Power conductor breakage fault detection apparatus and method of use
By employing multimodal sensing and environmental noise modeling techniques, the risk of power line breakage is identified and assessed, solving the problems of high false alarm rate and long response time in railway environments, and achieving efficient capture and rapid response to initial breakage signals.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SICHUAN TIANFU JIANGDONG TECH CO LTD
- Filing Date
- 2026-03-03
- Publication Date
- 2026-06-09
AI Technical Summary
Existing power conductor breakage fault detection equipment is susceptible to interference from complex electromagnetic environments and mechanical vibrations in railway environments, leading to false alarms or missed alarms. It is difficult to capture weak initial signals from the conductor, and the fixed high detection threshold prolongs the fault response time.
Multimodal sensing modules are used to collect electromagnetic field changes, vibration and acoustic data in a distributed manner. Combined with an environmental noise modeling module, a dynamic model is constructed. Data fusion and preprocessing are used to identify the initial wire breakage characteristics. An adaptive optimization module is used to assess the wire breakage risk and carry out graded treatment.
It significantly reduces the false alarm rate, improves the ability to capture initial line failure signals, shortens fault response time, and enhances the safety and reliability of railway power supply systems.
Smart Images
Figure CN121784615B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the technical field, specifically to a device for detecting broken power lines and a method for using it. Background Technology
[0002] Power line breakage fault detection equipment originated from the urgent need for stable power system operation and personal safety. Early methods relied on manual inspections, which were inefficient and risky. With the expansion of the power grid, the demand for rapid, accurate, and non-contact detection has become increasingly prominent. Its technological development has progressed from traditional electrical parameter monitoring to acoustic and fiber optic sensing, and even to intelligent and automated detection combining drones, infrared, machine vision, and artificial intelligence. It is now widely used in transmission and distribution networks, urban power grids, industrial parks, overhead lines in remote areas, and railway power supply systems. Key applications include severe weather warnings, inspection of aging lines, location of external damage faults, and real-time monitoring and fault self-healing in smart grids, significantly improving the reliability of power supply and maintenance efficiency.
[0003] In railway power supply systems, power conductor breakage detection equipment has the following technical drawbacks in practical applications:
[0004] Susceptible to interference from complex electromagnetic environments and mechanical vibrations, leading to false alarms or missed detections; the complex railway operating environment, with its strong electromagnetic fields generated by high-speed trains, rapid changes in traction current, and severe mechanical vibrations caused by passing trains, can significantly interfere with sensitive sensors based on electrical parameters, acoustics, or optics. This makes it difficult for the equipment to distinguish between real wire breakage signals and environmental noise, resulting in frequent false alarms or missed detections of some weak, transient wire breakage signals;
[0005] High detection thresholds or multiple confirmation mechanisms set to suppress environmental interference may reduce the ability to capture weak signals from incomplete wire breaks or the initial stages of a break, and prolong fault response time. To reduce the aforementioned frequent false alarms, system designers often increase the detection threshold or introduce complex signal processing algorithms and multi-sensor data fusion confirmation mechanisms. While this effectively filters out most environmental noise, a side effect is that it may sacrifice sensitivity to weak characteristic signals generated in the initial stages of a break (e.g., cracks appearing before the conductor is completely separated), incomplete wire breaks (e.g., poor high-impedance contact), or the instant of a break. This leads to a prolonged detection response time for actual wire break faults, especially in high-speed rail scenarios, where even a delay of a few seconds can have serious consequences, and it is difficult to provide early warning of potential safety hazards (such as an imminent conductor break).
[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention
[0007] The purpose of this invention is to provide a power conductor breakage fault detection device and a method for using it, so as to solve the problems mentioned in the background art.
[0008] To achieve the above objectives, the present invention provides the following technical solution:
[0009] A power conductor breakage fault detection device and its usage method, comprising:
[0010] Multimodal sensing modules are used to collect various physical quantities along the railway power supply contact network or catenary in a distributed manner, including electromagnetic field change data, high-frequency vibration data, acoustic characteristic data, and conductor micro-strain data.
[0011] The environmental noise modeling module is used to build a dynamic environmental noise model and analyze the background electromagnetic interference, mechanical vibration noise and other environmental noise data generated during real-time railway operation to obtain the background interference intensity factor.
[0012] The data fusion and preprocessing module is used to preprocess and fuse various physical quantity information with dynamic environmental noise models and transmit the data to the cloud platform; at the same time, it uses dimensionless processing calculations to perform data standardization and spatiotemporal synchronization calibration.
[0013] The wire breakage feature recognition module is used to identify and extract the initial weak feature signals and complete wire breakage feature signals of power conductor wire breakage faults based on standardized physical quantity information and dynamic environmental noise model, so as to construct a fault feature set and calculate and obtain the fault confidence index.
[0014] The fault location and early warning module is used to comprehensively judge the fault type, location and urgency based on the fault confidence index and the analysis results of the environmental noise modeling module, and issue early warning instructions of the corresponding level.
[0015] The system adaptive optimization module obtains the line break risk assessment index by correlating the fault confidence index with the background interference intensity factor, and pre-sets the assessment threshold range. It then compares and analyzes the line break risk assessment index with the assessment threshold to comprehensively assess the line break risk level of the current railway power supply system and makes corresponding handling suggestions based on the risk level.
[0016] Preferably, the multimodal sensing module includes a distributed sensing unit and a high-precision acquisition unit;
[0017] The distributed sensing unit is used along the railway power supply contact network, based on the uniformly distributed induction coils, fiber optic vibration sensors, micro acoustic sensor arrays and local impedance sensors on the catenary to obtain high-resolution and real-time multimodal data.
