Railway freight car fault prediction method

Abstract

This invention provides a method for predicting railway freight car faults, belonging to the field of freight car monitoring technology. The method includes: constructing a freight car model within a host computer; setting gravity parameters on the model to ensure that the weight of each module in the model is the same as the weight of the corresponding parts of the freight car; detecting the operating status of the freight car; and adjusting the model's operation according to the operating status. When the model fails before the freight car, the host computer generates a fault report based on the model's operating status, providing an early warning for the freight car. This railway freight car fault prediction method provides timely early warnings of freight car faults through the model, preventing major freight car failures and ensuring stable freight car operation and railway transportation safety.

Description

Technical Field

[0001] This invention belongs to the field of freight car monitoring technology, and more specifically, relates to a method for predicting railway freight car faults. Background Technology

[0002] Currently, my country's railway freight car maintenance and repair has long adhered to the principle of prevention-oriented, periodic planned maintenance. While this provides a certain level of safety assurance, it has also exposed some drawbacks. Periodic maintenance lacks specificity, failing to differentiate between the specific technical conditions of each car, resulting in over-maintenance. Furthermore, the existing periodic maintenance time is too long, and the maintenance content and processes are complex, leading to wasted costs and numerous inconveniences for shunting, severely impacting transportation efficiency.

[0003] Although installing multiple sensors on freight cars can provide timely warnings of malfunctions, it cannot predict them. More importantly, most of the malfunctions of existing railway freight cars are discovered during maintenance, which means that if a malfunction occurs while the freight car is in motion, it will pose a significant threat to the safety of railway operations and personnel. Summary of the Invention

[0004] The purpose of this invention is to provide a method for predicting railway freight car malfunctions, aiming to solve the problem that freight car malfunctions cannot be predicted, posing a significant threat to railway operation safety and personnel safety.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is: to provide a method for predicting railway freight car faults, comprising:

[0006] A model of a truck is constructed in the host computer, and gravity parameters are set on the model so that the weight of each module in the model is the same as the weight of the corresponding part of the truck.

[0007] The operating status of the truck is detected, and the host computer controls the model to operate in the corresponding state based on the operating status.

[0008] When the model fails before the truck, the host computer generates a fault report based on the model's operating status, and the fault report provides an early warning to the truck.

[0009] In one possible implementation, detecting the operating status of the truck includes:

[0010] During the operation of the truck, the measurement module records the applied quantity of the corresponding parts or systems of the truck; the verification module records the execution quantity of the truck under the applied quantity.

[0011] The applied amount and the executed amount constitute the operating status, which is uploaded to the host computer by the analysis module.

[0012] In one possible implementation, detecting the operating status of the truck further includes:

[0013] After the truck comes to a stop, maintenance personnel will inspect the measurement module, the calibration module, and the analysis module.

[0014] The maintenance personnel will report the current status of the truck to the host computer.

[0015] In one possible implementation, the host computer controlling the model to run in a corresponding state based on the operating status includes:

[0016] The host computer extracts parameters such as the weight of the truck, driving speed, applied amount, executed amount and vibration amount from the operating status, and makes the model run in a similar state.

[0017] In one possible implementation, the host computer controlling the model to run in a corresponding state based on the operating status includes:

[0018] During the model simulation, an interference term is set to simulate the losses in actual operation.

[0019] In one possible implementation, setting the disturbance term to make the model simulate the losses during actual operation includes:

[0020] The model was subjected to similar vibration conditions while the truck was in motion.

[0021] In one possible implementation, before the host computer generates a fault report based on the model's operating status, the following steps are also included:

[0022] A reference library is established, and multiple fault solutions are generated based on the parameters fed back by the measurement module, calibration module, and analysis module, combined with the reference library.

[0023] In one possible implementation, after generating multiple fault solutions based on the parameters fed back by the measurement module, verification module, and analysis module in conjunction with the reference library, the following further step is added:

[0024] Multiple fault solutions are tested and repaired in the model. Once the fault is resolved by the fault solution, the truck is repaired by maintenance personnel.

[0025] In one possible implementation, the host computer controlling the model to run in a corresponding state based on the operating status includes:

[0026] The host computer enables the model to run in the amplified state of the operating conditions. When the model malfunctions, the host computer generates a fault report.

