Full-life-cycle maintenance management information system for special vehicle

Abstract

The invention discloses a special vehicle full-life-cycle maintenance management information system, and belongs to the technical field of special vehicle maintenance. The vehicle information management unit comprises a data acquisition module and an archive management module, the data acquisition module comprises an automatic acquisition module, a manual acquisition module and a data verification module, and the archive management module is used for storing vehicle basic data and vehicle maintenance data; the vehicle health analysis unit comprises a health degree calculation module, a fault analysis module and a prediction module; and the maintenance cost analysis unit comprises an association module, an accounting module and a cost optimization module. Full-period management of the special vehicle is realized through each unit, the vehicle health analysis unit arranged in the system converts the maintenance work from passive response to active prevention, the fault occurrence rate and the maintenance lagging risk are effectively reduced, the maintenance cost analysis unit effectively integrates maintenance resources, waste is reduced, and the maintenance efficiency is improved. And the whole-cycle maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of special vehicle maintenance technology, and in particular to a special vehicle full life cycle maintenance management information system. Background Technology

[0002] Special vehicles (such as mobile power supply vehicles, insulated bucket trucks, mobile transformer substation vehicles, mobile ring network cabinet vehicles, cable bypass vehicles, UPS power supply vehicles, etc.) have extremely high requirements for the timeliness, accuracy and full life cycle coverage of maintenance management due to their special operating environment and complex functional requirements. In existing technologies, the maintenance and management of special vehicles mostly adopts a decentralized management model, which has many technical defects: First, vehicle information collection and management are not standardized. Data collection mostly relies on a single manual input method, which is inefficient and prone to errors. There is a lack of a collaborative verification mechanism between automatic and manual collection. At the same time, basic vehicle data and maintenance data are not uniformly archived and dynamically updated, resulting in data fragmentation and poor timeliness. Second, vehicle health status analysis lacks a systematic solution. It is impossible to assess vehicle health through multi-dimensional reference and quantitative evaluation. Fault analysis relies solely on manual experience, making it difficult to accurately locate high-frequency fault modes and root causes. Furthermore, there is a lack of effective prediction of component lifespan and fault occurrence time, leading to passive and delayed maintenance work. Third, maintenance cost control lacks a full-cycle perspective. An effective correlation between vehicles, components, and maintenance activities has not been established, making it impossible to accurately calculate the full-cycle maintenance costs of vehicles from purchase to retirement. It is even more difficult to generate targeted optimization solutions based on cost data, resulting in waste of maintenance resources and high costs.

[0003] In view of the shortcomings of the existing technologies, there is an urgent need for a special vehicle maintenance management information system that can achieve precise access control, standardized data management, intelligent health analysis, and full-cycle cost optimization, so as to improve the efficiency and scientific nature of maintenance management and ensure the reliable operation of special vehicles. Summary of the Invention

[0004] The purpose of this invention is to provide a special vehicle lifecycle maintenance management information system to solve the problems of difficult and inaccurate control and lack of scientific management and analysis of special vehicle maintenance data. This system has the advantages of accurate and comprehensive data control, intelligent analysis of health status, and management of maintenance costs.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] The special vehicle full life cycle maintenance management information system includes:

[0007] The user management unit is used to manage user access permissions;

[0008] The vehicle information management unit includes a data acquisition module and a file management module. The data acquisition module includes an automatic acquisition module, a manual acquisition module, and a data verification module. The file management module is used to store basic vehicle data and vehicle maintenance data.

[0009] The vehicle health analysis unit includes a health calculation module, a fault analysis module, and a prediction module. The health calculation module uses the frequency of fault occurrence, mean time between failures, component integrity rate, and the percentage of qualified test items as reference factors to calculate the vehicle health by weighting each of the reference factors. The fault analysis module is used to obtain fault characteristics and output fault results. The prediction module is used to predict the service life of components and the time of fault occurrence.

[0010] The maintenance cost analysis unit includes an association module, an accounting module, and a cost optimization module. The association module is used to establish the association between vehicles, components, and maintenance. The accounting module is used to calculate the full life cycle cost of a vehicle from purchase to retirement. The cost optimization module is used to analyze the maintenance cost of the vehicle and generate an optimized cost plan.

