Real-time green evaluation method and system for railway wagon operation process

Abstract

The invention provides a real-time green evaluation method and system for the operation process of a rail wagon, and relates to the technical field of rail transit green intelligence, the method comprises the steps of deploying a lightweight green evaluation model on a vehicle-mounted terminal, and carrying out real-time green evaluation, and the specific steps are as follows: obtaining multi-source sensing data of a target rail wagon in real time; on the basis of the multi-source sensing data, index scores of traction energy consumption, environmental influence, transportation efficiency and vehicle health are calculated through self-adaptive dynamic weights, and a final green evaluation result is obtained; according to the green evaluation result, carrying out full life cycle optimization on the operation process of the target rail wagon, and generating a green operation and maintenance decision; by integrating multi-source sensing data and a dynamic weight optimization algorithm technology, an energy-environment-efficiency-health real-time multi-dimensional dynamic green evaluation model is provided, and quantification and evaluation of traction energy consumption, environmental influence, transportation efficiency and vehicle health in the operation process are achieved.

Description

Technical Field

[0001] This disclosure relates to the field of green and intelligent technology for rail transit, specifically to a real-time green evaluation method and system for railway freight car operation. Background Technology

[0002] Monitoring the operation of railway freight cars plays an important role in ensuring the safety of rail transit, improving efficiency, reducing costs, and promoting intelligent development. The evaluation of the operation process is an effective way of monitoring, which can provide quantitative evaluation results and assist staff in making operation and maintenance decisions and switching operation modes.

[0003] Existing railway freight car operation evaluation technologies suffer from the following problems, resulting in poor timeliness and low accuracy, ultimately affecting the safe and efficient operation of railway freight cars: Limitations of static evaluation: Traditional green evaluation methods for railway freight cars are mostly based on fixed-cycle or fixed-weight indicators (such as annual average energy consumption), which cannot reflect changes in vehicle operating status in real time (such as short-term energy consumption fluctuations) and cannot respond to or deal with sudden operating events.

[0004] Single data dimension: Existing technologies rely on a single data source (such as vehicle energy consumption monitors) and lack integrated analysis of mechanical performance degradation and external environmental factors (such as weather and terrain).

[0005] Decision-making lag: Evaluation results are disconnected from operation and maintenance decisions, making it impossible to generate dynamic optimization suggestions in real time.

[0006] The indicators lack scientific rigor: the existing indicators do not take into account environmental impact and operational efficiency (such as the comprehensive environmental impact of unit cargo turnover). Summary of the Invention

[0007] To address the aforementioned issues, this disclosure proposes a real-time green evaluation method and system for railway freight car operation. By integrating multi-source sensor data and dynamic weight optimization algorithm technology, a real-time, multi-dimensional, dynamic green evaluation model of "energy-environment-efficiency-health" is proposed to quantify and evaluate traction energy consumption, environmental impact, transportation efficiency, and vehicle health during operation.

[0008] According to some embodiments, the present disclosure adopts the following technical solutions: A real-time green assessment method for railway freight car operation is proposed, which deploys a lightweight green assessment model on the onboard terminal to perform real-time green assessment. The specific steps are as follows: Real-time acquisition of multi-source sensor data of target railway freight cars; Based on multi-source sensor data, and using adaptive dynamic weights, scores for each indicator of traction energy consumption, environmental impact, transportation efficiency, and vehicle health are calculated to obtain the final green evaluation result. Based on the green assessment results, the entire lifecycle of the target railway freight car operation process is optimized to generate green operation and maintenance decisions; The real-time green evaluation method adopts an edge-cloud collaborative computing architecture, with the vehicle terminal as the edge and a local lightweight green evaluation model to perform real-time green evaluation of the operation process. The cloud uses federated learning technology to aggregate data from multiple vehicles and dynamically update the parameters of the lightweight green evaluation model, including adaptive dynamic weights.

[0009] Furthermore, the real-time acquisition of multi-source sensor data of the target railway freight car specifically includes: The system uses onboard sensors to collect data on the traction power, speed, acceleration, carbon emissions, tare weight, load, running distance, transport time, vibration signals, temperature, and sound of railway freight cars, and performs data preprocessing.

