signals.

[0094] The comparison process is parallel and comprehensive: the system continuously compares the amplitude of a specific frequency component (e.g., the 50Hz component) extracted from frequency domain analysis with its independently configured upper and lower amplitude limits; it compares the rate of change of the real-time signal value (e.g., the effective voltage value) calculated from time domain analysis with the corresponding rate of change threshold; and it also compares the offset of the extracted key signal frequency from its nominal frequency with the allowable frequency offset range threshold. If any of the above comparison results exceeds its corresponding preset threshold, the system determines that an anomaly has occurred and triggers an alarm. Once an alarm is triggered, the system automatically captures and permanently saves the complete original waveform data and related analysis results within a preset time window before and after the alarm time, and immediately generates a structured real-time analysis report. This report records in detail the alarm trigger time, channel, specific over-limit parameters (e.g., "50Hz current amplitude exceeds upper limit"), the over-limit value, and the associated waveform segment identifier, etc.

[0095] For example, at a railway site, the analysis host processes a real-time signal from channel 5 (current channel) of the acquisition unit. The frequency domain analysis module uses the ZOOM-FFT algorithm to perform high-resolution analysis within a narrow band around 50Hz, calculating in real time that the amplitude of the current component at 50Hz is 120A. The time domain analysis module simultaneously calculates that the effective value of the current in this channel changes at a rate of 15A per second. Meanwhile, the frequency analysis module measures the frequency of the current track circuit operating signal to be 1698.5Hz.

[0096] In the system's preset alarm threshold library, the upper limit of amplitude for the "50Hz traction current" component is set to 100A; the threshold for the rate of change of "current RMS value" is set to 10A per second; and the allowable frequency offset range for "ZPW-2000 track circuit signal" is set to nominal 1700Hz±1Hz.

[0097] During parallel comparison, the system found that: 1) the amplitude of the 50Hz current component (120A) was greater than the upper limit of 100A; 2) the effective rate of change of the current (15A / s) was greater than the threshold of 10A / s; 3) although the signal frequency of 1698.5Hz was still within the allowable frequency deviation range (1699Hz to 1701Hz), it did not exceed the limit. According to the rule that "an alarm is triggered if any comparison result exceeds the limit", since the first two conditions were met simultaneously, the system immediately determined that there was an anomaly.

[0098] The alarm was triggered at 14:30:05 on October 26, 2023. The system immediately and automatically captured and saved the complete raw waveform data of channel 5, as well as the real-time calculated spectrum and trend data, within the time window from 14:29:55 to 14:30:15 (a total of 20 seconds). Simultaneously, a real-time analysis report was generated, which structuredly recorded: the alarm trigger time was "2023-10-26 14:30:05"; the alarm channel was "CH5-Current"; the over-limit parameters and values ​​were listed in detail as "50Hz current component amplitude exceeds the upper limit, current value 120A, threshold 100A" and "current RMS value change rate exceeds the threshold, current value 15A / s, threshold 10A / s"; the report also associated the unique identifier of the saved waveform data segment.

[0099] This invention achieves accurate analysis of the frequency and amplitude of interference signals through the ZOOM-FFT algorithm. Combined with multi-dimensional and customizable threshold comparison, it significantly improves the accuracy and real-time performance of fault identification. It can promptly capture transient interference and slowly changing faults and automatically retain key evidence, providing a direct and reliable basis for on-site emergency response.

[0100] Step S40: Process the historical data. The processing includes: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results and generating a diagnostic analysis report containing the fault causes.

[0101] In this embodiment, a probability model for normal operating conditions is established, including:

[0102] Step S41: Select data from multiple periods when no faults occurred from the historical data as training data;

[0103] Step S42: Extract time-domain features and frequency-domain features from the composite signal corresponding to the training data;

[0104] Step S43: Perform mathematical transformation on the extracted features to make them conform to a Gaussian distribution;

[0105] Step S44: Based on the transformed feature data, the probability distribution of the composite signal features under normal operating conditions is modeled using a Gaussian mixture model, which is used to calculate the probability that the input features belong to the normal operating conditions.

[0106] In this embodiment, historical data of the fault occurrence time and the time period before and after it are input into the probability model for verification and probability calculation, and the identification of fault characteristics based on the calculation results includes:

[0107] Step S45: Input the historical data of the fault period into the Gaussian mixture model to calculate the probability value that it belongs to the normal operating condition;

[0108] Step S46: When the probability value is lower than a preset fault determination threshold, the fault period data is determined to be abnormal.