[0011] Preferably, the automatic acquisition module in the data acquisition module is used to obtain vehicle positioning data through the GPS module, obtain vehicle driving data through the vehicle diagnostic module, and transmit data through the GPRS module. The manual acquisition module is used to call the input system to save the vehicle data input by the user. The data verification module is used to verify whether the data transmitted by the automatic acquisition module and the manual acquisition module are consistent. If they are consistent, duplicate data is removed. The data verification module also includes sorting all data according to timestamps and sending supplementary reminders to the input system for missing data through the manual acquisition module.

[0012] Preferably, the file management module includes setting classification management dimensions and dynamic update rules, saving the data in the data acquisition module according to the classification management dimensions and updating the data in real time according to the dynamic update rules.

[0013] Preferably, the health calculation module uses the following formula to calculate the failure frequency, mean time between failures, component integrity rate, and percentage of qualified test items as reference factors:

[0014] ;

[0015] Mean Time Between Failures (MTBF) = ;

[0016] Component integrity rate = ;

[0017] Percentage of items passing the test = ;

[0018] The formula for calculating vehicle health by weighting each of the aforementioned reference factors is as follows:

[0019] Vehicle health rating = w1 × (1 - frequency of failure) + w2 × mean time between failures + w3 × component integrity rate + w4 × percentage of qualified test items;

[0020] Among them, w1, w2, w3, and w4 are preset weight values, and w1+w2+w3+w4=1.

[0021] Preferably, the fault analysis module includes a fault data preprocessing module and a fault correlation analysis module. The fault data preprocessing module is used to extract the vehicle's fault codes, faulty components, fault duration, and associated sensor data as fault features.

[0022] The fault association analysis module includes a density clustering algorithm and a root cause association algorithm. The density clustering algorithm is used to perform density clustering on fault features to form fault clusters of the same type. The root cause association algorithm is used to calculate the support and confidence of each fault cluster to generate association rules between fault events and fault features, and output the fault events as fault results.

[0023] Preferably, the prediction module includes a time-series data construction module, a prediction model training module, and a prediction output module;

[0024] The time-series data construction module is used to acquire historical time-series data and construct a sample set by combining the vehicle's cumulative running time and ambient temperature data. The sample set is divided into a test set and a training set.

[0025] The prediction model training module is an LSTM prediction model, which includes a forget gate, an input gate, a cell state, and an output gate. The mean squared error is used as the loss function, and the model is iteratively trained using the training set until the loss function converges to obtain the final prediction model.

[0026] The prediction output module is used to input the time series data of the component under test into the final prediction model and output the prediction results.

[0027] Preferably, the association module in the maintenance cost analysis unit is used to establish a three-way association mapping relationship between vehicle number, vehicle core components, and maintenance behavior, generate a unique association identifier, and bind the corresponding data.

[0028] Preferably, the accounting module in the maintenance cost analysis unit is used to obtain the vehicle's purchase stage cost, operation and maintenance stage maintenance cost, and decommissioning stage disposal cost through the association mapping relationship, and calculate the full cycle cost.

[0029] Preferably, the cost optimization module in the maintenance cost analysis unit analyzes the maintenance cost ratio and cost change trend of different vehicles and different components, combines vehicle health analysis data to mine the correlation between cost and health status, generates a cost optimization plan that includes maintenance strategy adjustment and component selection optimization, and quantifies and predicts the cost savings after optimization.

[0030] Preferably, the system also includes a database for storing data from each unit and setting security protection strategies for the data.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. This invention achieves full-cycle management of special vehicles through various units. The vehicle health analysis unit set up in the system transforms maintenance work from passive response to proactive prevention, effectively reducing the failure rate and maintenance delay risk. The maintenance cost analysis unit effectively integrates maintenance resources, reduces waste, and lowers the full-cycle maintenance cost.

