Safety operation system and safety operation method for catenary maintenance vehicle

Abstract

The application discloses a catenary maintenance operation vehicle safety operation system and safety operation method, comprising: a data acquisition system, comprising a plurality of temperature and vibration composite sensors, which are distributedly installed on each key component of a catenary maintenance operation vehicle running transmission system, and are used for collecting temperature and vibration acceleration signals of each monitoring point in real time; a signal transmission system, comprising a plurality of pre-acquisition boxes, each pre-acquisition box is connected with a plurality of temperature and vibration composite sensors, and is used for transmitting analog signals output by the temperature and vibration composite sensors to a vehicle-mounted host computer after preliminary signal conditioning and format conversion; an intelligent diagnosis system, comprising an intelligent diagnosis model carried on the vehicle-mounted host computer, the intelligent diagnosis system is used for receiving temperature digital signals and vibration digital signals to diagnose the running state of the component in real time, identifying a fault type, and giving on-site disposal measures based on the diagnosis result; and the intelligent diagnosis model matches corresponding differential diagnosis strategies according to corresponding component types.

Description

Technical Field

[0001] This invention relates to the field of rail vehicle condition monitoring and fault diagnosis technology, and in particular, to a safe operation and maintenance system for an overhead contact line maintenance vehicle. Furthermore, this invention also relates to a safe operation and maintenance method for an overhead contact line maintenance vehicle that includes the aforementioned safe operation and maintenance system. Background Technology

[0002] The overhead contact line maintenance vehicle is a key piece of equipment in the electrified railway power supply maintenance system, undertaking important tasks such as daily maintenance of overhead contact line equipment, emergency repair of faults, and line maintenance. With the continuous growth of my country's high-speed railway operating mileage and the continuous improvement of train operating speed, the frequency of use of the overhead contact line maintenance vehicle is increasing day by day, and its operational safety is directly related to the reliability of the railway power supply system and the safety of high-speed railway operation.

[0003] The safe operation and maintenance of overhead contact line maintenance vehicles faces multiple challenges. First, these vehicles are used frequently and require high efficiency, yet their maintenance still relies heavily on manual experience and regular inspections, resulting in low efficiency and a high rate of missed inspections. Therefore, online monitoring and intelligent troubleshooting of potential faults are urgently needed. Second, as the frequency of these vehicles' operation on high-speed rail lines increases, higher demands are placed on early warning and accurate diagnosis of faults—any operational malfunction could disrupt the normal operation of the high-speed rail line. Third, the continuous increase in the service life of these vehicles leads to component performance degradation and an increase in potential faults, necessitating safety precautions and full life-cycle health management. Fourth, these vehicles face the need for increased speeds, from 80 km / h to 120 km / h or even 160 km / h. Theoretical calculations and experimental studies have shown that increased speed leads to increased vehicle vibration and reduced component reliability; therefore, a comprehensive investigation of potential faults is required to prevent malfunctions.

[0004] For example, the technical solution disclosed in publication number CN215622011U is an online monitoring system for train running gear and a single-axle derailment monitoring system. This system includes a ground system and an on-board system. The on-board system comprises multiple on-board subsystems, each including an on-board host, a pre-processor, a temperature and vibration composite sensor, and connecting cables. This solution monitors the temperature and vibration of the running gear in real time during operation by installing a temperature and vibration composite sensor on each train bogie bearing and collecting signals through the pre-processor and the on-board host. However, this solution mainly focuses on monitoring the train bogie bearings, resulting in a limited monitoring scope and failing to cover all aspects of train operation. The system covers all key components of the entire travel transmission system, but its diagnostic functions are limited to signal acquisition and basic temperature and vibration alarms, lacking differentiated intelligent diagnostic capabilities for different component fault characteristics. For example, the technical solution disclosed in CN218858424U is a catenary work vehicle axle temperature monitoring system, including an axle temperature measurement and transmission module, an axle temperature monitoring and acquisition module, a human-machine interaction module, and an axle temperature background data management module. This solution only monitors axle temperature, with a single monitoring parameter, and cannot comprehensively analyze multi-dimensional state parameters such as vibration and impact, let alone achieve early identification and intelligent diagnosis of complex faults. In summary, existing technologies for overhead contact line maintenance vehicles lack comprehensive monitoring coverage, often limited to temperature and vibration monitoring of bogie bearings or axle boxes, failing to cover all critical components of the entire transmission system. Core power transmission components such as engines, torque converters, transfer cases, air compressors, and various gearboxes lack effective online monitoring methods. Diagnostic strategies are homogenized, using the same diagnostic algorithms and alarm thresholds for different components, failing to develop differentiated diagnostic models tailored to the unique fault mechanisms, signal characteristics, and operating conditions of each component. This results in insufficient diagnostic sensitivity for some components and a large number of false alarms for others. Furthermore, the lack of a multi-level alarm fusion mechanism means that existing solutions typically rely on single temperature threshold alarms or vibration amplitude alarms, leading to insufficient accuracy and reliability. Current technologies cannot meet the actual needs of overhead contact line maintenance vehicles for online monitoring, differentiated intelligent diagnosis, and precise operation and maintenance decision-making for critical components across the entire transmission chain. Summary of the Invention

[0005] This invention provides a safety operation and maintenance system and method for overhead contact line maintenance vehicles, in order to solve the technical problems of incomplete monitoring coverage and homogeneous diagnostic strategies of existing overhead contact line maintenance vehicles.

[0006] According to one aspect of the present invention, a safe operation and maintenance system for a catenary maintenance vehicle is provided, comprising: The data acquisition system includes multiple temperature and vibration composite sensors. The temperature and vibration composite sensors integrate temperature-sensitive elements and vibration-sensitive elements to simultaneously output temperature signals and vibration signals. The temperature and vibration composite sensors are distributed and installed on key components of the overhead contact line maintenance vehicle's travel transmission system to collect temperature and vibration acceleration signals at each monitoring point in real time. The signal transmission system includes multiple front-end acquisition boxes, each of which is connected to multiple temperature and vibration composite sensors. The system is used to perform preliminary signal conditioning and format conversion on the analog signals output by the temperature and vibration composite sensors and then transmit them to the vehicle host. The intelligent diagnostic system includes an intelligent diagnostic model mounted on the vehicle host. The intelligent diagnostic system is used to receive temperature digital signals and vibration digital signals transmitted by each of the front-end acquisition boxes to perform real-time diagnosis of the operating status of the components, identify the fault type, and provide on-site handling measures based on the diagnostic results. The intelligent diagnostic model matches corresponding differentiated diagnostic strategies according to the component type corresponding to each monitoring point.