[0109] Step S47: Using the data from the fault period and the calculated probability values, adjust the parameters of the Gaussian mixture model using a parameter optimization algorithm to update the probability model;

[0110] Step S48: Based on the updated probability model, recalculate the feature probability of the fault period data and identify at least one key feature that causes the probability value to decrease, the key feature corresponding to a specific signal frequency component or time domain parameter.

[0111] In this embodiment, the intelligent analysis and processing of historical data is mainly accomplished by the data intelligent analysis and processing logic module in the analysis host. The system reads the historical data stored in the acquisition unit and, combined with the actual operation of the railway line, selects multiple time periods without faults from the massive data as training data. These data should represent the normal operation status of the line under various working conditions. An unsupervised learning fault detection algorithm is used to systematically extract features that characterize the signal state from the composite signals corresponding to these training data. These features include both time-domain features, such as the effective value, peak value, mean value, and trend of the signal within a certain time window, and frequency-domain features, which need to be obtained by performing high-resolution spectrum analysis on the signal using the aforementioned ZOOM-FFT algorithm. For example, extracting the amplitude, specific frequency band energy, or harmonic component ratio of specific frequency components such as 50Hz traction current, 25Hz track circuit signal, and ZPW-2000 series (approximately 1700Hz) signal.

[0112] To adapt to the subsequent probability model, these extracted features need to undergo necessary mathematical transformations. Specifically, features that conform to a Gaussian distribution are selected whenever possible, while non-Gaussian features are transformed using linear or nonlinear methods to make their distribution shape closer to a Gaussian distribution. Then, based on these transformed feature data, a Gaussian mixture model is used to model the probability distribution of the composite signal feature vector under normal operating conditions. The model parameters are estimated using algorithms such as expectation-maximization, ultimately resulting in a model capable of calculating the probability that any input feature vector belongs to the distribution under that normal operating condition.

[0113] After establishing the probabilistic model, historical data from the fault occurrence time and the time periods before and after it are input into the model for verification and probability calculation, and fault characteristics are identified based on the calculation results. Specific steps include: First, based on the fault record or alarm time point, a data analysis period is defined, and various collected data from the time period before and after the fault point are selected. This data is then input into the established Gaussian mixture model for calculation to obtain the probability value of its belonging to normal operating conditions. When this probability value is lower than a preset fault judgment threshold, the system determines that the data for that fault period is abnormal. To continuously improve diagnostic accuracy, the system uses the data from this fault period to adjust the parameters of the Gaussian mixture model through an optimization algorithm, thereby updating the probabilistic model.

[0114] Based on the updated model, the feature probabilities of the fault period data are recalculated, and a deeper analysis is conducted to determine which specific features significantly deviated, leading to a decrease in the overall probability value. These identified key features correspond to specific signal frequency components (e.g., an abnormally increased 50Hz current component) or time-domain parameters (e.g., a sudden drop in the effective value of the track circuit voltage). Based on this analysis, the system automatically generates a structured diagnostic analysis report, which details the analyzed causes of the fault, such as "the fault was mainly caused by an abnormal increase in the 50Hz traction current to 120A, accompanied by a 15% decrease in the amplitude of the 1700Hz track circuit signal." This invention achieves intelligent and automated root cause analysis of historical fault data, improving the accuracy and efficiency of fault diagnosis.

[0115] Step S50: Output the real-time analysis report or the diagnostic analysis report.

[0116] In this embodiment, outputting the analysis report is the final step in the system's information presentation and decision support. Whether it is a real-time analysis report generated by real-time processing or a diagnostic analysis report generated after in-depth mining of historical data, its output is achieved through the human-computer interaction interface module of the analysis host and related data export functions.

[0117] Once an alarm is triggered and a real-time analysis report is generated during real-time processing, the report will be dynamically displayed immediately in a prominent area of ​​the main window of the human-machine interface of the analysis host, such as a dedicated alarm information bar or a pop-up window. The report clearly lists key information in a structured text format, typically including the timestamp of the alarm trigger, accurate to the second; the specific channel number where the alarm occurred, such as CH5; the signal type or frequency component determined to be abnormal, such as the 50Hz traction current component; the specific parameters exceeding the limit, such as the amplitude and its value, such as the current value of 120A; the preset threshold violated, such as the threshold of 100A; and the index or storage location of the associated original waveform data segment.