[0018] The high-precision acquisition unit is used to perform high-precision synchronous sampling and preliminary digital processing on the raw data collected by the distributed sensing unit, and is equipped with anti-strong electromagnetic interference and anti-vibration design.
[0019] Preferably, the environmental noise modeling module includes a noise feature acquisition unit and a dynamic modeling unit;
[0020] The noise feature acquisition unit is used to monitor and record background noise data in the railway operating environment in real time. The background noise data includes the traction current fluctuation characteristics when the train passes through each monitoring period, the acoustic spectrum generated by pantograph-catenary friction, the line vibration amplitude, and the environmental electromagnetic field intensity. Based on the background noise data, and combined with a statistical averaging algorithm, the average intensity of each noise type within the monitoring period is calculated.
[0021] The dynamic modeling unit is used to combine the average intensity of each noise type acquired by the noise feature acquisition unit with historical data and machine learning algorithms to construct a real-time updated dynamic environmental noise model, which is used to predict and identify the noise level and characteristics of the current environment in order to construct a background interference intensity factor.
[0022] Preferably, the broken wire feature recognition module includes an initial feature analysis unit and a complete feature recognition unit;
[0023] The initial characteristic analysis unit is used to calculate the initial wire breakage characteristic intensity index based on the fused data by analyzing signals including local micro-strain anomalies, high-frequency vibration mode changes, weak transient electromagnetic field disturbances, and subtle local impedance fluctuations. The initial wire breakage characteristic intensity index is obtained in the following way:
[0024] The initial line breakage characteristic intensity index is obtained by calculating the difference between the square of the conductor microstrain change rate during the monitoring period and the square of the mean conductor microstrain change rate during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration spectrum entropy during the monitoring period and the square of the mean high-frequency vibration spectrum entropy during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added.
[0025] Preferably, the complete feature recognition unit is used to pre-set a disconnection threshold and compare the disconnection threshold with the initial disconnection feature intensity index to preliminarily determine whether there is an initial disconnection risk in the current power conductor. The specific details are as follows:
[0026] If the initial disconnection characteristic intensity index exceeds the disconnection threshold, it is preliminarily determined that there is an initial disconnection risk in the current power conductor, indicating that there is a potential fault in the current line. At this time, an initial warning command will be issued.
[0027] If the initial line breakage characteristic intensity index does not exceed the line breakage threshold, it is preliminarily determined that there is no initial line breakage risk in the current power conductor, indicating that the current line is in good operating condition, and there is no need to issue additional initial warning instructions.
[0028] Preferably, the fault location and early warning module is used to monitor and record real-time data of a local area of the contact network in a timely manner after receiving the initial early warning command issued by the complete feature recognition unit. This includes electromagnetic field strength, vibration frequency, acoustic anomalies and conductor tension. The module also performs self-checks on the status of each sensor to determine whether false alarms are caused by sensor failure or strong external interference.
[0029] If the sensor is abnormal, it will be inspected and repaired or the detection parameters will be adjusted. If the sensor is normal, visual prompts will be provided to the dispatch center and maintenance personnel through the remote monitoring platform or on-site indicator lights, indicating that there is an initial risk of line failure and suggesting key inspections.
[0030] Preferably, the fault location and early warning module includes a multi-dimensional feature fusion unit and a confidence calculation unit;
[0031] The multi-dimensional feature fusion unit is used to analyze the line breakage characteristics under different environments using the fused data output by the data fusion and preprocessing module, so as to calculate the static line breakage characteristic factor and dynamic line breakage characteristic factor of the corresponding monitoring area, specifically obtained in the following manner:
[0032] The static line breakage characteristic factor is obtained by calculating the difference between the square of the transient electromagnetic field change rate at the monitoring distance and when the train is stationary and the square of the mean value of the parameter during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration energy at the monitoring distance and when the train is stationary and the square of the mean value of the parameter during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added.
[0033] The dynamic line breakage characteristic factor is obtained by calculating the difference between the square of the transient electromagnetic field change rate at the monitoring distance and when the train is moving and the square of the mean value of the parameter during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration energy at the monitoring distance and when the train is moving and the square of the mean value of the parameter during the monitoring period is calculated and then multiplied by another weighting coefficient; the two weighted differences are added together and a correction constant is added.
[0034] Preferably, the confidence calculation unit is used to comprehensively evaluate the wire breakage fault situation based on the static wire breakage characteristic factors and the dynamic wire breakage characteristic factors, so as to analyze and calculate the fault confidence index, which is obtained in the following way:
[0035] The fault confidence index is obtained by squaring the difference between the mean of the static fault characteristic factors at different monitoring distances when the train is stationary and the mean of the static fault characteristic factors during team training, and then multiplying by a weighting coefficient. At the same time, the fault confidence index is obtained by squaring the difference between the mean of the dynamic fault characteristic factors at different monitoring distances when the train is moving and the mean of the dynamic fault characteristic factors during team training, and then multiplying by another weighting coefficient. The two weighted squared differences are added together and a correction constant is added.
[0036] Preferably, the system adaptive optimization module includes a risk assessment unit and a strategy optimization unit;
[0037] The risk assessment unit is used to correlate the fault confidence index with the background interference intensity factor output by the environmental noise modeling module, and after dimensionless processing, fit and calculate the disconnection risk assessment index. The disconnection risk assessment index is obtained by multiplying the square of the fault confidence index by a weighting coefficient, adding the square of the background interference intensity factor multiplied by another weighting coefficient, and then subtracting a correction constant.