[0007] As a further improvement to the above technical solution, the key components include an engine, a torque converter, a transfer case, an air compressor, a primary gearbox, a secondary gearbox, and wheelsets.

[0008] As a further improvement to the above technical solution, the intelligent diagnostic model includes a temperature alarm module, a time-domain feature alarm module, and a fault feature frequency amplitude alarm module. The temperature alarm module generates an alarm by comparing the temperature value and temperature difference value of each monitoring point with a preset threshold. The time-domain feature alarm module generates an alarm by comparing the root mean square value and kurtosis index of the vibration signal at each monitoring point with a preset threshold. The fault feature frequency amplitude alarm module generates an alarm by constructing a fault feature frequency extraction strategy for different components based on their fault mechanisms and structural parameters. The most severe alarm result determines the maintenance measures.

[0009] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the engine: The original vibration signal is preprocessed to remove combustion impact interference. A dual sparse dictionary is constructed by combining the structural parameters of the crankshaft connecting rod bearing, the valve train mechanism and the reciprocating motion parameters of the piston. Combustion noise interference is removed by sparse decomposition algorithm, and independent fault frequencies of bearing wear and valve train mechanism abnormalities are separated and judged based on amplitude threshold.

[0010] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the torque converter: The original vibration signal is preprocessed to filter out hydraulic damping clutter. A dual sparse dictionary is constructed based on the structural parameters of the supporting bearing, gear parameters, and speed difference characteristics. The specific fault frequencies of bearing wear eccentricity, gear cavitation, and internal slippage are extracted through sparse decomposition and identified by amplitude determination.

[0011] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the transfer case: The original vibration signal is preprocessed to filter out interference noise caused by intermittent loading and load switching of the air compressor. A sparse dictionary library adapted to multiple working conditions is constructed based on the bearing structure of the multi-support shaft system and the transmission parameters of the dual-branch gear transmission of the running and power take-off. The mixed vibration signal is decomposed by sparse decomposition to separate the independent fault characteristic frequencies of the multi-shaft bearing and the power distribution gear, and the amplitude index is extracted to complete the threshold determination.

[0012] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for air compressors: The original vibration signal is normalized and noise-reduced to filter out interference noise caused by belt vibration, no-load start-up and air pressure switching. A dual sparse dictionary is constructed based on the structural parameters of the air compressor crankshaft support bearing, the reciprocating motion characteristics of the crankshaft connecting rod, and the working parameters of the piston cylinder and intake and exhaust valve group. The rotational fault component and the reciprocating compression mechanism fault component are separated by an adaptive matching tracking algorithm, and the judgment is made in combination with the total air cylinder pressure and the real-time load adaptive threshold.

[0013] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the secondary gearbox: The original vibration signal is subjected to high-frequency filtering and signal smoothing. A dual sparse dictionary adapted to high-speed working conditions is constructed based on the structural parameters of high-speed rolling bearings and the meshing transmission parameters of small-module high-speed gears. The independent fault components of bearing fatigue damage and high-speed gear meshing abnormality are reconstructed by an adaptive sparse decomposition algorithm and judged in combination with real-time speed and load adaptive threshold.

[0014] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the first-stage gearbox: The original vibration signal is enhanced by low-frequency effective signal processing and random impact noise is filtered out. Based on the mechanical parameters of heavy-duty roller bearings and the meshing characteristics of large-module low-speed transmission gears, a dual sparse dictionary adapted to heavy-duty low-speed working conditions is constructed. The independent fault frequency components of low-speed bearings and heavy-duty gears are reconstructed by an adaptive matching pursuit sparse decomposition algorithm, and the judgment is made in combination with the real-time output torque adaptive dynamic threshold.

[0015] As a further improvement to the above technical solution, the fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for wheelsets: The original vibration signal is detrended and digitally filtered to remove noise from rail gap impact and car body resonance. Based on the geometric parameters of the wheelset bearing, the real-time speed of the wheel, and the tread wear mechanism, a dual sparse dictionary is constructed for the bearing and the tread respectively. The bearing fault frequency and the tread damage characteristic frequency are separated by an adaptive matching tracking algorithm and judged by adaptive threshold of vehicle speed condition.

[0016] As a further improvement to the above technical solution, the data acquisition system includes a first measuring point located at the input shaft end of the secondary gearbox, a second measuring point located at the output shaft end of the secondary gearbox, a third measuring point located at the first end of the wheel axle of the secondary gearbox, a fourth measuring point located at the second end of the wheel axle of the secondary gearbox, a fifth measuring point located at the axle box of the second-end bogie, a sixth measuring point located at the axle box of the second-end bogie, a seventh measuring point located on the main beam, an eighth measuring point located on top of the torque converter, a ninth measuring point located on top of the engine drive end, a tenth measuring point located on the non-drive end of the engine, and a measuring point located on the air compressor. The eleventh measuring point, the twelfth measuring point located at the input shaft end of the first-stage gearbox, the thirteenth measuring point located at the first end of the wheel axle of the first-stage gearbox, the fourteenth measuring point located at the second end of the wheel axle of the first-stage gearbox, the fifteenth measuring point located at the axle box of the second-stage bogie, the sixteenth measuring point located at the axle box of the second-stage bogie, the seventeenth measuring point located at the torque converter, the eighteenth measuring point located at the axle box of the first-stage bogie, the nineteenth measuring point located at the axle box of the first-stage bogie, the twentieth measuring point located at the axle box of the first-stage bogie, the twenty-first measuring point located at the axle box of the first-stage bogie, and the twenty-second measuring point located at the transfer case.

[0017] As a further improvement to the above technical solution, the front-end acquisition boxes are connected in a serial communication topology.

[0018] According to another aspect of the present invention, a safe operation and maintenance method for a catenary maintenance vehicle is also provided, comprising the aforementioned safe operation and maintenance system for the catenary maintenance vehicle, wherein the operation and maintenance method includes: S1. Data preprocessing; S2. Temperature alarm determination; S3. Time-domain feature alarm determination; S4. Fault characteristic frequency amplitude alarm determination: Based on the component type corresponding to the monitoring point, call the corresponding differentiated diagnostic strategy; S5. Output of integrated alarm and maintenance measures.