[0033] 2. The vehicle information management unit adopts a dual collection interface that combines automatic and manual data collection, and is equipped with a data verification module. This module can perform data consistency verification, duplicate data removal, and missing data reminders. At the same time, through the classification management and dynamic update rules of the file management module, it ensures the standardization, integrity, and timeliness of vehicle basic data and maintenance data, thus solving the problems of data fragmentation and chaotic management in existing technologies. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the framework of the special vehicle full life cycle maintenance management information system of the present invention. Detailed Implementation

[0035] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0036] Example 1

[0037] like Figure 1 As shown, the special vehicle full life cycle maintenance management information system includes:

[0038] The user management unit manages user access permissions. Through precise role-based permission division, it avoids data corruption and information leaks caused by unauthorized operations, while simplifying the operational processes for different users and improving system efficiency. Three roles can be preset: user, administrator, and super administrator. Differential operational permissions are assigned to different roles through the permission configuration interface. For example, users only have permissions for manually entering vehicle data, submitting fault records, and viewing maintenance plans; administrators have permissions for viewing health analysis results, exporting cost reports, and approving optimization plans; and super administrators have full permissions, including adding / deleting users, modifying role permissions, and configuring system parameters. When a user logs into the system, the system automatically verifies their role identity and only opens the operation interface corresponding to their permissions.

[0039] The vehicle information management unit includes a data acquisition module and a file management module. The data acquisition module includes an automatic acquisition module, a manual acquisition module, and a data verification module. The file management module is used to store basic vehicle data and vehicle maintenance data. It provides dual-interface data acquisition, which solves the limitations of single data entry. Combined with the various data verification modules, it takes into account both real-time performance and completeness. The file management module ensures data quality and standardization, avoids data fragmentation, and provides a reliable data foundation for subsequent analysis.

[0040] The data acquisition module includes an automatic acquisition module for obtaining vehicle location data via GPS, vehicle driving data via vehicle diagnostics, and data transmission via GPRS. A manual acquisition module is used to call the input system to save user-inputted vehicle data. A data verification module verifies the consistency between the data transmitted by the automatic and manual acquisition modules; if consistent, duplicate data is removed. The data verification module also sorts all data by timestamp and sends supplementary reminders to the input system for missing data via the manual acquisition module. The automatic acquisition module, using location and driving data, can achieve the following functions: 1. Real-time monitoring of engineering vehicles, including real-time acquisition and display of vehicle location, speed, and driving status; 2. Recording and displaying the driving trajectory of engineering vehicles, including start point, end point, and waypoints, facilitating vehicle trajectory playback and path analysis for management personnel. The input system is a developed visual input interface that supports user input of maintenance work orders, fault descriptions, and inspection reports, ultimately transmitting the data to the data acquisition module. Since data collection is conducted in two modes, a data verification module is used to process the data collected in both modes to avoid data redundancy and conflicts. For example, comparing automatically collected and manually collected data of the same type (using power supply vehicles and insulated bucket trucks as examples), if the error between the automatically and manually collected vehicle mileage values ​​is ≤5%, they are considered consistent, and duplicate data is automatically removed. To ensure complete data, the data verification module sorts all collected data in ascending order by timestamp. If data is missing for a certain period, such as a power supply vehicle having no mileage data for a few days, a pop-up reminder is sent to the maintenance personnel's data entry system via the manual collection module, requesting supplementary data entry. This dual-mode collection and verification approach ensures data integrity and conciseness.

[0041] The file management module includes setting classification management dimensions and dynamic update rules. It saves data from the data acquisition module according to these classification dimensions and updates the data in real time based on the dynamic update rules. The file management module can classify and save data according to four dimensions: vehicle basic information, operational data, maintenance records, and inspection reports. Taking a power supply vehicle as an example, vehicle basic information includes static data such as vehicle number, brand, purchase date, rated power, generator set model, and fuel tank capacity. Operational data includes dynamic data such as voltage / frequency changes, oil consumption, and engine speed. Based on the dynamic update rules, when the data acquisition module adds new data, such as new maintenance records or real-time operational data, an update mechanism is automatically triggered to synchronize the new data to the corresponding classified file and update the file's last modification time.