[0038] The strategy optimization unit is used to pre-set an evaluation threshold range and compare the line break risk assessment index with the evaluation threshold to comprehensively assess the current line break risk level of the railway power supply system, and make corresponding handling suggestions based on the risk level. The specific content is as follows:
[0039] If the line break risk assessment index exceeds the assessment threshold range, it indicates that there is a high-risk line break fault in the current railway power supply system. At this time, the highest level emergency instruction will be immediately issued to the dispatch center, recommending that the relevant power supply area be cut off immediately and the emergency repair plan be activated.
[0040] If the line break risk assessment index is within the assessment threshold range, it indicates that the current railway power supply system has a medium-risk line break fault or a high-confidence initial line break risk. At this time, an early warning instruction will be issued to the dispatch center, suggesting speed limit operation or stopping the operation of trains on the relevant line, and arranging professional personnel to conduct on-site verification and prepare for emergency repairs.
[0041] If the line break risk assessment index does not reach the assessment threshold range, it indicates that the current railway power supply system is operating well or has low-risk potential hazards. In this case, a detailed report of this monitoring will be provided to the dispatch center, affirming the normal operation of the system and suggesting strengthening daily inspections and preventive maintenance.
[0042] A method for detecting power line breakage faults includes:
[0043] S1. Collect various physical quantity information in a distributed manner along the railway power supply contact network or catenary, including electromagnetic field change data, high-frequency vibration data, acoustic characteristic data and conductor micro-strain data;
[0044] S2. Construct a dynamic environmental noise model and analyze the background electromagnetic interference, mechanical vibration noise and other environmental noise data generated during real-time railway operation to obtain the background interference intensity factor.
[0045] S3. Perform data preprocessing and fusion of multiple physical quantity information and dynamic environmental noise model, and transmit the data to the cloud platform; at the same time, use dimensionless processing calculation to perform data standardization processing and perform spatiotemporal synchronization calibration.
[0046] S4. Based on the standardized processing of various physical quantity information and dynamic environmental noise model, identify and extract the initial weak characteristic signal and the complete disconnection characteristic signal of the power conductor open circuit fault, so as to construct the fault feature set and calculate and obtain the fault confidence index.
[0047] S5. Based on the fault confidence index and the analysis results of the environmental noise modeling module, comprehensively determine the fault type, location and urgency, and issue a warning command of the corresponding level.
[0048] S6. Correlate the fault confidence index with the background interference intensity factor to obtain the line break risk assessment index, and pre-set the assessment threshold range. Compare and analyze the line break risk assessment index with the assessment threshold to comprehensively assess the line break risk level of the current railway power supply system, and make corresponding handling suggestions based on the risk level.
[0049] Compared with the prior art, the beneficial effects of the present invention are:
[0050] This invention improves data quality from the source by using a distributed high-precision acquisition module with multimodal sensing modules and designs that resist strong electromagnetic interference and vibration. At the same time, the environmental noise modeling module builds a dynamic environmental noise model in real time and uses machine learning to accurately generate background interference intensity factors, thereby effectively filtering out complex environmental noise, significantly reducing the false alarm rate, and improving the ability to capture real disconnection signals.
[0051] This invention, through the initial feature analysis unit of the broken wire feature identification module, can accurately calculate the initial broken wire feature intensity index, thereby capturing initial weak feature signals such as local micro-strain anomalies in the conductor. Combining the static and dynamic broken wire feature factors calculated by the multi-dimensional feature fusion unit, the confidence calculation unit comprehensively evaluates the fault confidence index. The risk assessment unit of the system adaptive optimization module correlates the fault confidence index with the background interference intensity factor, fitting and obtaining the broken wire risk assessment index, achieving adaptive detection sensitivity adjustment based on real-time environmental noise, avoiding the drawbacks of a fixed high threshold. Finally, through the strategy optimization unit, the broken wire risk assessment index is compared with the assessment threshold range to achieve multi-level broken wire risk assessment and graded handling, greatly shortening the fault response time, improving the ability to capture incomplete broken wires or initial weak signals, and enhancing early warning efficiency, effectively ensuring the operational safety of the railway power supply system. Attached Figure Description
[0052] Figure 1 This is a schematic diagram of the overall equipment operation process of a power conductor breakage fault detection device according to the present invention;
[0053] Figure 2 This is a schematic diagram of the steps in the method for detecting broken wire faults in power conductors according to the present invention. Detailed Implementation
[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.
[0055] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0056] Example 1:
[0057] Please see Figure 1 The present invention provides a technical solution:
[0058] A power conductor breakage fault detection device, comprising:
[0059] Multimodal sensing modules are used to collect various physical quantities along the railway power supply contact network or catenary in a distributed manner, including electromagnetic field change data, high-frequency vibration data, acoustic characteristic data, and conductor micro-strain data.
[0060] The environmental noise modeling module is used to build a dynamic environmental noise model and analyze the background electromagnetic interference, mechanical vibration noise and other environmental noise data generated during real-time railway operation to obtain the background interference intensity factor.
[0061] The data fusion and preprocessing module is used to preprocess and fuse various physical quantity information with dynamic environmental noise models and transmit the data to the cloud platform; at the same time, it uses dimensionless processing calculations to perform data standardization and spatiotemporal synchronization calibration.
[0062] The wire breakage feature recognition module is used to identify and extract the initial weak feature signals and complete wire breakage feature signals of power conductor wire breakage faults based on standardized physical quantity information and dynamic environmental noise model, so as to construct a fault feature set and calculate and obtain the fault confidence index.