[0019] The present invention has the following beneficial effects: This safety operation and maintenance system utilizes a comprehensive deployment of multiple temperature and vibration composite sensors in the data acquisition system to fully monitor the core components of the entire travel transmission system of the overhead contact line maintenance vehicle. It achieves complete status awareness from the power source to the terminal actuators. Compared to existing technologies that only target localized monitoring of bogie bearings or axle boxes, this system significantly expands its monitoring coverage, enabling timely detection of early-stage faults in any link of the transmission chain. This avoids cascading failures caused by a single component malfunction. The signal transmission system facilitates signal conversion and transmission, while the front-end acquisition box allows for even distribution and redundancy of measurement points, simplifying batch installation, maintenance, and rapid fault diagnosis. Simplified wiring and adaptation to confined installation spaces simplify vehicle wiring and improve reliability. The data acquisition system provides ample input data for the intelligent diagnostic system, while the differentiated strategy of the intelligent diagnostic model ensures effective extraction and accurate identification of fault characteristics in different components, resulting in effective fault judgments. This differentiated diagnostic approach allows each component to achieve optimal diagnostic sensitivity and specificity under its unique operating conditions, significantly reducing false alarm and false negative rates. Compared to existing homogeneous solutions that use the same diagnostic algorithm for different components, this system demonstrates significantly improved diagnostic accuracy. Experimental verification shows a fault detection rate exceeding 95% and a false alarm rate controlled below 5%.

[0020] In addition to the objectives, features, and advantages described above, the present invention has other objectives, features, and advantages. The invention will now be described in further detail with reference to the figures. Attached Figure Description

[0021] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a simplified system connection diagram of a preferred embodiment of the present invention; Figure 2 This is a simplified structural diagram of the temperature and vibration composite sensor distribution according to a preferred embodiment of the present invention.

[0022] Legend: 1. First measuring point; 2. Second measuring point; 3. Third measuring point; 4. Fourth measuring point; 5. Fifth measuring point; 6. Sixth measuring point; 7. Seventh measuring point; 8. Eighth measuring point; 9. Ninth measuring point; 10. Tenth measuring point; 11. Eleventh measuring point; 12. Twelfth measuring point; 13. Thirteenth measuring point; 14. Fourteenth measuring point; 15. Fifteenth measuring point; 16. Sixteenth measuring point; 17. Seventeenth measuring point; 18. Eighteenth measuring point; 19. Nineteenth measuring point; 20. Twentieth measuring point; 21. Twenty-first measuring point; 22. Twenty-second measuring point; 23. On-board host; 24. Front-end data acquisition box; 25. Axle box; 26. First-stage gearbox; 27. Second-stage gearbox; 28. Torque converter; 29. ​​Engine; 30. Air compressor; 31. Transfer case. Detailed Implementation

[0023] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, the present invention can be implemented in many different ways as defined and covered below.

[0024] Figure 1 This is a simplified system connection diagram of a preferred embodiment of the present invention; Figure 2 This is a simplified structural diagram of the temperature and vibration composite sensor distribution according to a preferred embodiment of the present invention.

[0025] like Figure 1 and Figure 2 As shown, the safety operation and maintenance system for the overhead contact line maintenance vehicle in this embodiment includes: The data acquisition system includes multiple temperature and vibration composite sensors. The temperature and vibration composite sensors integrate temperature-sensitive elements and vibration-sensitive elements to simultaneously output temperature and vibration signals. The temperature and vibration composite sensors are distributed and installed on key components of the overhead contact line maintenance vehicle's travel transmission system to collect temperature and vibration acceleration signals at each monitoring point in real time. The signal transmission system includes multiple front-end acquisition boxes 24, each of which is connected to multiple temperature and vibration composite sensors. The system is used to perform preliminary signal conditioning and format conversion on the analog signals output by the temperature and vibration composite sensors and then transmit them to the vehicle host 23. The intelligent diagnostic system includes an intelligent diagnostic model mounted on the vehicle host 23. The intelligent diagnostic system is used to receive temperature and vibration digital signals transmitted by each front acquisition box 24, perform real-time diagnosis of the operating status of the components, identify the fault type, and provide on-site handling measures based on the diagnostic results. The intelligent diagnostic model matches the corresponding differentiated diagnostic strategy according to the component type corresponding to each monitoring point.

[0026] Among them, the vehicle-mounted host 23 is the vehicle-mounted host 23 of the overhead contact line maintenance vehicle. Through the built-in intelligent diagnostic model of the vehicle-mounted host 23, the entire process of data acquisition, signal processing, feature extraction, fault diagnosis and operation and maintenance measure suggestions can be automated. There is no need to rely on human experience and regular inspections. This not only greatly reduces the workload and technical threshold of operation and maintenance personnel, but also eliminates the risk of missed detection and misjudgment caused by human factors.

[0027] Understandably, this safety operation and maintenance system comprehensively monitors all core components of the overhead contact line maintenance vehicle's travel transmission system through a comprehensive deployment of multiple temperature and vibration composite sensors in the data acquisition system. This achieves complete status awareness from the power source to the terminal actuator. Compared to existing technologies that only target localized monitoring of bogie bearings or axle boxes 25, this system significantly expands its monitoring coverage, enabling timely detection of early-stage faults in any link of the transmission chain and preventing cascading failures caused by a single component malfunction. The signal transmission system facilitates signal conversion and transmission, while the front-end acquisition box 24 allows for even distribution and redundancy of measurement points, facilitating batch installation, maintenance, and rapid fault diagnosis. The system significantly simplifies wiring and adapts to confined installation spaces, simplifying vehicle wiring and improving reliability. The data acquisition system provides ample input data for the intelligent diagnostic system, while the differentiated strategy of the intelligent diagnostic model ensures effective extraction and accurate identification of fault characteristics in different components, resulting in effective fault judgments. This differentiated diagnostic approach allows each component to achieve optimal diagnostic sensitivity and specificity under its unique operating conditions, significantly reducing false alarm and false negative rates. Compared to existing homogeneous solutions that use the same diagnostic algorithm for different components, this system demonstrates significantly improved diagnostic accuracy. Experimental verification shows a fault detection rate exceeding 95% and a false alarm rate controlled below 5%.

[0028] In some preferred embodiments, the temperature and vibration composite sensor adopts an industrial-grade integrated temperature and vibration sensor, with a temperature measurement range covering -40℃ to +125℃ and a vibration measurement frequency range covering 0.5Hz to 10kHz, meeting the full-condition monitoring needs of the overhead contact line maintenance vehicle. The temperature and vibration composite sensor is installed by machining M12 threaded holes on the surface of each monitoring component, and the sensor is directly fixed to the monitoring component via the M12 thread, ensuring a rigid connection between the sensor and the monitoring component and the effectiveness of signal transmission. The front-end acquisition box 24 has an AD sampling resolution of no less than 16 bits and a sampling frequency of no less than 25.6kHz, ensuring accurate acquisition of high-frequency vibration signals. (Vehicle-mounted...) The intelligent diagnostic model built into the host 23 adopts an edge computing architecture, completing the entire process of data acquisition, feature extraction, and fault diagnosis at the vehicle end, achieving real-time diagnosis and alarm without relying on a ground server. The system also includes a human-machine interface display terminal connected to the vehicle host 23, which displays the temperature, vibration data, and abnormal fault information of each measuring point in real time, and graphically displays the health status of seven components. It also includes an event recorder connected to the vehicle host 23, which records the temperature, vibration data, and abnormal fault information on the bus in real time, and classifies and stores the data and faults according to system events, supporting data export for trend analysis and predictive fault analysis.