[0027] In one possible implementation, the step of the host computer generating a fault report when the model malfunctions includes:

[0028] Determine the current status of potentially faulty parts on the truck, and calculate the approximate timeframe for when these parts may fail based on the fault report.

[0029] The beneficial effects of the railway freight car fault prediction method provided by the present invention are as follows: Compared with the prior art, the railway freight car fault prediction method of the present invention first constructs a model of the freight car, and sets gravity parameters to make the weight of each module in the model the same as the weight of the corresponding parts of the freight car. That is, a freight car of the same specification is constructed in the host computer, and the host computer makes the model in a state corresponding to the operating condition by detecting the operating status of the freight car.

[0030] Since the truck and the model operate simultaneously, their operational status is relevant. When the model malfunctions, the host computer generates a fault report based on the model's operational status. At this point, the truck is given an early warning based on the fault report that first appears in the model.

[0031] In this application, a model is constructed and placed in a state corresponding to the operating conditions of a freight car. Since the weight of each module in the model is the same as that of the actual parts of the freight car, the model can provide timely warnings of freight car malfunctions, preventing major malfunctions and ensuring stable freight car operation and the safety of railway transportation. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A flowchart of a railway freight car fault prediction method provided in an embodiment of the present invention. Detailed Implementation

[0034] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0035] Please see Figure 1 The railway freight car fault prediction method provided by this invention will now be described. The railway freight car fault prediction method includes:

[0036] A truck model is built in the host computer, and gravity parameters are set on the model so that the weight of each module in the model is the same as the weight of the corresponding parts of the truck.

[0037] The computer monitors the truck's operating status and adjusts the model to the appropriate state based on that status.

[0038] When the model fails before the truck, the host computer generates a fault report based on the model's operation, and the fault report provides an early warning to the truck.

[0039] The beneficial effects of the railway freight car fault prediction method provided by the present invention are as follows: Compared with the prior art, the railway freight car fault prediction method of the present invention first constructs a model of the freight car, and sets gravity parameters to make the weight of each module in the model the same as the weight of the corresponding parts of the freight car. That is, a freight car of the same specification is constructed in the host computer, and the host computer makes the model in a state corresponding to the operating condition by detecting the operating status of the freight car.

[0040] Since the truck and the model operate simultaneously, their operational status is relevant. When the model malfunctions, the host computer generates a fault report based on the model's operational status. At this point, the truck is given an early warning based on the fault report that first appears in the model.

[0041] In this application, a model is constructed and placed in a state corresponding to the operating conditions of a freight car. Since the weight of each module in the model is the same as that of the actual parts of the freight car, the model can provide timely warnings of freight car malfunctions, preventing major malfunctions and ensuring stable freight car operation and the safety of railway transportation.

[0042] Safety and reliability are the most basic and important requirements for rail transit equipment. With the development of technology, rail transit equipment is becoming increasingly complex, and the diversity, integration level and intelligence of freight trains are constantly improving.

[0043] As the number of trains in operation increases and their complexity grows, more safety-influencing factors will inevitably be introduced, increasing the likelihood of component failures and malfunctions. Therefore, to ensure train safety, greater investment is needed in train design, production, maintenance, and monitoring. Currently, most railway freight car inspections rely on manual analysis of monitoring data and human error prediction, making accurate judgments difficult.

[0044] Existing technologies involve installing sensors at key locations on the vehicle body and transmitting the data collected by these sensors to corresponding monitoring systems. These systems analyze the received data to determine which systems or parts of the vehicle have malfunctioned. This predictive method is relatively simple; it uses sensors to replace manual inspection and relies on experience to judge abnormal data when anomalies are detected. However, this method only enables real-time data transmission and cannot predict potential truck malfunctions.

[0045] Fault prediction is based on existing collected data, combined with the truck's own structure and maintenance experience, to accurately predict possible faults or accurately locate faults that have already occurred, so as to promptly and specifically address the faults in the truck.

[0046] To address the aforementioned issues, this application provides a method for predicting railway freight car faults, which, compared to existing technologies, can more comprehensively perform dynamic and static detection of the freight car's status. Through dynamic and static detection, a freight car status profile is established. Furthermore, during freight car operation, the host computer simulates the freight car's status in real time using a model. This model can intuitively display potential faults and their causes, thereby providing more accurate data support for freight car improvement and fault diagnosis.