[0063] The fault location and early warning module is used to comprehensively judge the fault type, location and urgency based on the fault confidence index and the analysis results of the environmental noise modeling module, and issue early warning instructions of the corresponding level.
[0064] The system adaptive optimization module obtains the line break risk assessment index by correlating the fault confidence index with the background interference intensity factor, and pre-sets the assessment threshold range. It then compares and analyzes the line break risk assessment index with the assessment threshold to comprehensively assess the line break risk level of the current railway power supply system and makes corresponding handling suggestions based on the risk level.
[0065] Example 2:
[0066] The multimodal sensing module includes a distributed sensing unit and a high-precision acquisition unit;
[0067] The distributed sensing unit is used along the railway power supply contact network, based on the uniformly distributed induction coils, fiber optic vibration sensors, micro acoustic sensor arrays and local impedance sensors on the catenary to obtain high-resolution and real-time multimodal data.
[0068] Furthermore, induction coils are used to collect transient electromagnetic field changes, fiber optic vibration sensors are used to collect high-frequency vibrations and micro-strain of conductors, miniature acoustic sensor arrays are used to collect abnormal sounds, and local impedance sensors are used to monitor the continuity of conductors.
[0069] The high-precision acquisition unit is used to perform high-precision synchronous sampling and preliminary digital processing on the raw data collected by the distributed sensing unit, and is equipped with anti-strong electromagnetic interference and anti-vibration design.
[0070] The environmental noise modeling module includes a noise feature acquisition unit and a dynamic modeling unit;
[0071] The noise feature acquisition unit is used to monitor and record background noise data in the railway operating environment in real time. The background noise data includes the traction current fluctuation characteristics when the train passes through each monitoring period, the acoustic spectrum generated by pantograph-catenary friction, the line vibration amplitude, and the environmental electromagnetic field intensity. Based on the background noise data, and combined with a statistical averaging algorithm, the average intensity of each noise type within the monitoring period is calculated.
[0072] The dynamic modeling unit is used to combine the average intensity of each noise type acquired by the noise feature acquisition unit with historical data and machine learning algorithms to construct a real-time updated dynamic environmental noise model, which is used to predict and identify the noise level and characteristics of the current environment in order to construct a background interference intensity factor.
[0073] The broken wire feature recognition module includes an initial feature analysis unit and a complete feature recognition unit;
[0074] The initial characteristic analysis unit is used to calculate the initial wire breakage characteristic intensity index based on the fused data by analyzing signals including local micro-strain anomalies, high-frequency vibration mode changes, weak transient electromagnetic field disturbances, and subtle local impedance fluctuations. The initial wire breakage characteristic intensity index is obtained in the following way:
[0075] The initial line breakage characteristic intensity index is obtained by calculating the difference between the square of the conductor microstrain change rate during the monitoring period and the square of the mean conductor microstrain change rate during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration spectrum entropy during the monitoring period and the square of the mean high-frequency vibration spectrum entropy during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added.
[0076] Furthermore, the formula for calculating the fracture characteristic intensity index is as follows:
[0077]
[0078] In the formula, Cqtz represents the initial line breakage characteristic intensity index, which is the intensity of the initial line breakage characteristic of the power conductor during the current monitoring period. The higher the index, the greater the risk of initial line breakage. W1 is a weighting coefficient used to measure the contribution of the conductor's micro-strain change rate to the initial line breakage characteristic intensity. Its value is set based on actual system debugging and experience. W2 is a weighting coefficient used to measure the contribution of the high-frequency vibration spectrum entropy to the initial line breakage characteristic intensity. Its value is set based on actual system debugging and experience.
[0079] Sl i To monitor the rate of change of conductor microstrain during the monitoring period, the parameter source is the conductor microstrain data collected by the fiber optic vibration sensor in the multimodal sensing module. After being processed by the data preprocessing module, the instantaneous rate of change of conductor microstrain during the i-th monitoring period is calculated. The average rate of change of conductor microstrain within the monitoring period is the same as above. It is the average rate of change of conductor microstrain obtained by the noise feature acquisition unit or data processing module over a longer monitoring period, and is used as the benchmark under normal conditions.
[0080] H i The high-frequency vibration spectrum entropy during the monitoring period is obtained from high-frequency vibration data collected by fiber optic vibration sensors or miniature acoustic sensor arrays in the multimodal sensing module. After being processed by the data preprocessing module, the spectrum entropy of the high-frequency vibration signal in the i-th monitoring period is calculated, reflecting the complexity and randomness of the signal. The average value of the high-frequency vibration spectrum entropy within the monitoring period is obtained from the same source as above. It is the average value of the high-frequency vibration spectrum entropy obtained by the noise feature acquisition unit or data processing module over a longer monitoring period, and serves as a benchmark under normal conditions.
[0081] C1 is a correction constant used to adjust the baseline or bias of the initial fracture characteristic strength index to ensure that the calculation results are within a reasonable range. Its value is set according to system calibration and actual needs.
[0082] The complete feature recognition unit is used to pre-set a disconnection threshold and compare the disconnection threshold with the initial disconnection feature intensity index to preliminarily determine whether there is an initial disconnection risk in the current power conductor. The specific content is as follows:
[0083] If the initial disconnection characteristic intensity index exceeds the disconnection threshold, it is preliminarily determined that there is an initial disconnection risk in the current power conductor, indicating that there is a potential fault in the current line. At this time, an initial warning command will be issued.
[0084] If the initial line breakage characteristic intensity index does not exceed the line breakage threshold, it is preliminarily determined that there is no initial line breakage risk in the current power conductor, indicating that the current line is in good operating condition, and there is no need to issue additional initial warning instructions.