[0010] Furthermore, the traction energy consumption is calculated based on traction power, speed, and acceleration, using the instantaneous traction power and cumulative traction energy consumption of the vehicle's running resistance as a score. The environmental impact is calculated based on carbon emissions and sound data, taking into account the environmental impact of carbon emissions and noise levels; The transportation efficiency is calculated based on the relationship between cargo volume and delivery time, taking into account factors such as tare weight, load, travel distance, and delivery time.

[0011] Furthermore, the vehicle health score is calculated as follows: Based on the vibration signals of mechanical components, the real-time health status of the mechanical components is captured, and a vehicle health score is calculated.

[0012] Furthermore, the adaptive dynamic weights include the weights corresponding to each indicator and the weights corresponding to each component.

[0013] Furthermore, the dynamic update of the adaptive dynamic weights specifically includes: The cloud uses federated learning technology to aggregate data from multiple vehicles. Based on multi-source sensor data and environmental changes, it dynamically adjusts the weights of each indicator and each component to generate a global optimized weight matrix, which is periodically distributed to the edge.

[0014] According to some embodiments, the present disclosure adopts the following technical solutions: A real-time green assessment system for railway freight car operation deploys a lightweight green assessment model on an onboard terminal to perform real-time green assessments. The system includes: The acquisition module is configured to acquire multi-source sensor data of the target railway freight car in real time. The evaluation module is configured to: calculate the scores of each indicator of traction energy consumption, environmental impact, transportation efficiency and vehicle health based on multi-source sensor data and using adaptive dynamic weights, so as to obtain the final green evaluation result. The decision-making module is configured to: optimize the entire lifecycle of the target railway freight car operation process based on the green evaluation results, and generate green operation and maintenance decisions; The real-time green evaluation method adopts an edge-cloud collaborative computing architecture, with the vehicle terminal as the edge and a local lightweight green evaluation model to perform real-time green evaluation of the operation process. The cloud uses federated learning technology to aggregate data from multiple vehicles and dynamically update the parameters of the lightweight green evaluation model, including adaptive dynamic weights.

[0015] According to some embodiments, the present disclosure adopts the following technical solutions: A computer program product includes a computer program that, when executed by a processor, implements the aforementioned real-time green evaluation method for railway freight car operation.

[0016] According to some embodiments, the present disclosure adopts the following technical solutions: A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned real-time green evaluation method for railway freight car operation.

[0017] According to some embodiments, the present disclosure adopts the following technical solutions: An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the aforementioned real-time green evaluation method for railway freight car operation.

[0018] Compared with the prior art, the beneficial effects of this disclosure are as follows: This invention proposes a real-time green evaluation method and system for railway freight car operation. By integrating multi-source sensor data and dynamic weight optimization algorithm technology, it accurately quantifies and coordinates the traction energy consumption, environmental impact, transportation efficiency and vehicle health during the operation phase, thereby achieving real-time, efficient and accurate evaluation of railway freight car operation and generating optimal green operation and maintenance decisions.

[0019] This invention adopts an edge-cloud collaborative computing architecture, with the vehicle terminal as the edge and a local lightweight green evaluation model to perform real-time green evaluation of the operation process. The cloud uses federated learning technology to aggregate data from multiple vehicles and dynamically update the weights of each indicator, thus solving the core problems of poor timeliness and low accuracy of traditional methods.

[0020] This invention breaks through the traditional static evaluation framework and constructs a dynamic weighted green index evaluation model that includes traction energy consumption, environmental impact, transportation efficiency and vehicle health. The weights are adjusted in real time according to operating conditions, environment and regulations. A multi-objective optimization algorithm is used to balance traction energy consumption, environmental impact, transportation efficiency and vehicle health to achieve the optimal solution of comprehensive green performance index for railway freight cars during operation.