[0029] In some preferred embodiments, key components include engine 29, torque converter 28, transfer case 31, air compressor 30, secondary gearbox 27, primary gearbox 26, and wheelsets. It is understood that by comprehensively monitoring the seven core components (engine 29, torque converter 28, transfer case 31, air compressor 30, secondary gearbox 27, primary gearbox 26, and wheelsets) of the entire overhead contact line maintenance vehicle's travel transmission system, complete state perception from the power source to the terminal actuator is achieved. Compared to existing technologies that only target partial monitoring of bogie bearings or axle boxes 25, this system significantly expands its monitoring coverage, enabling timely detection of early fault hazards in any link of the transmission chain and preventing chain failures caused by a single component failure. In particular, it provides a systematic monitoring solution for components rarely addressed in existing technologies, such as transfer case 31 and air compressor 30, filling a technological gap. In some preferred embodiments, the intelligent diagnostic model includes a temperature alarm module, a time-domain feature alarm module, and a fault feature frequency amplitude alarm module. The temperature alarm module triggers an alarm by comparing the temperature and temperature difference values ​​at each monitoring point with preset thresholds. The time-domain feature alarm module triggers an alarm by comparing the root mean square value and kurtosis index of the vibration signals at each monitoring point with preset thresholds. The fault feature frequency amplitude alarm module constructs fault feature frequency extraction strategies for different components based on their fault mechanisms and structural parameters. The most severe alarm result determines the maintenance measures. It is understood that this system establishes a multi-level alarm system integrating temperature alarm, time-domain feature alarm, and fault feature frequency amplitude alarm. Temperature alarm provides macro-trend judgment, time-domain feature alarm provides signal statistical characteristic judgment, and fault feature frequency amplitude alarm provides accurate frequency domain fault mode identification. The three alarms evaluate the component status from different dimensions, and the most severe result determines the maintenance measures, forming a multi-level, multi-dimensional security protection. The multi-level alarm fusion mechanism effectively avoids the one-sidedness and uncertainty of a single alarm mechanism, providing maintenance personnel with sufficient and reliable decision-making basis.

[0030] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategies for engine 29: The original vibration signal is preprocessed to remove combustion impact interference. A dual sparse dictionary is constructed by combining crankshaft connecting rod bearing structural parameters, valve train and piston reciprocating motion parameters. Combustion noise interference is stripped away using a sparse decomposition algorithm, and independent fault frequencies of bearing wear and valve train abnormalities are separated and determined based on amplitude thresholds. It is understandable that engine 29 was selected as the vehicle's sole power source, bearing multiple loads including travel, operation, and aerodynamics. Low-speed full-load conditions are prone to wear and thermal failures, which can directly lead to vehicle paralysis. The diagnostic challenges are strong coupling interference from combustion and multiple loads, with fault characteristics masked by noise; large temperature drift under varying operating conditions; and fixed thresholds. The false alarm rate is high, and early bearing wear is difficult to identify; the vibration signal of engine 29 contains strong noise interference such as combustion impact and fan rotation; firstly, the original vibration signal is preprocessed to remove strong interference noise and purify the mechanical fault characteristic signal; combined with the structural parameters of crankshaft connecting rod bearing, valve train and piston reciprocating motion parameters, a dual sparse dictionary adapted to the power mechanism is constructed; the combustion noise interference is removed by sparse decomposition algorithm, and the independent fault frequencies of bearing wear and valve train abnormality are separated; based on the amplitude threshold, the early structural faults of the core transmission and motion mechanism of engine 29 are accurately determined, and finally the fault judgment result of engine 29 is obtained by comprehensive analysis.

[0031] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategy for the torque converter 28: The original vibration signal is preprocessed to filter hydraulic damping noise. A dual sparse dictionary is constructed based on the structural parameters of the supporting bearing, gear parameters, and speed difference characteristics. Specific fault frequencies of bearing wear eccentricity, gear cavitation, and internal slippage are extracted through sparse decomposition and identified by amplitude determination. It is understandable that the torque converter 28 is selected based on its core torque buffer amplification function. Frequent low-speed slippage results in high losses, and internal faults easily generate debris that contaminates the entire transmission system, with no direct inspection conditions. The diagnostic challenge lies in the weak early fault signals caused by oil vibration absorption. The boundary mode between normal slippage and fault slippage conditions... The problem is that oil temperature and pressure are easily affected by low-temperature gas generation, leading to numerous false alarms. Torque converter 28 suffers from hydraulic medium vibration absorption and extremely low signal-to-noise ratio of fault signals. The pre-processing stage focuses on filtering hydraulic damping noise and purifying weak and effective fault signals. Based on the structural parameters of the supporting bearing, gear parameters, and speed difference characteristics, a dual sparse dictionary is constructed. By breaking through the hydraulic vibration absorption interference through sparse decomposition, the specific fault frequencies of bearing wear eccentricity, gear cavitation, and internal slippage are effectively extracted. Through precise amplitude determination, the early structural faults of torque converter 28, which are highly concealed, are identified, avoiding the risks of oil circuit contamination and transmission failure.

[0032] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategy for the transfer case 31: The original vibration signal is preprocessed to filter interference noise caused by intermittent loading and load switching of the air compressor 30. A sparse dictionary library adapted to multiple working conditions is constructed based on the multi-support shaft bearing structure and the dual-branch gear transmission parameters for travel and power take-off. The mixed vibration signal is decomposed through sparse decomposition to separate the independent fault characteristic frequencies of the multi-shaft bearings and power distribution gears, and amplitude indicators are extracted to complete threshold judgment. It is understandable that the transfer case 31 is selected as a power distribution hub, simultaneously outputting power to both the travel and air compressor 30. It has multiple internal shafts and meshing structures, and a fault can simultaneously cut off the air source power for both the travel and operation. The diagnostic challenge lies in the dual-load power coupling and the mixed vibration signals. The source of the fault is difficult to locate; intermittent loading of the air compressor 30 causes fluctuations in operating conditions, which can easily generate false fault alarms; the transfer case 31 has the characteristics of dual load coupling and frequent fluctuations in operating conditions. First, the original vibration signal is preprocessed to filter out the interference noise caused by the intermittent loading and load switching of the air compressor 30; based on the bearing structure of the multi-support shaft system and the transmission parameters of the dual-branch gear transmission of the travel and power take-off, a sparse dictionary library adapted to multiple operating conditions is constructed. By sparsely decomposing and disassembling the mixed vibration signal, the independent fault characteristic frequencies of the multi-shaft bearings and power distribution gears are separated, and the amplitude index is extracted to complete the threshold judgment, so as to accurately locate coupled faults such as bearing loosening and wear, abnormal gear meshing, and power distribution imbalance.