[0085] The fault location and early warning module is used to monitor and record real-time data of a local area of the contact network in a timely manner after receiving the initial early warning command issued by the complete feature recognition unit. This includes electromagnetic field strength, vibration frequency, acoustic anomalies and conductor tension. The module also performs self-checks on the status of each sensor to determine whether false alarms are caused by sensor failure or strong external interference.
[0086] If the sensor is abnormal, it will be inspected and repaired or the detection parameters will be adjusted. If the sensor is normal, visual prompts will be provided to the dispatch center and maintenance personnel through the remote monitoring platform or on-site indicator lights, indicating that there is an initial risk of line failure and suggesting key inspections.
[0087] The fault location and early warning module includes a multi-dimensional feature fusion unit and a confidence calculation unit;
[0088] The multi-dimensional feature fusion unit is used to analyze the line breakage characteristics under different environments using the fused data output by the data fusion and preprocessing module, so as to calculate the static line breakage characteristic factor and dynamic line breakage characteristic factor of the corresponding monitoring area, specifically obtained in the following manner:
[0089] The static line breakage characteristic factor is obtained by calculating the difference between the square of the transient electromagnetic field change rate at the monitoring distance and when the train is stationary and the square of the mean value of the parameter during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration energy at the monitoring distance and when the train is stationary and the square of the mean value of the parameter during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added.
[0090] The dynamic line breakage characteristic factor is obtained by calculating the difference between the square of the transient electromagnetic field change rate at the monitoring distance and when the train is moving and the square of the mean value of the parameter during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration energy at the monitoring distance and when the train is moving and the square of the mean value of the parameter during the monitoring period is calculated and then multiplied by another weighting coefficient; the two weighted differences are added together and a correction constant is added.
[0091] Furthermore, the specific formula for calculating the static disconnection characteristic factor Jjyz is as follows:
[0092]
[0093] In the formula, Jjyz is the static line breakage characteristic factor, which represents the intensity of the line breakage characteristic at a specific monitoring distance when the train is stationary;
[0094] W3 is a weighting coefficient used to measure the contribution of the transient electromagnetic field change rate to the static wire breakage characteristic factor under static conditions; W4 is a weighting coefficient used to measure the contribution of high-frequency vibration energy to the static wire breakage characteristic factor under static conditions.
[0095] E k，st The transient electromagnetic field change rate at a monitoring distance of k and with the train stationary is calculated by taking electromagnetic field data collected by the induction coil in the multi-modal sensing module and then processing the data through the data preprocessing and fusion module.
[0096] Est is the average rate of change of transient electromagnetic field under stationary conditions within the monitoring period: The parameter source is the same as above, which is the average rate of change of transient electromagnetic field under stationary conditions of the train over a longer monitoring period.
[0097] Vk,st represents the high-frequency vibration energy at a monitoring distance of k and when the train is stationary. The parameter is obtained from vibration data collected by fiber optic vibration sensors or miniature acoustic sensor arrays in the multimodal sensing module. After data preprocessing and fusion, the energy of the high-frequency vibration signal at a monitoring distance of k and when the train is stationary is calculated.
[0098] Vst is the average high-frequency vibration energy under stationary conditions within the monitoring period. The parameter source is the same as above, and it is the average value of high-frequency vibration energy under stationary conditions of the train over a longer monitoring period.
[0099] C2 is a correction constant used to adjust the baseline or bias of the static disconnection characteristic factor;
[0100] Furthermore, the specific formula for calculating the dynamic disconnection characteristic factor is as follows:
[0101]
[0102] In the formula, Yjyz is the dynamic line breakage characteristic factor, which represents the intensity of the line breakage characteristic at a specific monitoring distance when the train is in motion;
[0103] W5 is a weighting coefficient used to measure the contribution of the transient electromagnetic field change rate to the dynamic wire breakage characteristic factor under moving conditions; W6 is a weighting coefficient used to measure the contribution of high-frequency vibration energy to the dynamic wire breakage characteristic factor under moving conditions.
[0104] E j,mo The transient electromagnetic field change rate is given at a monitoring distance of j and while the train is in motion. The parameter source is the same as E. k,st However, data is collected and calculated when the monitoring distance is j and the train is moving;
[0105] E moTo monitor the average rate of change of transient electromagnetic field under moving conditions within the monitoring period, the parameter source is the same as E. k,st However, over a longer monitoring period, the average value of the transient electromagnetic field change rate under train movement conditions is observed.
[0106] V j,mo This refers to the high-frequency vibration energy at a monitoring distance of j while the train is in motion. The parameter source is the same as V. k,st However, data is collected and calculated when the monitoring distance is j and the train is moving;
[0107] V mo To monitor the average high-frequency vibration energy during the movement period, the parameter source is the same as V. k,st However, over a longer monitoring period, the average value of high-frequency vibration energy under train movement conditions;
[0108] C3 is a correction constant used to adjust the baseline or bias of the dynamic disconnection characteristic factor.
[0109] The confidence calculation unit is used to comprehensively evaluate the wire breakage fault situation based on static wire breakage characteristic factors and dynamic wire breakage characteristic factors, so as to analyze and calculate the fault confidence index, which is obtained in the following way:
[0110] The fault confidence index is obtained by squaring the difference between the mean of the static fault characteristic factors at different monitoring distances when the train is stationary and the mean of the static fault characteristic factors during team training, and then multiplying by a weighting coefficient. At the same time, the fault confidence index is obtained by squaring the difference between the mean of the dynamic fault characteristic factors at different monitoring distances when the train is moving and the mean of the dynamic fault characteristic factors during team training, and then multiplying by another weighting coefficient. The two weighted squared differences are added together and a correction constant is added.