[0021] This invention employs a self-sustaining green power supply system that utilizes the vibration characteristics of vehicles during operation. Through piezoelectric vibration energy harvesting and intelligent power management, it solves the current technical problem of railway freight cars not being powered, while simultaneously achieving zero-cable deployment of sensor networks.

[0022] This invention, based on fault characteristic frequencies and real-time monitoring and quantification algorithms of full-frequency domain spectral energy, identifies and accurately captures vehicle vibration problems in advance, realizing an early warning vibration fault diagnosis technology. It solves the problem that traditional railway freight car fault inspection and diagnosis rely on manual inspection, improving the efficiency and accuracy of fault identification.

[0023] This invention employs a low-cost, readily available hardware solution and software platform, applicable to both railway and urban rail trains, enhancing safety while significantly reducing maintenance costs. This system solution possesses enormous market potential, significant technological advantages, low cost, and high reliability, making it valuable for long-term, large-scale adoption. Attached Figure Description

[0024] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0025] Figure 1 This is a flowchart of the method in Example 1. Detailed Implementation

[0026] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0027] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0028] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0029] Example 1 One embodiment of this disclosure provides a real-time green evaluation method for railway freight car operation. Based on an edge-cloud collaborative computing architecture, it proposes a real-time, multi-dimensional, dynamic green evaluation model of "energy-environment-efficiency-health" by integrating multi-source sensor data and dynamic weight optimization algorithms. This model quantifies and evaluates traction energy consumption, environmental impact, transportation efficiency, and vehicle health during operation. The edge-cloud collaborative computing architecture uses the on-board terminal as the edge, on which a lightweight green evaluation model is deployed to perform real-time green evaluation of the operation process. In the cloud, federated learning technology is used to aggregate data from multiple vehicles, and the weights of each indicator are dynamically adjusted according to multi-source sensor data and environmental changes, and periodically distributed to the edge. The specific implementation process is described below from two aspects: real-time processing at the edge and deep optimization in the cloud.

[0030] I. Real-time processing at the edge A lightweight green evaluation model is deployed on the onboard terminal of the target railway freight car to realize the collection, preprocessing, index evaluation, and decision generation of multi-source sensor data, such as... Figure 1 As shown, the specific steps are as follows: Step S1: Acquire multi-source sensor data of the target railway freight car in real time.

[0031] The system uses onboard sensors to collect data on the traction power, speed, acceleration, carbon emissions, tare weight, load, running distance, transport time, vibration signals, temperature, and sound of railway freight cars, and performs data preprocessing.

[0032] The selection and configuration of the vehicle-mounted sensors are as follows: (1) Vibration detection: MEMS accelerometer (ADXL355, range ±20g, power consumption 1.8mW@100Hz).

[0033] (2) Noise detection: Digital microphone (INMP441, SNR 61dB, I²S output, power consumption 1.2mW).

[0034] (3) Speed ​​acquisition: Hall sensor (A1324, detects wheel axle speed, calculates speed by combining wheel circumference, power consumption 0.5mW).

[0035] (4) Environmental monitoring: Temperature and humidity sensor (SHT35, accuracy ±1.5%RH, power consumption 1μA@1Hz).

[0036] The sensor in this embodiment adopts a vibration energy harvesting power supply scheme. By utilizing the vibration characteristics of the vehicle during operation, and through the self-generating technology of piezoelectric vibration energy harvesting + intelligent power management, it solves the current technical problem of railway freight cars not being powered, and at the same time realizes the zero-cable deployment of the sensor network.

[0037] The vibration energy harvesting and power supply scheme is as follows: 1. Energy Harvesting Module Design (1) Selection of piezoelectric material: PZT-5H piezoelectric ceramic sheet with size 30mm×30mm and resonant frequency 120Hz is adopted, which is suitable for the typical vibration frequency of railway freight cars (5-100Hz).

[0038] (2) Mechanical structure: cantilever beam structure design with a counterweight mass block (10g) at the end to improve the energy conversion efficiency under low frequency vibration.

[0039] (3) Circuit design: 1) Rectifier circuit: Full-bridge rectifier (diode SS34, voltage drop 0.3V) converts AC output to DC.