[0033] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategies for the air compressor 30: The original vibration signal is normalized and noise-reduced to filter out interference noise caused by belt vibration, no-load start-up, and air pressure switching. A dual sparse dictionary is constructed based on the structural parameters of the crankshaft support bearing of the air compressor 30, the reciprocating motion characteristics of the crankshaft connecting rod, and the working parameters of the piston cylinder and intake / exhaust valve assembly. An adaptive matching tracking algorithm separates rotational fault components from reciprocating compression mechanism fault components, and the judgment is made in conjunction with the total cylinder pressure and real-time load adaptive threshold. It is understandable that the selection criteria for the air compressor 30 are core components for operational safety, ensuring braking, platform locking, and pantograph pneumatic functions. Faults directly trigger safety interlocks, endangering high-altitude work safety. The diagnostic challenges include intermittent start-stop non-steady-state operation, rendering conventional spectrum diagnostics ineffective; tunnel moisture easily causes oil emulsification and sensor scaling; and the air tank buffer masks early minor faults. The core moving components of the air compressor 30 include the crankshaft support bearing, crankshaft connecting rod mechanism, two- or three-cylinder piston cylinders, and intake / exhaust valve assembly, relying on the diesel engine transfer case. The 31 V-belt indirect drive provides pressurized air to the vehicle's braking and operating mechanisms. The equipment is characterized by frequent starts and stops, intermittent compression, and strong belt-driven disturbances. First, the original vibration signal is normalized and noise-reduced, filtering out irrelevant interference noise from belt vibration, no-load starts, and air pressure switching. Based on the crankshaft support bearing structural parameters of the air compressor 30, the reciprocating motion characteristics of the crankshaft connecting rod, and the working parameters of the piston cylinder and intake / exhaust valve assembly, a dual sparse dictionary adapted to the piston air compressor 30 is constructed. An adaptive matching tracking algorithm is used to perform sparse decomposition of the vibration signal, effectively separating rotational fault components such as bearing wear and crankshaft runout from reciprocating compression mechanism fault components such as piston wear, connecting rod loosening, and valve assembly jamming. The equipment's specific fault characteristic frequency points are purified and amplitude indicators are quantified. Combined with the total cylinder pressure and real-time load adaptive threshold, accurate comparison is achieved, effectively identifying early and minor structural faults in the piston air compressor 30, ensuring stable air supply and operational safety for the entire vehicle's pneumatic system.

[0034] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategy for the secondary gearbox 27: The original vibration signal undergoes high-frequency filtering and signal smoothing. A dual sparse dictionary adapted to high-speed operating conditions is constructed based on the structural parameters of high-speed rolling bearings and the meshing transmission parameters of small-module high-speed gears. An adaptive sparse decomposition algorithm reconstructs the independent fault components of bearing fatigue damage and high-speed gear meshing anomalies, and the judgment is made in conjunction with real-time speed and adaptive load thresholds. It is understandable that the secondary gearbox operates at high speed and experiences frequent alternating loads, resulting in a high risk of pitting and tooth breakage in the gear bearings. Fault debris can easily damage the rear secondary gearbox 27, causing significant cascading damage. Diagnostic challenges include the easy superposition of high-frequency vibration with vehicle body resonance; wheel-rail impact noise masking early fault signals; limited installation space; insufficient monitoring points; and the high-speed operation of the secondary gearbox 27, the high-speed end of the vehicle transmission, with strong high-frequency vibration interference, and faults primarily characterized by high-frequency fatigue damage. Given the characteristics of high-speed operation, vehicle body resonance, and severe high-frequency impact interference from wheels and rails, the preprocessing stage employs high-frequency filtering and signal smoothing to effectively eliminate redundant interference signals under high-speed conditions and improve the purity of fault signals. Based on the structural parameters of the 27 high-speed rolling bearings in the secondary gearbox and the meshing transmission parameters of small-module high-speed gears, a dual sparse dictionary adapted to high-speed conditions is built. Through an adaptive sparse decomposition algorithm, complex composite high-frequency vibration signals are decomposed, and independent fault components of bearing fatigue damage and high-speed gear meshing abnormalities are accurately reconstructed. The inherent fault characteristic frequency points of the high-speed transmission system are purified and the amplitude index is quantified. Combined with real-time speed and load adaptive thresholds, comparison and judgment are completed. This can accurately identify high-frequency fatigue high-risk structural faults such as bearing spalling, gear pitting, micro-tooth breakage, and abnormal meshing impact under high-speed conditions, avoiding running faults caused by high-speed transmission failure.

[0035] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategy for the first-stage gearbox 26: The original vibration signal is enhanced by low-frequency effective signal processing and random impact noise is filtered out. Based on the mechanical parameters of heavy-duty roller bearings and the meshing characteristics of large-module low-speed transmission gears, a dual sparse dictionary adapted to heavy-duty low-speed conditions is constructed. The independent fault frequency components of low-speed bearings and heavy-duty gears are reconstructed through an adaptive matching pursuit sparse decomposition algorithm, and the judgment is made in combination with the real-time output torque adaptive dynamic threshold. It is understandable that the selection basis of the first-stage gearbox 26 is to directly drive the wheelset, which bears the maximum impact and traction torque. Faults can easily lead to vehicle lock-up, circuit blockage, and extremely difficult rescue in tunnels. The diagnostic difficulties are that the low-speed fault pulse energy is extremely low and the signal is weak; the load fluctuation is extremely large, and the fixed threshold is completely ineffective; the environment under the vehicle is harsh, the sensor protection is difficult, and early cracks have no temperature rise characteristics; the first-stage gearbox 26 is the low-speed heavy-duty end of the vehicle transmission, which bears a large torque traction load, and the fault signal is weak. The interference under heavy load conditions is prominent. In the preprocessing stage, the focus is on enhancing the effective low-frequency fault signals and filtering out random impact noise generated by slope traction, curved driving, and heavy load start-up, so as to solve the problem of low signal-to-noise ratio of low-speed fault signals and easy masking by the operating conditions. Based on the mechanical parameters of the 26 heavy-load roller bearings of the first-stage gearbox and the meshing characteristics of the large-module low-speed transmission gears, a dedicated dual sparse dictionary adapted to heavy load low-speed conditions is constructed. The weak hidden fault features are amplified by the adaptive matching tracking sparse decomposition algorithm, effectively reconstructing the independent fault frequency components of low-speed bearings and heavy-load gears, accurately purifying the inherent fault frequency points under low-speed heavy load conditions and quantifying the amplitude index. Combined with the real-time output torque adaptive dynamic threshold to complete the accurate comparison, it can effectively identify early fatal structural faults such as bearing indentation, tooth root micro-cracks, gear plastic deformation, and heavy load meshing damage that are difficult to capture by conventional diagnostic methods, and avoid the risk of low-speed heavy load transmission jamming.