[0042] The vehicle health analysis unit includes a health calculation module, a fault analysis module, and a prediction module. The health calculation module uses fault occurrence frequency, mean time between failures (MTBF), component integrity rate, and the percentage of qualified inspection items as reference factors to calculate the vehicle's health level by weighting each reference factor. The fault analysis module acquires fault characteristics and outputs fault results. The prediction module predicts component lifespan and fault occurrence time. The health calculation module achieves a quantitative assessment of the vehicle's health status, avoiding the subjectivity of human experience judgment. The fault analysis module can accurately locate high-frequency faults and their causes based on quantified data. The prediction module adapts to the prediction needs of time-series data, improving the accuracy of component lifespan and fault time prediction, and enabling maintenance work to shift from passive response to proactive prevention.

[0043] The specific working principles of the aforementioned health calculation module, fault analysis module, and prediction module are as follows:

[0044] The health status calculation module uses the following formula to calculate the failure frequency, mean time between failures, component integrity rate, and percentage of qualified test items as reference factors:

[0045] ;

[0046] Mean Time Between Failures (MTBF) = ;

[0047] Component integrity rate = ;

[0048] Percentage of items passing the test = ;

[0049] The formula for calculating vehicle health by weighting each of the aforementioned reference factors is as follows:

[0050] Vehicle health rating = w1 × (1 - frequency of failure) + w2 × mean time between failures + w3 × component integrity rate + w4 × percentage of qualified test items;

[0051] Among them, w1, w2, w3, and w4 are preset weight values, and w1+w2+w3+w4=1.

[0052] This module calculates vehicle health by setting four reference factors and weighting them. Taking the JY-001 special vehicle insulated bucket truck as an example, with a statistical period of 30 days, the first step is to calculate each reference factor:

[0053] Fault frequency: JY-001 experienced one hydraulic system fault during the period, with a fault frequency of approximately 0.033 times per day (1 / 30).

[0054] Mean time between failures (MTBF): The duration of hydraulic system failures during the statistical period was 4 hours. The total time without failures was 30 × 24 - 4 = 716 hours. The number of failures was 1. The mean time between failures was 716 / (1 + 1) = 358 hours. The number of failures was increased by 1 to avoid situations where no failures occurred and the calculation could not be performed.

[0055] The component integrity rate is as follows: JY-001 has a total of 6 core components, such as insulated arm, hydraulic system, outriggers, electrical system, engine and wheels. Currently, there are 5 components without faults, so the component integrity rate = 5 / 6 ≈ 0.833.

[0056] The pass rate of the test items is approximately 0.833% because 6 tests were completed within the statistical period and 5 of them passed.

[0057] The final weighted health score is calculated as follows:

[0058] With preset weights w1=0.2, w2=0.3, w3=0.3, and w4=0.2, substituting them into the formula: Vehicle Health Score = 0.2×(1-0.033)+0.3×358+0.3×0.833+0.2×0.833≈0.92, and selecting a percentage system, 92 points is considered a good health score. The vehicle's health status can be directly viewed using this quantified value.

[0059] The fault analysis module includes a fault data preprocessing module and a fault correlation analysis module. The fault data preprocessing module is used to extract the vehicle's fault codes, faulty components, fault duration, and associated sensor data as fault features. Taking the aforementioned insulated bucket truck JY-001 as an example with a statistical period of 30 days, the fault data preprocessing module extracts the fault codes of JY-001, including the hydraulic system fault code 0x0021, the faulty component being the hydraulic oil pipe, the fault duration being 4 hours, and the associated sensor data showing that the faulty component is hydraulic oil pressure lower than the rated value by 30%, as fault features. The fault codes and faulty components are then quantified into feature values ​​in the range of 0-1.

[0060] The fault correlation analysis module includes a density clustering algorithm and a root cause correlation algorithm. The density clustering algorithm performs density clustering on fault features to form clusters of similar faults. The root cause correlation algorithm calculates the support and confidence of each fault cluster to generate association rules between fault events and fault features, outputting the fault events as the fault results. The DBSCAN density clustering algorithm is selected. This algorithm uses two parameters, ε-neighborhood radius and MinPts (minimum number of points), to divide the data into core points, boundary points, and noise points, ultimately forming density-connected clusters. The formula is as follows:

[0061] The ε-neighborhood is: for a sample point Its ε-neighborhood is all the pairs of ε-neighborhoods.