[0040] 2) Energy storage unit: supercapacitor (5.5V / 10F) in parallel with lithium battery (3.7V / 500mAh), supporting instantaneous high power and continuous power supply.

[0041] 3) Power management: The LTC3588 chip enables energy storage and regulated output (3.3V / 5V adjustable).

[0042] 2. Power supply performance indicators (1) Output power: Under a vibration acceleration of 0.5g, the output power of a single module is about 8mW.

[0043] (4) Energy storage capacity: The supercapacitor can provide power to the system for 2 hours when fully charged, and the lithium battery can provide up to 48 hours of additional power.

[0044] (3) Switching logic: Supercapacitors are used as the primary power source, and the system seamlessly switches to lithium batteries when the voltage is low.

[0045] Step S2: Based on multi-source sensor data, and using adaptive dynamic weights, calculate the scores of each indicator for traction energy consumption, environmental impact, transportation efficiency, and vehicle health to obtain the final green evaluation result.

[0046] This embodiment proposes a real-time, multi-dimensional, dynamic green evaluation model encompassing "energy, environment, efficiency, and health" to calculate the comprehensive green performance score of railway freight car operation, expressed by the formula:

[0047] In the formula, The comprehensive green performance score for railway freight car operation ranges from 0 to 1. This is the standardized value for traction energy consumption indicators; These are the standardized values ​​for environmental impact indicators; This refers to the standardized value of the transportation efficiency indicator. Standardized values ​​for vehicle health indicators; , , , The weights of each indicator must satisfy the following conditions: .

[0048] The steps for calculating the overall green performance score are as follows: (1) Normalization: First, calculate each sub-indicator, and then normalize it to the comparable range (0~1) according to the preset benchmark value. (2) Weighted summation: The standardized values ​​are linearly weighted according to their weights to generate a comprehensive green performance index. ; (3) Interpretation of indicators: It has excellent green performance; It has good green performance; It needs to be optimized and improved. The four indicators are explained below.

[0049] 1. Traction energy consumption index

[0050] Traction energy consumption during railway freight car operation refers to the energy consumed by the railway freight traction system to maintain or change the motion state of the vehicle. It is usually related to factors such as traction force, speed, and running resistance, and can be expressed by the formula:

[0051]

[0052]

[0053] in: Cumulative traction energy consumption of the traction locomotive, in kWh; The baseline traction energy consumption is in kWh. The total mass of a railway freight car; The speed of the trains in the formation, in km / h; The distance traveled by the trains in a formation can be obtained by measuring the speed with equipment or by integrating the speed over time, in km.

[0054] Cumulative traction energy consumption Based on the calculation of vehicle running resistance, it can be expressed by the formula:

[0055]

[0056] in: Let be the instantaneous traction power at time t, in kW; These are empirical coefficients related to vehicle type, wheel-rail contact, and aerodynamic drag. Let be the vehicle speed at time t, in km / h; The gradient angle of the track at the location where the vehicle is operating; This is the coefficient for additional resistance on the slope. It is set to 1 when the railway freight equipment is on the slope after calculation, and 0 otherwise. The additional resistance ratio of the slope is the ratio of the mass of the railway freight equipment on the slope to the total mass of the assembled railway freight cars. The radius of the curve at the location where the vehicle is operating, in meters (m). This is the curve additional resistance effective coefficient, which is 1 when the railway freight equipment is on a curve segment after calculation, and 0 otherwise. is the additional resistance ratio coefficient for curves, which is the ratio of the mass of railway freight equipment on the curve segment to the total mass of the assembled railway freight cars; Let be the vehicle acceleration at time t, in m / s². The transmission efficiency of the traction locomotive transmission system.

[0057] 2. Environmental Impact Indicators

[0058] The environmental impact assessment indicators for railway freight car operation take into account the impact of noise levels on the environment, and the calculation method is as follows:

[0059]

[0060] in: The reference noise level is set at 60 dB(A). The vehicle's operating speed is The noise level at that time, in dB(A).

[0061] 3. Transportation efficiency indicators

[0062] The transportation efficiency evaluation index during railway freight car operation relates to the relationship between freight volume and delivery time.