[0118] The interface will synchronously highlight or automatically switch to waveform and spectrum graphs near the alarm time point, allowing maintenance personnel to intuitively see the specific morphological changes before and after the signal anomaly occurred. Furthermore, this real-time analysis report will be automatically saved as a separate data file in the storage directory specified by the analysis host. Its format can be JSON, XML, or a specific log file format, and the filename typically includes information such as time and channel for traceability. In some implementations, the system can also be configured to send alarm information and report summaries to a remote server or mobile terminal in real time via a network interface, enabling remote monitoring.

[0119] For diagnostic analysis reports, the output is typically completed after the user initiates an intelligent analysis command for a specific historical fault period. After the analysis host completes fault feature identification and root cause analysis based on a Gaussian mixture model, it generates a more detailed report view in the interface.

[0120] This report not only includes basic information such as the time, location, and affected channels of the fault, but more importantly, it includes a section analyzing the cause of the fault. This section presents analytical conclusions with clear textual descriptions and key data. For example, the analysis determined that the main cause of this fault was 50Hz traction current interference, whose amplitude remained consistently higher than the normal range, reaching a maximum of 125A during the fault period; simultaneously, the amplitude of the 1700Hz track circuit signal was suppressed and decreased by approximately 20%. The report also embeds key evidence used to reach this conclusion, such as a comparison curve of the 50Hz current amplitude during the fault period and the normal period, and a comparison graph of the 1700Hz signal spectrum, in the form of charts or provides convenient links. The complete diagnostic analysis report can be exported and saved in standard document formats, such as PDF or Word, for easy inclusion in formal maintenance records or technical archives. The system can also update the knowledge base or adjust the threshold parameters of real-time alarms based on repeatedly verified diagnostic rules or characteristic patterns, enabling the system to learn and optimize itself.

[0121] This invention transforms complex signal processing and data analysis results into standardized, structured information that maintenance personnel can directly understand and use for decision-making and to guide maintenance actions. This significantly shortens the time from anomaly detection to fault location, improves the targeting and efficiency of electrical maintenance operations, and also accumulates valuable standardized data for fault statistics, pattern research, and preventative maintenance.

[0122] In this embodiment, the method further includes:

[0123] The system receives and responds to external control commands, including configuration commands for setting the signal type and range of each acquisition channel, control commands for starting or stopping the signal acquisition process, and status query commands for querying the current working status.

[0124] In this embodiment, the method, during operation, possesses the ability to interact with and be controlled by the operator. This is achieved by receiving and responding to a series of external control commands. These commands are primarily generated by the analysis host and distributed to the data acquisition unit via wireless WiFi or wired USB connection.

[0125] Specifically, configuration commands are used to remotely set the hardware parameters of the data acquisition unit. Operators can independently select the signal type for each of the eight acquisition channels on the human-machine interface of the analysis host, specifying whether the channel is used to measure voltage or current signals. Simultaneously, appropriate ranges can be set for each channel, such as setting a voltage channel to a 0-10V range or a current channel to a 0-5A range, to ensure the accuracy and safety of signal acquisition. After receiving these configuration commands through the transmission module, the digital signal processor of the acquisition unit parses the command content and configures the corresponding analog-to-digital converter parameters and signal conditioning circuits for the appropriate channels. Upon completion, it returns configuration confirmation information to the analysis host.

[0126] Control commands are used to directly manage the lifecycle of the signal acquisition process. The most important control commands include "Start Acquisition" and "Stop Acquisition." When the analysis host issues a "Start Acquisition" command, the digital signal processor of the acquisition unit initiates the synchronous data acquisition and storage process for all configured channels. Correspondingly, the "Stop Acquisition" command causes the acquisition unit to sequentially terminate data acquisition and enter standby mode. This enables precise remote control of the testing process without requiring manual operation by personnel at the equipment.