[0062] The core point is: If ≥ MinPts, where MinPts is a preset threshold, then As the core point;

[0063] The density can be achieved if a core point sequence exists. And if each adjacent point is within the ε-neighborhood of the other, then and Density can be achieved;

[0064] A cluster is a set of all mutually density-reachable sample points.

[0065] Five historical fault records of JY-001 were input into the DBSCAN density clustering algorithm. Taking ε neighborhood radius = 0.5 and MinPts = 2 as an example, the five fault records of JY-001 in the past year were clustered to generate hydraulic system fault clusters and insulation component fault clusters.

[0066] Root Cause Association Algorithm: Calculates the support and confidence of the "hydraulic system fault cluster". The support of "faulty component: hydraulic oil pressure 30% below rated value" and "hydraulic system fault" is 30 / 100 = 30%, and the confidence is 30 / 40 = 75%. This generates the association rule "faulty component: hydraulic oil pressure 30% below rated value → hydraulic system fault", which is output as the fault result. The root cause association algorithm is the Apriori algorithm, used to discover association rules between fault events and sensor data. The core metrics are support (the frequency of rule occurrence) and confidence (the reliability of the rule). The calculation formulas for these metrics are as follows:

[0067] Support: Where A is the precondition and B is the fault event;

[0068] Confidence level: .

[0069] Taking JY-001 as an example, 100 sensor data and fault records were taken as a sample, and the occurrence of faulty components such as hydraulic oil pressure being 30% lower than the rated value (A) and hydraulic system faults (B) were statistically analyzed:

[0070] Number of records containing both A and B: 30; Number of records containing A: 40; Total number of records: 100.

[0071] Support: This indicates that the faulty component is a hydraulic oil pressure that is 30% lower than the rated value, and the probability of a hydraulic system failure is 30%.

[0072] Confidence level: This indicates that when the hydraulic oil pressure is 30% lower than the rated value, the probability of a hydraulic system failure is 75%. This correlation can be used as a condition for early warning of the root cause of the failure.

[0073] The prediction module includes a time series data construction module, a prediction model training module, and a prediction output module;

[0074] The time-series data construction module is used to acquire historical time-series data and construct a sample set by combining the vehicle's cumulative running time and ambient temperature data. The sample set is divided into a test set and a training set. Taking JY-001 as an example, the historical leakage current data (0.2mA→0.5mA) of its faulty component, the insulating arm, is collected over 40 days. This data is then combined with the vehicle's cumulative running time and ambient temperature data to construct a sample set, which is divided into a training set and a test set in a 7:3 time ratio. Specific data values ​​will not be listed individually.

[0075] The prediction model training module uses an LSTM prediction model, which includes a forget gate, input gate, cell state gate, and output gate. It uses mean squared error as the loss function and iteratively trains on the training set until the loss function converges to obtain the final prediction model. The core of the LSTM prediction model is to capture long-term dependencies in time series through the gate structure to predict component lifespan or failure time. The LSTM prediction model is set with 16 hidden layer neurons, a learning rate of 0.001, and uses mean squared error as the loss function. It is iteratively trained 50 times on the training set until the loss function converges to obtain the final prediction model. The LSTM prediction model is also an existing neural network model, so its principles will not be elaborated further here.

[0076] The prediction output module is used to input the time series data of the component under test into the final prediction model and output the prediction results. The time series data of the test set is input into the final model, which predicts that the faulty component is the wear of the hydraulic oil pipe. The remaining service life is calculated based on the leakage current threshold; for example, service life = (1.0 - 0.5) / 0.03 ≈ 16.7 days. Simultaneously, the module outputs that the insulating arm is highly likely to fail in the next 17 days.

[0077] Multi-reference factor weighted calculation enables quantitative assessment of vehicle health status, avoiding the subjectivity of human experience judgment; density clustering combined with association rules can accurately locate high-frequency faults and root causes; LSTM model adapts to the prediction needs of time series data, improving the accuracy of component life and failure time prediction, and enabling maintenance work to shift from passive response to proactive prevention.

[0078] The maintenance cost analysis unit includes an association module, an accounting module, and a cost optimization module. The association module is used to establish the association between vehicles, components, and maintenance. The accounting module is used to calculate the full life cycle cost of a vehicle from purchase to retirement. The cost optimization module is used to analyze the maintenance cost of a vehicle and generate an optimized cost plan.