[0047] In some embodiments of the railway freight car fault prediction method provided in this application, detecting the operating status of the freight car includes:

[0048] During the truck's operation, the measurement module records the applied quantity of the corresponding parts or systems of the truck; the verification module records the execution quantity of the truck under the applied quantity.

[0049] The applied quantity and the executed quantity constitute the operational status, which is then uploaded to the host computer by the analysis module.

[0050] To enable dynamic monitoring of trucks, multiple sensors and instruments are installed on them to detect malfunctions. Taking the truck's steering system as an example, to accurately assess its condition, a measurement module is first installed to determine the steering wheel's rotation angle, and a verification module records the actual wheel rotation angle. In other words, the measurement module records the steering wheel's rotation angle, while the verification module records the actual wheel sway angle. An analysis module is then used to determine the stress and sway angle experienced by each component during the truck's steering process.

[0051] During the truck's movement, this data is transmitted in real time to the host computer via the communication module. The host computer stores the execution standards determined by the measurement and calibration modules; that is, it stores the standards for the stress and angle of each part's oscillation motion when the truck rotates at a corresponding angle. The host computer compares the actual collected data with the standards and then analyzes the cause of the deviation. If the problem is serious, more detailed testing is required during static inspection.

[0052] It should be noted that there are many components and systems on the truck, and the types and functions of the corresponding measurement and calibration modules are also different. That is, the measurement and calibration modules are only limited to one function, and different instruments or devices are needed to realize the corresponding functions for different systems.

[0053] In some embodiments of the railway freight car fault prediction method provided in this application, detecting the operating status of the freight car further includes:

[0054] After the truck comes to a stop, maintenance personnel will inspect the measurement module, calibration module, and analysis module.

[0055] The maintenance personnel will report the current status of the truck to the host computer.

[0056] The static inspection in this application mainly refers to the inspection of the truck while it is stationary, which is currently mostly carried out manually. During the static inspection process, staff not only need to check and record the basic functions of the truck, but also need to check and verify the problems found during dynamic inspection. The results recorded by maintenance personnel need to be uploaded to a host computer, which analyzes and summarizes the problems found during dynamic inspection based on the feedback data.

[0057] For some issues, maintenance personnel can upload the truck's status or the movement of parts to a host computer via video and images. The host computer will then compare the transmitted information with the standard to analyze the cause of the deviation.

[0058] In addition, maintenance personnel need to check each measurement module, calibration module, and analysis module to determine whether they are in normal working order.

[0059] To conduct a more detailed inspection, the feedback of each function under dynamic driving conditions can be tested on-site, such as the steering wheel rotation angle and wheel swing angle when stationary, thereby providing more comprehensive data for the host computer to solve problems.

[0060] In some embodiments of the railway freight car fault prediction method provided in this application, the host computer keeps the model running in a corresponding state according to the operating conditions, including:

[0061] The host computer extracts parameters such as the weight of the truck's load, driving speed, applied amount, executed amount, and vibration amount from the operating status, and runs the model in a similar state.

[0062] Trucks are assembled from multiple parts. If one function malfunctions, other functions or parts may also be affected. Existing truck fault analysis systems only analyze and monitor the operation of the target part or system through sensors. However, the cause of the above problems may be the abnormal state of other functional parts, which ultimately leads to frequent faults. That is, after the repair is completed, the fault will reappear.

[0063] To address the aforementioned issues, this application constructs corresponding models based on the type of truck, incorporating gravity parameters to ensure that the weight of each component in the model matches the actual weight. To more clearly analyze the truck's operating status and potential malfunctions, before the truck is driven, the parameters determined by the measurement and verification modules under standard conditions must be input into the model. Furthermore, after actual driving, the weight of the cargo loaded on the truck and its distribution must also be input into the model.

[0064] The measurement and verification modules clearly show the truck's current operating status. This status is then input into the model, allowing the model to operate at the same speed while loading the same cargo. Based on these settings, the host computer can comprehensively determine the truck's current condition and provide early warnings about potential fault locations and types.