[0111] Furthermore, the specific formula for calculating the fault confidence index is as follows:
[0112]
[0113] In the formula, Gzzd is the fault confidence index, representing the reliability or confidence level of the currently detected open circuit fault signal. The higher the index, the greater the probability of the fault occurring;
[0114] W7 is a weighting coefficient used to measure the contribution of static disconnection characteristic factor deviation to the fault confidence index; W8 is a weighting coefficient used to measure the contribution of dynamic disconnection characteristic factor deviation to the fault confidence index.
[0115] The mean value of static line breakage feature factors at different monitoring distances when the train is stationary is given. The parameter is the average value of multiple static line breakage feature factors Jjyz calculated by the multi-dimensional feature fusion unit, which represents the overall line breakage feature performance under the current stationary state.
[0116] The mean value of static disconnection feature factors during team training is the average value of static disconnection feature factors obtained by the system from a large amount of training data with no faults or known faults, which serves as a benchmark under normal or typical fault modes; "team training" here refers to the model training and calibration process of the system before deployment.
[0117] The mean value of dynamic line breakage feature factors at different monitoring distances when the train is in motion is the parameter source, which is the average value of multiple dynamic line breakage feature factors Yjyz calculated by the multi-dimensional feature fusion unit, representing the overall line breakage feature performance under the current moving state.
[0118] The mean of the dynamic disconnection feature factors during team training; the parameter source is the same as... , is the average value of the dynamic disconnection feature factors obtained by the system from the training data;
[0119] C4 is a correction constant used to adjust the baseline or bias of the fault confidence index;
[0120] The system adaptive optimization module includes a risk assessment unit and a strategy optimization unit;
[0121] The risk assessment unit is used to correlate the fault confidence index with the background interference intensity factor output by the environmental noise modeling module, and after dimensionless processing, fit and calculate the disconnection risk assessment index. The disconnection risk assessment index is obtained by multiplying the square of the fault confidence index by a weighting coefficient, adding the square of the background interference intensity factor multiplied by another weighting coefficient, and then subtracting a correction constant.
[0122] Furthermore, the specific formula for calculating the disconnection risk assessment index Dxpg is as follows:
[0123]
[0124] In the formula, Dxpg is the line break risk assessment index, which comprehensively assesses the risk level of a power line break fault in the current railway power supply system; the higher the index, the higher the risk level.
[0125] W9 is a weighting coefficient used to measure the contribution of the fault confidence index to the line break risk assessment index; W10 is a weighting coefficient used to measure the contribution of the background interference intensity factor to the line break risk assessment index.
[0126] Gzzd is the fault confidence index, whose parameter is derived from the output of the confidence calculation unit, reflecting the reliability of the detected fault signal; Bjgf is the background interference intensity factor, whose parameter is derived from the output of the environmental noise modeling module, reflecting the noise level and characteristics of the current environment.
[0127] C5 is a correction constant used to adjust the baseline or bias of the disconnection risk assessment index;
[0128] The strategy optimization unit is used to pre-set an evaluation threshold range and compare the line break risk assessment index with the evaluation threshold to comprehensively assess the current line break risk level of the railway power supply system, and make corresponding handling suggestions based on the risk level. The specific content is as follows:
[0129] If the line break risk assessment index exceeds the assessment threshold range, it indicates that there is a high-risk line break fault in the current railway power supply system. At this time, the highest level emergency instruction will be immediately issued to the dispatch center, recommending that the relevant power supply area be cut off immediately and the emergency repair plan be activated.
[0130] If the line break risk assessment index is within the assessment threshold range, it indicates that the current railway power supply system has a medium-risk line break fault or a high-confidence initial line break risk. At this time, an early warning instruction will be issued to the dispatch center, suggesting speed limit operation or stopping the operation of trains on the relevant line, and arranging professional personnel to conduct on-site verification and prepare for emergency repairs.
[0131] If the line break risk assessment index does not reach the assessment threshold range, it indicates that the current railway power supply system is operating well or has low-risk potential hazards. In this case, a detailed report of this monitoring will be provided to the dispatch center, affirming the normal operation of the system and suggesting strengthening daily inspections and preventive maintenance.
[0132] Example 3:
[0133] Please see Figure 2 A method for detecting power line breakage faults, comprising:
[0134] S1. Collect various physical quantity information in a distributed manner along the railway power supply contact network or catenary, including electromagnetic field change data, high-frequency vibration data, acoustic characteristic data and conductor micro-strain data;
[0135] S2. Construct a dynamic environmental noise model and analyze the background electromagnetic interference, mechanical vibration noise and other environmental noise data generated during real-time railway operation to obtain the background interference intensity factor.
[0136] S3. Perform data preprocessing and fusion of multiple physical quantity information and dynamic environmental noise model, and transmit the data to the cloud platform; at the same time, use dimensionless processing calculation to perform data standardization processing and perform spatiotemporal synchronization calibration.
[0137] S4. Based on the standardized processing of various physical quantity information and dynamic environmental noise model, identify and extract the initial weak characteristic signal and the complete disconnection characteristic signal of the power conductor open circuit fault, so as to construct the fault feature set and calculate and obtain the fault confidence index.
[0138] S5. Based on the fault confidence index and the analysis results of the environmental noise modeling module, comprehensively determine the fault type, location and urgency, and issue a warning command of the corresponding level.
[0139] S6. Correlate the fault confidence index with the background interference intensity factor to obtain the line break risk assessment index, and pre-set the assessment threshold range. Compare and analyze the line break risk assessment index with the assessment threshold to comprehensively assess the line break risk level of the current railway power supply system, and make corresponding handling suggestions based on the risk level.