[0063]

[0064]

[0065] in: The transport efficiency index for assembling railway freight cars; The total mass of goods transported by rail freight cars is expressed in tons (t). The travel time for assembling railway freight cars, in hours (h); The maximum permissible cargo mass (t) for assembling railway freight cars; The design speed for assembling railway freight cars, in km / h. This refers to the load factor of freight trains.

[0066] 4. Vehicle health indicators

[0067] Vehicle health indicators are core indicators in the green evaluation system for railway freight car operation, used to quantify the impact of the real-time health status of key mechanical components on overall green performance. Their innovation lies in reducing resource waste and environmental risks caused by unplanned downtime through predictive maintenance, achieving closed-loop optimization of "mechanical reliability - environmental performance - economic benefits." Vehicle health indicators comprehensively capture the real-time health status of mechanical components through vibration detection and analysis techniques. The calculation formula is as follows:

[0068] in: Vibration analysis health score for potential fault i; Let be the weight of the vibration analysis health score for potential fault i; m be the number of potential faults monitored for vibration; each weight coefficient is allocated based on the degree of impact caused by the fault risk, and satisfies the following conditions: .

[0069] (1) Background principle of vibration health scoring Mechanical components of railway freight cars (such as bearings in the running gear) generate vibration signals during operation due to friction, impact, and deformation. The vibration characteristics differ significantly under different health conditions. - Normal state: The vibration spectrum shows a stable background noise and natural frequency components, and the energy distribution is uniform; - Early failure: Weak harmonic components appear at specific failure frequencies (such as the frequency of defects in the inner ring of the bearing); - Severe fault: The fault frequency amplitude increases significantly, accompanied by an increase in sideband and broadband noise energy.

[0070] In the analysis of vibration signals from railway freight cars, the Fast Fourier Transform (FFT) is used to convert the time-domain vibration signal into a frequency-domain signal to extract fault characteristic frequencies. For example, the fault characteristic frequencies of bearings are identified as follows: - Inner loop fault frequency (BPFI):

[0071] - Outer ring failure frequency (BPFO):

[0072] - Rolling element failure frequency (BSF):

[0073] This refers to the number of rolling elements in the bearing. The bearing rolling element diameter is in mm. The bearing pitch circle diameter is in mm. θ is the contact angle, °; RPM is the bearing speed, r / min.

[0074] (2) Vibration health score calculation model - Reference Spectrum Energy Quantization: Multiple sets of vibration data are collected under healthy equipment conditions, and the average energy within the three fault characteristic identification frequency bands (i.e., the range of fault characteristic frequency ± 3 times the sampling frequency) for the inner ring, outer ring, and rolling element faults is calculated respectively. (Usually, data from the previous 100 hours of trouble-free operation is used).

[0075] - Fault spectrum energy quantization: When the i-th type of fault occurs, extract the average energy of the real-time fault spectrum within the fault characteristic frequency band (i.e., the range of fault characteristic frequency ± 3 times the sampling frequency). Then calculate the energy deviation of the i-th type of fault.

[0076]

[0077] Identify reference energy deviation within the frequency band by identifying fault characteristics. Normalize the vibration analysis health score: ,

[0078] - Physical meaning: When monitoring the potential fault frequency of a certain vehicle structure - The current spectrum is consistent with the benchmark, and the equipment is in good condition. - Significant energy anomaly triggered an early warning.

[0079] Step S3: Based on the green assessment results, optimize the entire lifecycle of the target railway freight car operation process to generate green operation and maintenance decisions; An intelligent closed-loop decision-making system is adopted to optimize the entire lifecycle of the target railway freight car based on the green assessment results, generating green operation and maintenance decisions. This intelligent closed-loop decision-making system analyzes the dynamic changes of traction energy consumption indicators (EII), environmental impact indicators (ETE), transport efficiency indicators (TEI), and vehicle health indicators (MHI) in real time, generates optimization decision instructions, and forms a closed-loop feedback to achieve continuous improvement in the green performance of railway freight cars throughout their entire lifecycle. The synergistic effect between indicators and decisions is specifically reflected in the following four layers of logic: 3.1 Traction Energy Consumption Indicators and Energy Optimization Decisions (1) Data-driven: - Real-time monitoring of data such as traction power and regenerative braking energy recovery rate, and calculation of traction energy consumption; - Dynamically identify high-energy-consuming operating conditions (such as steep slope acceleration and frequent start-stop).