[0065] In some embodiments of the railway freight car fault prediction method provided in this application, the host computer keeps the model running in a corresponding state according to the operating conditions, including:

[0066] During model simulation, interference terms are set to simulate the losses in actual operation.

[0067] Because manufacturers produce overall truck models, but each truck's usage varies, installation and assembly may differ. To make the created model more realistic, maintenance personnel test various measurement, calibration, and analysis modules during each static inspection. During testing, the truck model's condition is also monitored to see if the actual truck and model are synchronized, i.e., whether the swing angles and stresses of corresponding components are the same. Furthermore, a series of interference factors, such as impurities, noise, deformation, and reduced accuracy, are added to specific locations in the model, as it is under relatively ideal conditions compared to reality.

[0068] During each static inspection, corresponding images or videos can be captured, and the model can be adjusted based on the actual shooting situation. The ultimate goal is to make it closer to the real situation.

[0069] In some embodiments of the railway freight car fault prediction method provided in this application, setting interference terms to enable the model to simulate the losses during actual operation includes:

[0070] The model was subjected to similar vibration conditions while the truck was in motion.

[0071] Under normal circumstances, trucks are most likely to break down while in motion, and they also experience the most severe load impacts and wear and tear during operation. Therefore, real-time monitoring of trucks during operation is crucial for early prediction of potential malfunctions.

[0072] However, it should be noted that freight cars generate periodic vibrations relative to the railway when traveling on it. The simulation model on the host computer is in a relatively idealized state, which leads to longer failure times or the model remaining in a normal state throughout, making effective fault prediction impossible. For these reasons, a camera can be installed in front of the freight car to collect information such as the unevenness and curvature angle of the railway. In other scenarios, vibration sensors can be installed on the freight car to detect the current vibration and transmit this information to the host computer via a communication module. The host computer then adjusts the model to operate under the same vibration conditions based on the freight car's vibration data.

[0073] To more closely resemble real-world conditions, the truck's operational status is fed back to the host computer in real time during its journey. This operational status includes parameters such as speed, acceleration, steering angle, braking force, cooling capacity, and power generation. The host computer then sets relevant parameters to keep the model in the same state.

[0074] In some embodiments of the railway freight car fault prediction method provided in this application, the method further includes the following steps before the host computer generates a fault report based on the model's operation:

[0075] A reference library is established, and multiple fault solutions are generated based on the parameters fed back by the measurement module, calibration module, and analysis module, combined with the reference library.

[0076] Because trucks have numerous components, interference may occur between different functional modules when a malfunction occurs. Furthermore, malfunctions develop gradually, requiring early warning to prevent them from escalating. To effectively identify malfunctions, a reference library needs to be established on the host computer beforehand, containing various malfunction types and identification methods. During truck operation, parameters determined by the measurement, verification, and analysis modules are fed back to the host computer in real time. The host computer analyzes these parameters against the reference library while simultaneously maintaining a consistent state across the truck.

[0077] The causes of the faults and solutions can be analyzed using reference libraries and models.

[0078] In some embodiments of the railway freight car fault prediction method provided in this application, after generating multiple fault solutions based on parameters fed back by the measurement module, verification module, and analysis module in conjunction with a reference library, the method further includes:

[0079] Multiple fault solutions are tested and repaired in the model. Once the fault is resolved by the model through the fault solution, the maintenance personnel will repair the truck.

[0080] To illustrate more clearly, when the parameters reported by the truck exceed the standards in the reference library, the host computer can determine several solutions to the current fault based on the information recorded in the reference library. However, it is still impossible to know which solution is most effective in solving the current problem, nor can it know how the truck's state will change after adopting a certain solution. However, since this application constructs a truck model within the host computer, and this model closely resembles the truck's actual state, the model can be used to simulate fault diagnosis before repairing or troubleshooting the truck.

[0081] In other words, before repairing the truck, multiple scenarios are tested within the model. The parameters of each part in the model can be precisely expressed digitally. After applying a repair scenario, the model in the host computer is driven at a certain simulated speed, and the model's operation is observed. The host computer then checks the state before and after the repair to determine whether the fault has been resolved. If not, the next scenario is applied.