[0079] The association module in the maintenance cost analysis unit is used to establish a mapping relationship between vehicle number, vehicle core components, and maintenance activities, generating a unique association identifier and binding the corresponding data. Taking JY-001 as an example, the unique association identifier is generated using the format of vehicle number + maintenance event type code + timestamp. For example, the hydraulic oil replacement maintenance association identifier for JY-001 is JY-001-01-202X0X1X1030. Association binding: The association identifier is used to bind the basic information of JY-001, the faulty component being the hydraulic system, and the hydraulic oil replacement work order, forming a three-dimensional association data of vehicle-component-maintenance. It also includes an association verification step: automatically verifying the consistency of the association data. If problems such as the vehicle number corresponding to the association identifier not existing are found, it is marked as an association anomaly and the user is notified to handle it.

[0080] The accounting module in the maintenance cost analysis unit is used to obtain the vehicle's purchase cost, operation and maintenance cost, and decommissioning cost through correlation mapping relationships, and calculate the total lifecycle cost. Cost data for each stage is obtained: the purchase cost, operation and maintenance cost, and decommissioning cost of JY-001 are extracted through a correlation data table; the staged cost calculation method is as follows:

[0081] Purchase stage cost = Purchase price 1.8 million yuan + Purchase tax 159,300 yuan + Initial registration fee 3,000 yuan = 1,962,300 yuan;

[0082] Maintenance cost during operation and maintenance phase = cost of each operation and maintenance event = hydraulic oil replacement (oil + labor) 15,000 yuan + cleaning of insulating components 6,000 yuan + grease filling of slewing bearing 3,000 yuan + withstand pressure test fee 8,000 yuan + outrigger maintenance 4,000 yuan = 36,000 yuan / year, estimated to be used for 10 years, total operation and maintenance cost 360,000 yuan;

[0083] The cost of disposal during the decommissioning phase = disposal fee of 20,000 yuan + dismantling fee of 5,000 yuan - residual value recovery of 35,000 yuan = -5,000 yuan;

[0084] Full lifecycle cost calculation: Full lifecycle cost = 196.23 + 36 - 0.5 = 231.73 million yuan.

[0085] The cost optimization module in the maintenance cost analysis unit analyzes the maintenance cost ratio and cost change trends of different vehicles and components. It then combines this with vehicle health analysis data mining to mine the correlation between cost and health status, generating a cost optimization plan that includes maintenance strategy adjustments and component selection optimization. The module also quantifies and predicts the cost savings after optimization. Using the above example as a basis for calculation, a cost analysis is first performed. The analysis of the maintenance cost ratio of different components of JY-001 reveals that the hydraulic system accounts for 41.7% of the maintenance cost, and this cost shows an increasing trend year by year.

[0086] Optimization solution generation: Based on the analysis results, optimization solutions are generated, such as shortening the hydraulic oil replacement cycle from 12 months to 10 months and selecting low-temperature resistant hydraulic oil to adapt to the northern winter environment. The corresponding solution is selected from the solution library and pushed to the system.

[0087] Cost savings: Based on historical data fitting, a 17% reduction in replacement cycle corresponds to a 16% cost contribution rate. It is predicted that the annual maintenance cost will be reduced by 16% after optimization, resulting in an annual saving of 3.6 × 16% = 0.576 million yuan, and a total saving of 5.76 million yuan over 10 years.

[0088] The maintenance cost analysis unit establishes a precise mapping between costs and vehicles and components, solving the problems of fragmented and untraceable cost data; full-cycle cost accounting enables full-process control of vehicle maintenance costs; and cost optimization solutions combined with health data are more targeted, effectively reducing maintenance costs and improving resource utilization efficiency.

[0089] The system also includes a database for storing data from each unit and setting security measures for that data. The database is an essential data storage unit in any system, and the security measures prevent data loss.