[0127] Status query commands are used to obtain real-time operating status information from the data acquisition unit. These commands can include a "time query" command to query the data acquisition unit's internal real-time clock, a "remaining battery query" command to assess remaining battery power, and a "maximum test duration query" command to calculate how much more testing time the current storage space can support. Upon receiving a query command, the data acquisition unit reads the current status data from the corresponding sensors or storage management system and feeds it back to the analysis host via the communication link, clearly displaying it to the operator on the human-machine interface.

[0128] This invention empowers maintenance personnel with flexible, remote control capabilities over testing equipment. Users can adjust measurement parameters, start and stop data acquisition tasks, and monitor equipment status at any time according to on-site testing needs, thereby significantly improving the efficiency and adaptability of test deployment. It truly realizes intelligent and convenient management of on-site testing, overcoming the limitations of existing equipment that requires continuous on-site operation and configuration.

[0129] Although the steps in the above embodiments are described in the above order, those skilled in the art will understand that in order to achieve the effect of this embodiment, different steps do not need to be executed in such order. They can be executed simultaneously (in parallel) or in reverse order. These simple variations are all within the protection scope of this invention.

[0130] A track circuit interference testing and analysis system according to a second embodiment of the present invention, based on a track circuit interference testing and analysis method, the system includes:

[0131] The data acquisition module is configured to synchronously acquire voltage and current signals from the track circuit through multiple channels to obtain the original time-domain waveform signal.

[0132] The data storage module is configured to store the original time-domain waveform signal in real time to form historical data;

[0133] The real-time analysis report generation module is configured to perform real-time processing on the original time-domain waveform signal. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze its different frequency components, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time signal value and rate of change; comparing the analyzed frequency components, real-time value, and rate of change with a preset alarm threshold, triggering an alarm when the threshold is exceeded, capturing and saving waveform data before and after the alarm time, and generating a real-time analysis report containing fault information.

[0134] The diagnostic analysis report generation module is configured to process the historical data, the processing including: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results and generating a diagnostic analysis report containing the fault cause.

[0135] The output module is configured to output the real-time analysis report or the diagnostic analysis report. Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related explanations of the method described above can be found in the corresponding processes of the foregoing system embodiments, and will not be repeated here.

[0136] It should be noted that the track circuit interference testing and analysis system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.

[0137] A device according to a third embodiment of the present invention includes:

[0138] At least one processor;

[0139] and a memory communicatively connected to at least one of the processors;

[0140] The memory stores instructions that can be executed by the processor to implement the above-described method for testing and analyzing track circuit interference.

[0141] A fourth embodiment of the present invention provides a computer-readable storage medium storing computer instructions, which are executed by the computer to implement the above-described method for testing and analyzing track circuit interference.

[0142] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the storage device and processing device described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0143] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0144] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0145] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for testing and analyzing interference in track circuits, characterized in that, include: The voltage and current signals of the track circuit are acquired synchronously through multiple channels to obtain the original time-domain waveform signals; The original time-domain waveform signal is stored in real time to form historical data; The original time-domain waveform signal is processed in real time. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze the different frequency components it contains, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time value and rate of change; comparing the analyzed frequency components, real-time value and rate of change with a preset alarm threshold, and triggering an alarm when the threshold is exceeded; capturing and saving waveform data before and after the alarm time and generating a real-time analysis report containing fault information. The historical data is processed, including: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results and generating a diagnostic analysis report containing the fault causes. Output the real-time analysis report or the diagnostic analysis report; The establishment of the probabilistic model for normal operating conditions includes: Data from multiple periods during which no faults occurred were selected from the historical data as training data; Extract time-domain and frequency-domain features from the composite signal corresponding to the training data; The extracted features are mathematically transformed to conform to a Gaussian distribution. Based on the transformed feature data, the probability distribution of composite signal features under normal operating conditions is modeled using a Gaussian mixture model, which is used to calculate the probability that the input features belong to normal operating conditions. The step of inputting historical data of the fault occurrence time and the time period before and after it into the probability model for verification and probability calculation, and the step of identifying fault characteristics based on the calculation results, includes: The historical data of the fault period is input into the Gaussian mixture model to calculate the probability value that it belongs to the normal operating condition; When the probability value is lower than the preset fault determination threshold, the data during the fault period is determined to be abnormal. Using the data from the fault period and the calculated probability values, the parameters of the Gaussian mixture model are adjusted through a parameter optimization algorithm to update the probability model; Based on the updated probability model, the characteristic probability of the data during the fault period is recalculated, and at least one key feature that causes the probability value to decrease is identified. The key feature corresponds to a specific signal frequency component or time-domain parameter.