[0140] It should be noted that all calculation formulas in this application employ regression analysis, including but not limited to machine learning algorithms, to deeply analyze the collected parameters and identify their natural trends and interrelationships. Specialized software, such as Python's Scikit-learn library or the R language, is used to automatically generate mathematical models that match the data. Then, cross-validation and other methods are used to objectively evaluate the model performance, and continuous feedback and optimization are combined to ensure that the created formulas truly reflect the inherent laws of the data, thereby guaranteeing their effectiveness and accuracy. In all calculation formulas in this application, the parameters in each formula undergo dimensionless processing within a consistent range to ensure that different physical quantities are compared on the same scale; dimensionless processing techniques include, but are not limited to, min-max normalization and Z-score standardization.
[0141] The technical solution of this invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of this invention.
[0142] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.
[0143] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
[0144] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A power conductor breakage fault detection device, characterized in that, include: Multimodal sensing modules are used to collect various physical quantities along the railway power supply contact network or catenary in a distributed manner, including electromagnetic field change data, high-frequency vibration data, acoustic characteristic data, and conductor micro-strain data. The environmental noise modeling module is used to build a dynamic environmental noise model and analyze the background electromagnetic interference, mechanical vibration noise and other environmental noise data generated during real-time railway operation to obtain the background interference intensity factor. The data fusion and preprocessing module is used to preprocess and fuse various physical quantity information with dynamic environmental noise models and transmit the data to the cloud platform; at the same time, it uses dimensionless processing calculations to perform data standardization and spatiotemporal synchronization calibration. The wire breakage feature recognition module is used to identify and extract the initial weak feature signals and complete wire breakage feature signals of power conductor wire breakage faults based on standardized physical quantity information and dynamic environmental noise model, so as to construct a fault feature set and calculate and obtain the fault confidence index; the wire breakage feature recognition module includes an initial feature analysis unit and a complete feature recognition unit. The initial characteristic analysis unit is used to calculate the initial wire breakage characteristic intensity index based on the fused data by analyzing signals including local micro-strain anomalies, high-frequency vibration mode changes, weak transient electromagnetic field disturbances, and subtle local impedance fluctuations. The initial wire breakage characteristic intensity index is obtained in the following way: The initial line breakage characteristic intensity index is obtained by calculating the difference between the square of the conductor microstrain change rate during the monitoring period and the square of the mean conductor microstrain change rate during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration spectrum entropy during the monitoring period and the square of the mean high-frequency vibration spectrum entropy during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added. The fault location and early warning module is used to comprehensively judge the fault type, location and urgency based on the fault confidence index and the analysis results of the environmental noise modeling module, and issue early warning instructions of the corresponding level. The system adaptive optimization module obtains the line break risk assessment index by correlating the fault confidence index with the background interference intensity factor, and pre-sets the assessment threshold range. It then compares and analyzes the line break risk assessment index with the assessment threshold to comprehensively assess the line break risk level of the current railway power supply system and makes corresponding handling suggestions based on the risk level.
2. The power conductor breakage fault detection device according to claim 1, characterized in that: The multimodal sensing module includes a distributed sensing unit and a high-precision acquisition unit; The distributed sensing unit is used along the railway power supply contact network, based on the uniformly distributed induction coils, fiber optic vibration sensors, micro acoustic sensor arrays and local impedance sensors on the catenary to obtain high-resolution and real-time multimodal data. The high-precision acquisition unit is used to perform high-precision synchronous sampling and preliminary digital processing on the raw data collected by the distributed sensing unit, and is equipped with anti-strong electromagnetic interference and anti-vibration design.
3. The power conductor breakage fault detection device according to claim 2, characterized in that: The environmental noise modeling module includes a noise feature acquisition unit and a dynamic modeling unit; The noise feature acquisition unit is used to monitor and record background noise data in the railway operating environment in real time. The background noise data includes the traction current fluctuation characteristics when the train passes through each monitoring period, the acoustic spectrum generated by pantograph-catenary friction, the line vibration amplitude, and the environmental electromagnetic field intensity. Based on the background noise data, and combined with a statistical averaging algorithm, the average intensity of each noise type within the monitoring period is calculated. The dynamic modeling unit is used to combine the average intensity of each noise type acquired by the noise feature acquisition unit with historical data and machine learning algorithms to construct a real-time updated dynamic environmental noise model, which is used to predict and identify the noise level and characteristics of the current environment in order to construct a background interference intensity factor.
4. The power conductor breakage fault detection device according to claim 3, characterized in that: The complete feature recognition unit is used to pre-set a disconnection threshold and compare the disconnection threshold with the initial disconnection feature intensity index to preliminarily determine whether there is an initial disconnection risk in the current power conductor. The specific content is as follows: If the initial disconnection characteristic intensity index exceeds the disconnection threshold, it is preliminarily determined that there is an initial disconnection risk in the current power conductor, indicating that there is a potential fault in the current line. At this time, an initial warning command will be issued. If the initial line breakage characteristic intensity index does not exceed the line breakage threshold, it is preliminarily determined that there is no initial line breakage risk in the current power conductor, indicating that the current line is in good operating condition, and there is no need to issue additional initial warning instructions.