[0080] (2) Decision triggering: - Driving behavior optimization: Based on the historical curve of traction energy consumption and digital twin simulation, a constant speed driving mode or a limit on the maximum vehicle speed is recommended; - Energy Management: Switch to hybrid mode (internal combustion + energy storage) in non-electrified sections to reduce peak instantaneous traction energy consumption.

[0081] (3) Closed-loop verification: - After implementing the optimization strategy, the system recalculates the traction energy consumption value to verify the energy saving / fuel saving effect (e.g., a certain optimization reduces the traction energy consumption value by 15%).

[0082] Case: When the system detects that a train's traction energy consumption reaches 0.25 kWh / ton-km (exceeding the threshold by 20%) while running on a 10‰ gradient, it automatically reduces the traction output gradient and activates energy storage assistance, reducing the traction energy consumption to 0.21 kWh / ton-km.

[0083] 3.2 Environmental Impact Indicators and Ecological Protection Decisions (1) Dynamic perception: - GIS maps mark ecologically sensitive areas (such as nature reserves, water sources, and residential areas).

[0084] (2) Decision triggering: - Emission control: Reduce operating speed in sensitive areas to reduce vehicle operating noise.

[0085] (3) Closed-loop verification: - Optimized real-time monitoring of changes in environmental impact indicators to generate environmental performance reports. Case: The environmental impact index of a certain railway freight train is 58. The system detects that the train is passing through a residential area ahead. The environmental impact index of the train is reduced to 42 by adopting a speed reduction scheme, which is a 27% reduction in the environmental impact index.

[0086] 3.3 Transportation efficiency indicators and economic-environmental collaborative decision-making (1) Multi-objective optimization: By linking the volume of goods transported with the timeliness of delivery, we can find the combination of operating parameters that corresponds to the maximum value of the transportation efficiency index.

[0087] (2) Decision triggering: - Loading optimization: Recommend the best load ratio (e.g., the transportation efficiency index is highest when the load is 80%). - Timetable adjustment: Increase the proportion of electric locomotives in operation during low-carbon periods of the power grid (such as peak wind power periods).

[0088] (3) Closed-loop verification: - Compare the transportation efficiency index values ​​before and after optimization to quantify the synergistic gains in economic and environmental benefits (e.g., an 18% increase in transportation efficiency index). Case Study: System analysis of historical data for a freight route revealed that the carbon intensity of electricity was 30% lower at night. It was recommended that 60% of the trips be rescheduled to nighttime, which improved the transportation efficiency index from 0.85 to 1.02.

[0089] 3.4 Vehicle Health Indicators and Preventive Green Maintenance (1) State perception: - By integrating vibration, temperature and acoustic emission data, vehicle health index values ​​are calculated to predict the remaining life (RUL) of components. - Identify the potential impact of latent faults on energy consumption and emissions (such as increased frictional energy consumption due to bearing wear).

[0090] (2) Decision triggering: - Maintenance strategy optimization: When the vehicle health index is <0.7, generate spare parts replacement work orders and low-carbon maintenance plans (such as using remanufactured parts). - Degraded operating mode: Limit the maximum speed when the vehicle's health indicators are too low to prevent the malfunction from escalating and causing an environmental accident.

[0091] (3) Closed-loop verification: - Reassess the correlation between vehicle health indicators and traction energy consumption / environmental impact indicators after maintenance (e.g., energy consumption decreased by 8% after bearing replacement).

[0092] Case: The health index of a train dropped from 0.82 to 0.58 (due to brake pad wear). The system recommended replacing the parts with recycled materials. After the repair, the health index of the train returned to 0.88, and brake dust emissions were reduced by 40%.