[0090] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0091] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A special vehicle full life cycle maintenance management information system, characterized in that, include: The user management unit is used to manage user access permissions; The vehicle information management unit includes a data acquisition module and a file management module. The data acquisition module includes an automatic acquisition module, a manual acquisition module, and a data verification module. The file management module is used to store basic vehicle data and vehicle maintenance data. The vehicle health analysis unit includes a health calculation module, a fault analysis module, and a prediction module. The health calculation module uses the frequency of fault occurrence, mean time between failures, component integrity rate, and the percentage of qualified test items as reference factors to calculate the vehicle health by weighting each of the reference factors. The fault analysis module is used to obtain fault characteristics and output fault results. The prediction module is used to predict the service life of components and the time of fault occurrence. The maintenance cost analysis unit includes an association module, an accounting module, and a cost optimization module. The association module is used to establish the association between vehicles, components, and maintenance. The accounting module is used to calculate the full life cycle cost of a vehicle from purchase to retirement. The cost optimization module is used to analyze the maintenance cost of the vehicle and generate an optimized cost plan.

2. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The automatic acquisition module in the data acquisition module is used to obtain vehicle positioning data through the GPS module, vehicle driving data through the vehicle diagnostic module, and data transmission through the GPRS module. The manual acquisition module is used to call the input system to save the vehicle data entered by the user. The data verification module is used to verify whether the data transmitted by the automatic acquisition module and the manual acquisition module are consistent. If they are consistent, duplicate data is removed. The data verification module also includes sorting all data according to timestamps and sending supplementary reminders to the input system for missing data through the manual acquisition module.

3. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The file management module includes setting classification management dimensions and dynamic update rules, saving the data in the data acquisition module according to the classification management dimensions and updating the data in real time according to the dynamic update rules.

4. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The health status calculation module uses the following formula to calculate the failure frequency, mean time between failures, component integrity rate, and percentage of qualified test items as reference factors: ； Mean Time Between Failures (MTBF) = ; Component integrity rate = ; Percentage of items passing the test = ; The formula for calculating vehicle health by weighting each of the aforementioned reference factors is as follows: Vehicle health rating = w1 × (1 - frequency of failure) + w2 × mean time between failures + w3 × component integrity rate + w4 × percentage of qualified test items; Among them, w1, w2, w3, and w4 are preset weight values, and w1+w2+w3+w4=1.

5. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The fault analysis module includes a fault data preprocessing module and a fault correlation analysis module. The fault data preprocessing module is used to extract the vehicle's fault codes, faulty components, fault duration, and associated sensor data as fault features. The fault association analysis module includes a density clustering algorithm and a root cause association algorithm. The density clustering algorithm is used to perform density clustering on fault features to form fault clusters of the same type. The root cause association algorithm is used to calculate the support and confidence of each fault cluster to generate association rules between fault events and fault features, and output the fault events as fault results.

6. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The prediction module includes a time series data construction module, a prediction model training module, and a prediction output module; The time-series data construction module is used to acquire historical time-series data and construct a sample set by combining the vehicle's cumulative running time and ambient temperature data. The sample set is divided into a test set and a training set. The prediction model training module is an LSTM prediction model, which includes a forget gate, an input gate, a cell state, and an output gate. The mean squared error is used as the loss function, and the model is iteratively trained using the training set until the loss function converges to obtain the final prediction model. The prediction output module is used to input the time series data of the component under test into the final prediction model and output the prediction results.

7. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The association module in the maintenance cost analysis unit is used to establish a mapping relationship between vehicle number, vehicle core components, and maintenance behavior, generate a unique association identifier, and bind the corresponding data.

8. The special vehicle full life cycle maintenance management information system according to claim 7, characterized in that, The accounting module in the maintenance cost analysis unit is used to obtain the vehicle's purchase cost, operation and maintenance cost, and decommissioning cost through association mapping relationships, and calculate the total life cycle cost.

9. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The cost optimization module in the maintenance cost analysis unit analyzes the maintenance cost ratio and cost change trend of different vehicles and different components, and combines vehicle health analysis data to mine the correlation between cost and health status, generating a cost optimization plan that includes maintenance strategy adjustment and component selection optimization, and quantitatively predicting the cost savings after optimization.

10. The special vehicle full life cycle maintenance management information system according to claim 1, characterized in that, The system also includes a database for storing data from each unit and setting security protection strategies for the data.

Images