5. The power conductor breakage fault detection device according to claim 4, characterized in that: The fault location and early warning module is used to monitor and record real-time data of a local area of the contact network in a timely manner after receiving the initial early warning command issued by the complete feature recognition unit. This includes electromagnetic field strength, vibration frequency, acoustic anomalies and conductor tension. The module also performs self-checks on the status of each sensor to determine whether false alarms are caused by sensor failure or strong external interference. If the sensor is abnormal, it will be inspected and repaired or the detection parameters will be adjusted. If the sensor is normal, visual prompts will be provided to the dispatch center and maintenance personnel through the remote monitoring platform or on-site indicator lights, indicating that there is an initial risk of line failure and suggesting key inspections.
6. The power conductor breakage fault detection device according to claim 5, characterized in that: The fault location and early warning module includes a multi-dimensional feature fusion unit and a confidence calculation unit; The multi-dimensional feature fusion unit is used to analyze the line breakage characteristics under different environments using the fused data output by the data fusion and preprocessing module, so as to calculate the static line breakage characteristic factor and dynamic line breakage characteristic factor of the corresponding monitoring area, specifically obtained in the following manner: The static line breakage characteristic factor is obtained by calculating the difference between the square of the transient electromagnetic field change rate at the monitoring distance and when the train is stationary and the square of the mean transient electromagnetic field change rate during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration energy at the monitoring distance and when the train is stationary and the square of the mean high-frequency vibration energy during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added. The dynamic line breakage characteristic factor is obtained by calculating the difference between the square of the transient electromagnetic field change rate at the monitoring distance and when the train is moving and the square of the mean transient electromagnetic field change rate during the monitoring period, and then multiplying it by a weighting coefficient; at the same time, the difference between the square of the high-frequency vibration energy at the monitoring distance and when the train is moving and the square of the mean high-frequency vibration energy during the monitoring period is calculated and then multiplied by another weighting coefficient; these two weighted differences are added together and a correction constant is added.
7. The power conductor breakage fault detection device according to claim 6, characterized in that: The confidence calculation unit is used to comprehensively evaluate the wire breakage fault situation based on static wire breakage characteristic factors and dynamic wire breakage characteristic factors, so as to analyze and calculate the fault confidence index, which is obtained in the following way: The fault confidence index is obtained by squaring the difference between the mean of the static fault characteristic factors at different monitoring distances when the train is stationary and the mean of the static fault characteristic factors during team training, and then multiplying by a weighting coefficient. At the same time, the fault confidence index is obtained by squaring the difference between the mean of the dynamic fault characteristic factors at different monitoring distances when the train is moving and the mean of the dynamic fault characteristic factors during team training, and then multiplying by another weighting coefficient. The two weighted squared differences are added together and a correction constant is added.
8. The power conductor breakage fault detection device according to claim 7, characterized in that: The system adaptive optimization module includes a risk assessment unit and a strategy optimization unit; The risk assessment unit is used to correlate the fault confidence index with the background interference intensity factor output by the environmental noise modeling module, and after dimensionless processing, fit and calculate the disconnection risk assessment index. The disconnection risk assessment index is obtained by multiplying the square of the fault confidence index by a weighting coefficient, adding the square of the background interference intensity factor multiplied by another weighting coefficient, and then subtracting a correction constant. The strategy optimization unit is used to pre-set an evaluation threshold range and compare the line break risk assessment index with the evaluation threshold to comprehensively assess the current line break risk level of the railway power supply system, and make corresponding handling suggestions based on the risk level. The specific content is as follows: If the line break risk assessment index exceeds the assessment threshold range, it indicates that there is a high-risk line break fault in the current railway power supply system. At this time, the highest level emergency instruction will be immediately issued to the dispatch center, recommending that the relevant power supply area be cut off immediately and the emergency repair plan be activated. If the line break risk assessment index is within the assessment threshold range, it indicates that the current railway power supply system has a medium-risk line break fault or a high-confidence initial line break risk. At this time, an early warning instruction will be issued to the dispatch center, suggesting speed limit operation or stopping the operation of trains on the relevant line, and arranging professional personnel to conduct on-site verification and prepare for emergency repairs. If the line break risk assessment index does not reach the assessment threshold range, it indicates that the current railway power supply system is operating well or has low-risk potential hazards. In this case, a detailed report of this monitoring will be provided to the dispatch center, affirming the normal operation of the system and suggesting strengthening daily inspections and preventive maintenance.
9. A method for detecting power line breakage faults, characterized in that, The method is used to execute the power conductor breakage fault detection device according to any one of claims 1-8, comprising: S1. Collect various physical quantity information in a distributed manner along the railway power supply contact network or catenary, including electromagnetic field change data, high-frequency vibration data, acoustic characteristic data and conductor micro-strain data; S2. Construct a dynamic environmental noise model and analyze the background electromagnetic interference, mechanical vibration noise and other environmental noise data generated during real-time railway operation to obtain the background interference intensity factor. S3. Perform data preprocessing and fusion of multiple physical quantity information and dynamic environmental noise model, and transmit the data to the cloud platform; at the same time, use dimensionless processing calculation to perform data standardization processing and perform spatiotemporal synchronization calibration. S4. Based on the standardized processing of various physical quantity information and dynamic environmental noise model, identify and extract the initial weak characteristic signal and the complete disconnection characteristic signal of the power conductor open circuit fault, so as to construct the fault feature set and calculate and obtain the fault confidence index. S5. Based on the fault confidence index and the analysis results of the environmental noise modeling module, comprehensively determine the fault type, location and urgency, and issue a warning command of the corresponding level. S6. Correlate the fault confidence index with the background interference intensity factor to obtain the line break risk assessment index, and pre-set the assessment threshold range. Compare and analyze the line break risk assessment index with the assessment threshold to comprehensively assess the line break risk level of the current railway power supply system, and make corresponding handling suggestions based on the risk level.