[0036] In some preferred embodiments, the fault characteristic frequency amplitude alarm module employs the following diagnostic strategy for wheelsets: It performs detrending processing and digital filtering on the original vibration signal to eliminate rail gap impact and vehicle body resonance interference noise; it constructs dedicated dual sparse dictionaries for bearings and treads based on the wheelset bearing geometry parameters, real-time wheel speed, and tread wear mechanism; it separates bearing fault frequencies and tread damage characteristic frequencies using an adaptive matching pursuit algorithm; and it makes judgments based on adaptive thresholds for vehicle speed conditions. It is understood that wheelset selection is based on the vehicle's running load and terminal execution core, directly bearing rail impact, braking, and curve loads, posing a derailment safety risk; wheelset damage can have a reverse impact on the transmission system, causing secondary faults in the gearbox and bearings, making them key monitoring components for driving safety; the diagnostic challenge lies in the large random impact noise from the rail surface, which easily masks early, weak bearing faults; and the tread... Damaged vibration characteristics are highly similar to those under normal operating conditions, making them difficult to distinguish. Large fluctuations in vehicle speed and harsh environmental interference further complicate the early identification of latent cracks in wheelsets. First, vibration signals from the entire wheelset travel are collected. Basic preprocessing, including detrending and digital filtering, is performed to accurately remove strong interference noise such as rail gap impact and vehicle body resonance, improving the purity of the original signal. Based on the geometric parameters of the wheelset bearings, real-time wheel speed, and tread wear mechanism, dedicated dual sparse dictionaries for bearings and treads are constructed. An adaptive matching pursuit algorithm is used to perform sparse decomposition of the signal, reconstructing the effective fault components and completely separating the bearing fault frequency and tread damage characteristic frequency coupled in the composite vibration. The core fault frequencies are further refined, and the corresponding amplitude indicators are quantified. Combined with adaptive thresholds for vehicle speed conditions, accurate comparison is achieved, ultimately enabling the classification, identification, and status determination of early latent faults in the wheelset.

[0037] Understandably, for each of the seven different components, specific diagnostic strategies and fault feature frequency extraction methods were developed based on their respective fault mechanisms, signal characteristics, and operating conditions. For example, the engine diagnostic strategy (29) uses a dual sparse dictionary to isolate combustion noise to address strong combustion impact interference; the torque converter diagnostic strategy (28) uses sparse decomposition to overcome hydraulic damping interference to address hydraulic vibration absorption; the transfer case diagnostic strategy (31) constructs a multi-condition sparse dictionary library to address dual-load coupling issues; the air compressor diagnostic strategy (30) uses adaptive matching tracking to separate rotational and reciprocating compression faults to address intermittent start-stop issues; the secondary gearbox diagnostic strategy (27) uses high-frequency filtering and sparse decomposition to address high-speed, high-frequency interference; and the primary gearbox diagnostic strategy (26) addresses low-speed... For heavy-load weak signals, signal enhancement and adaptive matching tracking are used; the wheelset diagnostic strategy adopts detrending processing and dual-dictionary separation of bearing and tread faults for rail impact noise; although the diagnostic strategies of each key component are independent, they all follow a unified technical framework of signal preprocessing → dual sparse dictionary construction → sparse decomposition / adaptive matching tracking → amplitude threshold determination. At the same time, differentiated adjustments are made for the unique fault mechanisms and operating conditions of each component. The differentiated implementation under the unified framework not only ensures the systematicness and reproducibility of the technical solution, but also ensures the pertinence and accuracy of the diagnosis of each component, so that each component can obtain the optimal diagnostic sensitivity and specificity under its unique operating conditions, which greatly reduces the false alarm rate and the missed alarm rate. In some preferred embodiments, the data acquisition system includes a first measuring point 1 located at the input shaft end of the secondary gearbox 27, a second measuring point 2 located at the output shaft end of the secondary gearbox 27, a third measuring point 3 located at the first end of the wheel axle of the secondary gearbox 27, a fourth measuring point 4 located at the second end of the wheel axle of the secondary gearbox 27, a fifth measuring point 5 located at the axle box 25 of the second-end bogie, a sixth measuring point 6 located at the axle box 25 of the second-end bogie, a seventh measuring point 7 located on the main beam, an eighth measuring point 8 located on top of the torque converter 28, a ninth measuring point 9 located on top of the drive end of the engine 29, a tenth measuring point 10 located on the non-drive end of the engine 29, an eleventh measuring point 11 located on the air compressor 30, and a third measuring point 10 located on the output shaft end of the secondary gearbox 27. The twelfth measuring point 12 at the input shaft end of the first-stage gearbox 26, the thirteenth measuring point 13 at the first end of the wheel axle of the first-stage gearbox 26, the fourteenth measuring point 14 at the second end of the wheel axle of the first-stage gearbox 26, the fifteenth measuring point 15 at the bogie axle box 25 at end II, the sixteenth measuring point 16 at the bogie axle box 25 at end II, the seventeenth measuring point 17 at the torque converter 28, the eighteenth measuring point 18 at the bogie axle box 25 at end I, the nineteenth measuring point 19 at the bogie axle box 25 at end I, the twentieth measuring point 20 at the bogie axle box 25 at end I, the twenty-first measuring point 21 at the bogie axle box 25 at end I, and the twenty-second measuring point 22 at the transfer case 31; The comprehensive layout of 22 sensors provides ample input data for the intelligent diagnostic model, while the differentiated strategy of the intelligent diagnostic model ensures the effective extraction and accurate identification of fault characteristics of different components. Without a comprehensive sensor layout, the diagnostic model lacks a data foundation; without a differentiated diagnostic strategy, comprehensive sensor data cannot be transformed into effective fault judgment results.