2. The method according to claim 1, characterized in that, The multi-channel synchronous acquisition of voltage and current signals from the track circuit includes: Voltage signals are acquired synchronously through multiple electrically isolated voltage acquisition channels, and current signals are acquired synchronously through multiple electrically isolated current acquisition channels. The signals acquired from each acquisition channel are converted from analog to digital to generate digitized raw time-domain waveform signals. The acquired frequency band covers the range from the normal operating signal frequency of the track circuit to the frequency of potential interference signals.

3. The method according to claim 1, characterized in that, The original time-domain waveform signal is stored in real time, including: A time identifier is added to the original time-domain waveform signal, and the signal is buffered in the form of data units with the time identifier. The cached data units are continuously written to a non-volatile storage medium to form a continuous historical data sequence with a preset time length. When the storage capacity reaches the preset upper limit, the earliest stored historical data is overwritten according to the time order.

4. The method according to claim 1, characterized in that, Frequency domain analysis is performed on the original time-domain waveform signal to analyze its different frequency components, including: Determine the specific frequency band to be analyzed and its center frequency; based on the center frequency and the original sampling frequency, digitally shift the original time-domain waveform signal to shift the spectrum of the specific frequency band to the baseband to obtain the frequency-shifted signal; The frequency-shifted signal is subjected to an anti-aliasing low-pass filter, and the cutoff frequency of the low-pass filter is determined according to a predetermined frequency resolution amplification factor and the original sampling frequency. The low-pass filtered signal is resampled at a reduced sampling rate, and the sampling interval of the resampling is equal to the frequency resolution amplification factor. Perform a fast Fourier transform on the resampled signal and calculate its spectrum to obtain a spectrum with improved frequency resolution within the specific frequency band. Based on the spectrum, signal components of different frequencies and their amplitude information are separated and analyzed from the original time-domain waveform signal.

5. The method according to claim 1, characterized in that, The preset alarm thresholds include upper and lower amplitude thresholds set independently according to the type of different frequency signal components being analyzed, a real-time signal value change rate threshold set based on historical data or theoretical values, and a frequency offset range threshold set for a specific frequency component. The comparison includes: comparing the amplitude of the analyzed specific frequency component with its corresponding upper amplitude threshold and lower amplitude threshold, comparing the calculated real-time signal value change rate with the change rate threshold, and / or comparing the analyzed signal frequency with its corresponding frequency offset range threshold, wherein any comparison result exceeding the threshold will trigger an alarm.

6. The method according to claim 1, characterized in that, The method further includes: The system receives and responds to external control commands, including configuration commands for setting the signal type and range of each acquisition channel, control commands for starting or stopping the signal acquisition process, and status query commands for querying the current working status.

7. A track circuit interference testing and analysis system, based on the track circuit interference testing and analysis method according to any one of claims 1-6, characterized in that, The system includes: The data acquisition module is configured to synchronously acquire voltage and current signals from the track circuit through multiple channels to obtain the original time-domain waveform signal. The data storage module is configured to store the original time-domain waveform signal in real time to form historical data; The real-time analysis report generation module is configured to perform real-time processing on the original time-domain waveform signal. The real-time processing includes: performing frequency domain analysis on the original time-domain waveform signal to analyze its different frequency components, and performing time domain analysis on the original time-domain waveform signal to calculate its real-time signal value and rate of change; comparing the analyzed frequency components, real-time value, and rate of change with a preset alarm threshold, triggering an alarm when the threshold is exceeded, capturing and saving waveform data before and after the alarm time, and generating a real-time analysis report containing fault information. The diagnostic analysis report generation module is configured to process the historical data, the processing including: reading the historical data and performing time-domain and frequency-domain analysis on the composite signal under normal operating conditions to establish a probability model of normal operating conditions; inputting historical data of the fault occurrence time and the time periods before and after it into the probability model for verification and probability calculation; identifying fault characteristics based on the calculation results, and generating a diagnostic analysis report containing the fault cause. The output module is configured to output the real-time analysis report or the diagnostic analysis report.

8. An electronic device, characterized in that, include: At least one processor; A memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by the processor to implement the track circuit interference test and analysis method according to any one of claims 1-6.

Images