[0093] II. Deep Cloud Optimization Federated learning technology is used to aggregate data from multiple vehicles. Based on multi-source sensor data and environmental changes, the weights of each indicator and each component are dynamically adjusted to generate a global optimized weight matrix, which is periodically distributed to the edge.

[0094] The dynamic weight optimization algorithm, deployed in the cloud, is the core decision engine in the green evaluation system for railway freight cars. Its core objective is to dynamically adjust the weights of each evaluation indicator based on real-time operational data and environmental changes, ensuring that the evaluation results accurately reflect the current state of the vehicles and drive intelligent decision-making. This algorithm solves the key problem that traditional static weighting methods cannot respond to dynamic operating conditions through multi-source data fusion, combined weighting, and adaptive feedback mechanisms. The following section elaborates on the algorithm framework, mathematical model, and technical implementation: 1. Input data: - Real-time operating data: traction power, speed, acceleration, weight, load, and component health; -Environmental data: temperature and humidity, rain and snow weather, terrain slope; - External constraints: Environmental regulations, transportation task priorities (such as noise requirements).

[0095] 2. Output results: - Dynamic weight matrix: Real-time weight values ​​of each green evaluation indicator.

[0096] 3. Processing procedure: (1) Data preparation and standardization Based on the input data, the values ​​of each indicator are calculated using the method in step S2 to obtain the original indicator matrix. Where m is the number of evaluation objects and n is the number of evaluation indicators.

[0097] By standardizing the data and eliminating dimensions, a standardized matrix is ​​obtained. Among them, positive indicators are used Negative indicators .

[0098] (2) CRITIC method Calculate objective weights based on the conflict and information content among indicators:

[0099]

[0100] ,

[0101] In the formula, The information content of indicator j; The standard deviation of indicator j (reflecting volatility); The Pearson correlation coefficient between indicators i and j (reflecting conflict). (3) Entropy weight method Using information entropy to measure the dispersion of indicators:

[0102]

[0103]

[0104] In the formula, This represents the percentage of the standardized indicator value.

[0105] (4) Combinatorial weighting model: The two weights calculated above are combined, and an environmental compensation factor is introduced. :

[0106] In the formula, As an indicator The weight, and Empirical coefficient (usually) + = 1); The weight of "energy consumption" is dynamically adjusted based on real-time environmental data (e.g., increasing the weight of "energy consumption" during heavy rain).

[0107] (5) Dynamic update mechanism of sliding time window The data samples are updated on a rolling basis with a fixed time (e.g., 5 minutes) or mileage interval (e.g., 500 kilometers) as the window length, so as to dynamically acquire multi-source sensor data and environmental changes, which are the input data of this algorithm.

[0108] (6) Global optimization driven by intelligent learning - Edge end: Each train train trains are trained locally with weighted models, while preserving private data (such as the location of wear on specific components).

[0109] - Cloud: Aggregate parameters from multiple vehicle models to generate a globally optimized weight matrix. It is periodically distributed to the edge.

[0110] Example 2 One embodiment of this disclosure provides a real-time green assessment system for railway freight car operation, which deploys a lightweight green assessment model on an onboard terminal to perform real-time green assessments. The system includes: The acquisition module is configured to acquire multi-source sensor data of the target railway freight car in real time. The evaluation module is configured to: calculate the scores of each indicator of traction energy consumption, environmental impact, transportation efficiency and vehicle health based on multi-source sensor data and using adaptive dynamic weights, so as to obtain the final green evaluation result. The decision-making module is configured to: optimize the entire lifecycle of the target railway freight car operation process based on the green evaluation results, and generate green operation and maintenance decisions; The real-time green evaluation method adopts an edge-cloud collaborative computing architecture, with the vehicle terminal as the edge and a local lightweight green evaluation model to perform real-time green evaluation of the operation process. The cloud uses federated learning technology to aggregate data from multiple vehicles and dynamically update the parameters of the lightweight green evaluation model, including adaptive dynamic weights.