[0082] In some embodiments of the railway freight car fault prediction method provided in this application, the host computer keeps the model running in a corresponding state according to the operating conditions, including:

[0083] The host computer runs the model in an amplified state of operation. When a model malfunctions, the host computer generates a fault report.

[0084] To predict actual truck malfunctions, the measurement, verification, and analysis modules transmit parameters to the host computer. The host computer then amplifies these parameters appropriately, subjecting the model to these amplified parameters. For example, the model's speed, vibrations, rotation angles, and driving resistance are all greater than those experienced by the actual truck. In this scenario, the model will malfunction before the truck itself.

[0085] When a model malfunctions, the host computer generates a corresponding fault report based on the location of the fault and sends the fault report to the maintenance personnel. During static testing, the maintenance personnel inspect the relevant components of the truck based on the fault report.

[0086] In some embodiments of the railway freight car fault prediction method provided in this application, when a model malfunctions, the host computer generates a fault report including:

[0087] Determine the current status of potentially faulty parts on the truck, and calculate the approximate timeframe for when these parts might fail based on the fault reports.

[0088] When a fault occurs in the model, it indicates that the structural strength or service life of some components has been affected under the current application environment. Since the model resides within a host computer, the host computer can detect any component within the model and digitally display its status. Based on these functions, the host computer records the status of each component or system within the entire model when the model begins its simulated driving, along with parameters such as the model's speed and cargo weight.

[0089] When a model malfunctions, the host computer uses parameters such as travel speed and cargo weight as a timeline, and refines the fault report with the state of the target part or system as a reference. The fault report shows how long the part or system traveled and at what speed before the malfunction occurred. During static inspections of the truck, potentially faulty parts or systems are detected and their current state is determined. The fault report then roughly estimates the time of malfunction, indicating the next loading speed, cargo weight, and travel time. Through these fault reports, early warnings of potential malfunctions can be provided.

[0090] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting railway freight car faults, characterized in that, include: A model of a truck is constructed in the host computer, and gravity parameters are set on the model so that the weight of each module in the model is the same as the weight of the corresponding part of the truck. The operating status of the truck is detected, and the host computer controls the model to operate in the corresponding state based on the operating status. When the model fails before the truck, the host computer generates a fault report based on the model's operating status, and the fault report provides an early warning to the truck. The detection of the truck's operating status includes: During the operation of the truck, the measurement module records the applied quantity of the corresponding parts or systems of the truck; the verification module records the execution quantity of the truck under the applied quantity. The applied amount and the executed amount constitute the operating status, which is uploaded to the host computer by the analysis module; The detection of the truck's operating status also includes: After the truck comes to a stop, maintenance personnel will inspect the measurement module, the calibration module, and the analysis module. The maintenance personnel will report the current status of the truck to the host computer. The host computer controls the model to run in a corresponding state based on the operating status, including: The host computer extracts parameters such as the weight of the truck, driving speed, applied amount, executed amount, and vibration from the operating status, and runs the model in the same state as the truck.

2. The railway freight car fault prediction method as described in claim 1, characterized in that, The host computer controls the model to run in a corresponding state based on the operating status, including: During the model simulation, an interference term is set to simulate the losses in actual operation.

3. The railway freight car fault prediction method as described in claim 2, characterized in that, The setting of interference terms to simulate the losses during actual operation of the model includes: When the truck is in motion, the model is subjected to the same vibration conditions as the truck.

4. The railway freight car fault prediction method as described in claim 3, characterized in that, Before the host computer generates a fault report based on the model's operating status, the following steps are also included: A reference library is established, and multiple fault solutions are generated based on the parameters fed back by the measurement module, calibration module, and analysis module, combined with the reference library.

5. The railway freight car fault prediction method as described in claim 4, characterized in that, After generating multiple fault solutions based on the parameters fed back by the measurement module, calibration module, and analysis module, and in conjunction with the reference library, the process further includes: Multiple fault solutions are tested and repaired in the model. Once the fault is resolved by the fault solution, the truck is repaired by maintenance personnel.

6. The railway freight car fault prediction method as described in claim 1, characterized in that, The host computer controls the model to run in a corresponding state based on the operating status, including: The host computer enables the model to run in the amplified state of the operating conditions. When the model malfunctions, the host computer generates a fault report.

Images