[0038] In some preferred embodiments, the front-end acquisition boxes 24 are connected in a series communication topology. Taking an 8-terminal front-end acquisition box 24 as an example, it can be understood that the standardized interface design of the 8-terminal front-end acquisition box 24 provides the hardware foundation for the series communication topology, while the series communication topology fully utilizes the power supply and data forwarding functions of terminals 7 and 8 in the 8-terminal structure. Together, they simplify the vehicle wiring and improve reliability. Specifically, each front-end acquisition box 24 distributes the signal lines of the 22 temperature and vibration composite sensors evenly to the four front-end acquisition boxes 24 according to the principle of proximity sensor access. Terminals 1-6 of each front-end acquisition box 24 are connected to the temperature signal line and vibration signal line of the corresponding sensor, respectively. Terminal 7 is connected to the vehicle's 24V DC power supply to power the acquisition circuit board, and simultaneously communicates via Ethernet. The system transmits the collected temperature and vibration data externally; terminal 8 outputs 24V DC voltage and simultaneously receives data from other front-end acquisition boxes 24; the front-end acquisition boxes 24 communicate in series, with terminal 8 of the first acquisition box connected to terminal 7 of the second acquisition box, terminal 8 of the second acquisition box connected to terminal 7 of the third acquisition box, and so on, with terminal 8 of the last acquisition box connected to the vehicle host 23; data is transmitted from upstream acquisition boxes to downstream, and finally aggregated to the vehicle host 23; this system comprehensively monitors 22 measuring points of seven components: engine 29, torque converter 28, transfer case 31, air compressor 30, secondary gearbox 27, primary gearbox 26, and wheelsets, achieving online status perception of the entire drivetrain system without any monitoring blind spots.

[0039] On the other hand, a preferred embodiment of the present invention also provides a safe operation and maintenance method for a catenary maintenance vehicle, which is applied as described in the above-mentioned safe operation and maintenance system for a catenary maintenance vehicle. The operation and maintenance method includes: S1. Data preprocessing: Perform preprocessing operations such as filtering, noise reduction, and format conversion on the received raw temperature and vibration signals; S2. Temperature alarm determination: Calculate the temperature value and temperature difference value (temperature of measuring point minus ambient temperature) of each measuring point and compare them with the preset temperature threshold and temperature difference threshold; if the threshold is exceeded, an electrical temperature alarm is triggered. S3. Time-domain feature alarm determination: Calculate the root mean square value and kurtosis index of the vibration signal at each measuring point and compare them with the preset threshold; if the threshold is exceeded, a time-domain feature alarm is triggered. S4. Fault Characteristic Frequency Amplitude Alarm Determination: Based on the component type corresponding to the monitoring point, the appropriate differentiated diagnostic strategy is invoked. For engine 29 test point, the engine 29 diagnostic strategy is executed, signal preprocessing removes combustion impact interference, constructs a dual sparse dictionary, sparsely decomposes and strips combustion noise, separates independent fault frequencies of bearing wear and valve train abnormalities, and determines the amplitude threshold. For torque converter 28 test point, the torque converter 28 diagnostic strategy is executed, filtering hydraulic damping noise, constructing a dual sparse dictionary, sparsely decomposes and extracts specific fault frequencies of bearing wear eccentricity, gear cavitation, and internal slippage, and determines the amplitude. For transfer case 31 test point, the transfer case 31 diagnostic strategy is executed, filtering intermittent loading interference from air compressor 30, constructing a multi-condition sparse dictionary library, sparsely decomposes and disassembles mixed signals, separates independent fault characteristic frequencies of multi-shaft bearings and power distribution gears, and determines the amplitude. For air compressor 30 test point, the air compressor 30 diagnostic strategy is executed, with normalization and noise reduction processing, constructing a dual sparse dictionary, and adaptive... The system performs matching and tracking to separate rotational and reciprocating compression fault components, and then uses the total air cylinder pressure and real-time load to adaptively determine thresholds. For the second-stage gearbox 27 measuring point, the system implements the second-stage gearbox 27 diagnostic strategy, including high-frequency filtering and signal smoothing, and constructs a dual sparse dictionary adapted to high-speed conditions. It then adaptively sparsely decomposes and reconstructs independent fault components of bearing fatigue damage and high-speed gear meshing abnormalities, and uses the real-time speed and load to adaptively determine thresholds. For the first-stage gearbox 26 measuring point, the system implements the first-stage gearbox 26 diagnostic strategy, enhancing low-frequency effective signals, filtering out random impact noise, and constructing a dual sparse dictionary adapted to heavy-load low-speed conditions. It then adaptively matches and tracks to amplify weak, latent fault features, and uses the real-time output torque to adaptively determine dynamic thresholds. For the wheelset and axle box 25 measuring point, the system implements the wheelset diagnostic strategy, including detrending processing and digital filtering, and constructs a dual sparse dictionary specifically for bearings and treads. It then adaptively matches and tracks to separate bearing fault frequencies and tread damage characteristic frequencies, and uses the vehicle speed to adaptively determine thresholds. S5. Comprehensive alarm and maintenance measures output; integrates three results: temperature alarm, time-domain characteristic alarm, and fault characteristic frequency amplitude alarm. The maintenance measures are determined based on the most severe alarm result. The diagnostic results and on-site handling suggestions are output to the maintenance personnel through the human-machine interaction display terminal.

[0040] In the description of this invention, it should be noted that the terms "upper", "lower", "front", "rear", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0041] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0042] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A safety operation and maintenance system for overhead contact line maintenance vehicles, characterized in that, include: The data acquisition system includes multiple temperature and vibration composite sensors. The temperature and vibration composite sensors integrate temperature-sensitive elements and vibration-sensitive elements to simultaneously output temperature signals and vibration signals. The temperature and vibration composite sensors are distributed and installed on key components of the overhead contact line maintenance vehicle's travel transmission system to collect temperature and vibration acceleration signals at each monitoring point in real time. The signal transmission system includes multiple front-end acquisition boxes (24), each of which is connected to multiple temperature and vibration composite sensors, and is used to transmit the analog signals output by the temperature and vibration composite sensors to the vehicle host (23) after preliminary signal conditioning and format conversion. The intelligent diagnostic system includes an intelligent diagnostic model mounted on the vehicle host (23). The intelligent diagnostic system is used to receive temperature digital signals and vibration digital signals transmitted by each of the front acquisition boxes (24) to perform real-time diagnosis of the operating status of the components, identify the fault type, and provide on-site handling measures based on the diagnostic results. The intelligent diagnostic model matches the corresponding differentiated diagnostic strategy according to the component type corresponding to each monitoring point.

2. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 1, characterized in that, The key components include an engine (29), a torque converter (28), a transfer case (31), an air compressor (30), a primary gearbox (26), a secondary gearbox (27), and wheelsets.

3. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 2, characterized in that, The intelligent diagnostic model includes a temperature alarm module, a time-domain feature alarm module, and a fault feature frequency amplitude alarm module. The temperature alarm module generates an alarm by comparing the temperature value and temperature difference value of each monitoring point with a preset threshold. The time-domain feature alarm module generates an alarm by comparing the root mean square value and kurtosis index of the vibration signal at each monitoring point with a preset threshold. The fault feature frequency amplitude alarm module generates an alarm by constructing a fault feature frequency extraction strategy for different components based on their fault mechanisms and structural parameters. The most severe alarm result determines the maintenance measures.

4. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the engine (29): The original vibration signal is preprocessed to remove combustion impact interference. A dual sparse dictionary is constructed by combining the structural parameters of the crankshaft connecting rod bearing, the valve train mechanism and the reciprocating motion parameters of the piston. Combustion noise interference is removed by sparse decomposition algorithm, and independent fault frequencies of bearing wear and valve train mechanism abnormalities are separated and judged based on amplitude threshold.

5. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the torque converter (28): The original vibration signal is preprocessed to filter out hydraulic damping clutter. A dual sparse dictionary is constructed based on the structural parameters of the supporting bearing, gear parameters, and speed difference characteristics. The specific fault frequencies of bearing wear eccentricity, gear cavitation, and internal slippage are extracted through sparse decomposition and identified by amplitude determination.

6. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the transfer case (31): The original vibration signal is preprocessed to filter the interference noise caused by the intermittent loading and load switching of the air compressor (30). A sparse dictionary library adapted to multiple working conditions is constructed based on the bearing structure of the multi-support shaft system and the transmission parameters of the dual-branch gear for travel and power take-off. The mixed vibration signal is decomposed by sparse decomposition, the independent fault characteristic frequencies of the multi-shaft bearing and the power distribution gear are separated, and the amplitude index is extracted to complete the threshold judgment.

7. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the air compressor (30): The original vibration signal is normalized and noise is reduced to filter out interference noise caused by belt vibration, no-load start-up and air pressure switching. Based on the structural parameters of the crankshaft support bearing of the air compressor (30), the reciprocating motion characteristics of the crankshaft connecting rod and the working parameters of the piston cylinder and intake and exhaust valve group, a dual sparse dictionary is constructed. The rotational fault component and the reciprocating compression mechanism fault component are separated by the adaptive matching tracking algorithm, and the judgment is made in combination with the total air cylinder pressure and the real-time load adaptive threshold.

8. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the secondary gearbox (27): The original vibration signal is subjected to high-frequency filtering and signal smoothing. A dual sparse dictionary adapted to high-speed working conditions is constructed based on the structural parameters of high-speed rolling bearings and the meshing transmission parameters of small-module high-speed gears. The independent fault components of bearing fatigue damage and high-speed gear meshing abnormality are reconstructed by an adaptive sparse decomposition algorithm and judged in combination with real-time speed and load adaptive threshold.

9. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module adopts the following diagnostic strategy for the first-stage gearbox (26): The original vibration signal is enhanced by low-frequency effective signal processing and random impact noise is filtered out. Based on the mechanical parameters of heavy-duty roller bearings and the meshing characteristics of large-module low-speed transmission gears, a dual sparse dictionary adapted to heavy-duty low-speed working conditions is constructed. The independent fault frequency components of low-speed bearings and heavy-duty gears are reconstructed by an adaptive matching pursuit sparse decomposition algorithm, and the judgment is made in combination with the real-time output torque adaptive dynamic threshold.

10. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 3, characterized in that, The fault characteristic frequency amplitude alarm module employs the following diagnostic strategy for wheelsets: The original vibration signal is detrended and digitally filtered to remove noise from rail gap impact and car body resonance. Based on the geometric parameters of the wheelset bearing, the real-time speed of the wheel, and the tread wear mechanism, a dual sparse dictionary is constructed for the bearing and the tread respectively. The bearing fault frequency and the tread damage characteristic frequency are separated by an adaptive matching tracking algorithm and judged by adaptive threshold of vehicle speed condition.

11. The safety operation and maintenance system for overhead contact line maintenance vehicles according to any one of claims 2-10, characterized in that, The data acquisition system includes a first measuring point (1) located at the input shaft end of the secondary gearbox (27), a second measuring point (2) located at the output shaft end of the secondary gearbox (27), a third measuring point (3) located at the first end of the wheel axle of the secondary gearbox (27), a fourth measuring point (4) located at the second end of the wheel axle of the secondary gearbox (27), a fifth measuring point (5) located at the axle box (25) of the second-end bogie, a sixth measuring point (6) located at the axle box (25) of the second-end bogie, a seventh measuring point (7) located on the main beam, an eighth measuring point (8) located on top of the torque converter (28), a ninth measuring point (9) located on top of the drive end of the engine (29), a tenth measuring point (10) located on the non-drive end of the engine (29), an eleventh measuring point (11) located on the air compressor (30), and a measuring point located on the first-end gearbox. (26) The twelfth measuring point (12) at the input shaft end, the thirteenth measuring point (13) at the first end of the wheel axle of the first gearbox (26), the fourteenth measuring point (14) at the second end of the wheel axle of the first gearbox (26), the fifteenth measuring point (15) at the bogie axle box (25) at the second end, the sixteenth measuring point (16) at the bogie axle box (25) at the second end, the seventeenth measuring point (17) at the torque converter (28), the eighteenth measuring point (18) at the bogie axle box (25) at the first end, the nineteenth measuring point (19) at the bogie axle box (25) at the first end, the twentieth measuring point (20) at the bogie axle box (25) at the first end, the twenty-first measuring point (21) at the bogie axle box (25) at the first end, and the twenty-second measuring point (22) at the transfer case (31).

12. The safety operation and maintenance system for overhead contact line maintenance vehicles according to claim 1, characterized in that, Each of the aforementioned front-end acquisition boxes (24) is connected via a serial communication topology.

13. A safe operation and maintenance method for a catenary maintenance vehicle, characterized in that, The operation and maintenance method of the overhead contact line maintenance vehicle safety operation and maintenance system as described in any one of claims 3-10 includes: S1. Data preprocessing; S2. Temperature alarm determination; S3. Time-domain feature alarm determination; S4. Fault characteristic frequency amplitude alarm determination: Based on the component type corresponding to the monitoring point, call the corresponding differentiated diagnostic strategy. S5. Output of integrated alarm and maintenance measures.

Images