[0111] Example 3 One embodiment of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned real-time green evaluation method for railway freight car operation.

[0112] Example 4 One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the aforementioned real-time green evaluation method for railway freight car operation.

[0113] Example 5 One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the aforementioned real-time green evaluation method for railway freight car operation.

[0114] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0115] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0116] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A real-time green evaluation method for railway freight car operation, characterized in that, Deploying a lightweight green assessment model on the vehicle terminal for real-time green assessment involves the following steps: Real-time acquisition of multi-source sensor data of target railway freight cars; Based on multi-source sensor data, and using adaptive dynamic weights, scores for each indicator of traction energy consumption, environmental impact, transportation efficiency, and vehicle health are calculated to obtain the final green evaluation result. Based on the green assessment results, the entire lifecycle of the target railway freight car operation process is optimized to generate green operation and maintenance decisions; The real-time green evaluation method adopts an edge-cloud collaborative computing architecture, with the vehicle terminal as the edge and a local lightweight green evaluation model to perform real-time green evaluation of the operation process. The cloud uses federated learning technology to aggregate data from multiple vehicles and dynamically update the parameters of the lightweight green evaluation model, including adaptive dynamic weights.

2. The real-time green evaluation method for railway freight car operation as described in claim 1, characterized in that, The real-time acquisition of multi-source sensor data of the target railway freight car specifically includes: The system uses onboard sensors to collect data on the traction power, speed, acceleration, carbon emissions, tare weight, load, running distance, transport time, vibration signals, temperature, and sound of railway freight cars, and performs data preprocessing.

3. The real-time green evaluation method for railway freight car operation as described in claim 2, characterized in that, The traction energy consumption is calculated based on traction power, speed, and acceleration, using the instantaneous traction power and cumulative traction energy consumption of the vehicle's running resistance as the score is obtained. The environmental impact is calculated based on carbon emissions and sound data, taking into account the environmental impact of carbon emissions and noise levels; The transportation efficiency is calculated based on the relationship between cargo volume and delivery time, taking into account factors such as tare weight, load, travel distance, and delivery time.

4. The real-time green evaluation method for railway freight car operation as described in claim 1, characterized in that, The vehicle health score is calculated as follows: Based on the vibration signals of mechanical components, the real-time health status of the mechanical components is captured, and a vehicle health score is calculated.

5. A real-time green evaluation method for railway freight car operation as described in claim 4, characterized in that, The adaptive dynamic weights include the weights corresponding to each indicator and the weights corresponding to each component.

6. The real-time green evaluation method for railway freight car operation as described in claim 5, characterized in that, The dynamic update of the adaptive dynamic weights is specifically as follows: The cloud uses federated learning technology to aggregate data from multiple vehicles. Based on multi-source sensor data and environmental changes, it dynamically adjusts the weights of each indicator and each component to generate a global optimized weight matrix, which is periodically distributed to the edge.

7. A real-time green evaluation system for railway freight car operation, characterized in that, A lightweight green assessment model is deployed on an in-vehicle terminal to perform real-time green assessments. The system includes: The acquisition module is configured to acquire multi-source sensor data of the target railway freight car in real time. The evaluation module is configured to: calculate the scores of each indicator of traction energy consumption, environmental impact, transportation efficiency and vehicle health based on multi-source sensor data and using adaptive dynamic weights, so as to obtain the final green evaluation result. The decision-making module is configured to: optimize the entire lifecycle of the target railway freight car operation process based on the green evaluation results, and generate green operation and maintenance decisions; The real-time green evaluation method adopts an edge-cloud collaborative computing architecture, with the vehicle terminal as the edge and a local lightweight green evaluation model to perform real-time green evaluation of the operation process. The cloud uses federated learning technology to aggregate data from multiple vehicles and dynamically update the parameters of the lightweight green evaluation model, including adaptive dynamic weights.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the real-time green evaluation method for railway freight car operation as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement a real-time green evaluation method for railway freight car operation as described in any one of claims 1-6.

10. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement a real-time green evaluation method for railway freight car operation as described in any one of claims 1-6.

Images