Intelligent automobile authentication test method and system based on cooperation of various laws and regulations

Abstract

The invention belongs to the technical field of automobile testing, and provides an intelligent automobile authentication testing method and system based on cooperation of multiple laws and regulations. According to the technical scheme, data obtained through analysis are stored according to a hierarchical storage architecture to obtain a multi-laws and regulation database; constructing an association knowledge graph of the vehicle parameters and the regulation test items, and traversing the test items matched with the core attribute parameters of the vehicle type to be tested in the knowledge graph; dynamically adjusting the weight of each test item according to the priority of the target market, and obtaining a test list according to the screening associated with the test items; acquiring real-time feedback data of each test item in the test list, calling a test parameter configuration strategy corresponding to the target area, and adjusting configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy; and training the constructed compliance risk prediction model, predicting the configured test parameters based on the trained compliance risk prediction model, and generating a critical test condition conforming to the law and regulation boundary conditions. And the test of multi-regulation scenes is met.

Description

Technical Field

[0001] The present invention belongs to the technical field of automotive testing, and particularly relates to an intelligent vehicle certification testing method and system based on the coordination of multiple regulations. Background Technique

[0002] The statements in this part only provide background technical information related to the present invention and do not necessarily constitute prior art.

[0003] The test platforms for vehicles in different regions need to match the certification standards of the corresponding regions. However, the current vehicle certification platforms use manual parsing when analyzing the certification standards, making it difficult to track version updates. To solve this problem, intelligent and automated testing methods have been proposed. However, the existing automated testing methods have the problems that device parameters need to be repeatedly configured during cross-regional standard certification, the testing cycle for a single vehicle model with multiple regulations is long, and there is a lack of mining of multi-source heterogeneous data. They are aimed at use case generation and script execution under a single regulation and scenario, and cannot meet the scenarios of multi-regulation system testing, resulting in low testing efficiency. Summary of the Invention

[0004] To solve at least one of the technical problems in the above background technique, the present invention provides an intelligent vehicle certification testing method and system based on the coordination of multiple certification standards, which integrates data of multiple vehicle models and multiple regulations to generate critical test conditions that meet the legal boundary conditions and can meet the testing scenarios of multiple regulations.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] The first aspect of the present invention provides an intelligent vehicle certification testing method based on the coordination of multiple regulations, including the following steps:

[0007] Analyze the key test items and parameter constraint data in the obtained multiple regulation documents, and store the analyzed data in a hierarchical storage architecture to obtain a multi-regulation database;

[0008] Combined with the multi-regulation database, construct an association knowledge graph of vehicle parameters and regulation test items, obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle to be tested, and traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle to be tested; dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening associated with the test item;

[0009] Obtain the real-time feedback data of each test item in the test list, retrieve the test parameter configuration strategy corresponding to the target region, and adjust the configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy;

[0010] Train the constructed compliance risk prediction model based on historical test data, and predict the configured test parameters based on the trained compliance risk prediction model to generate critical test conditions that meet the regulatory boundary conditions.

[0011] Furthermore, construct an association knowledge graph between vehicle parameters and regulatory test items, including:

[0012] Define the nodes of the knowledge graph and construct three types of core nodes: vehicle components, regulatory test items, and constraint rules;

[0013] Define the relationships of the knowledge graph and establish logical relationships based on a multi-regulatory database with a hierarchical storage architecture;

[0014] Import the nodes and relationships into the Neo4j graph database to form a dynamically scalable network structure; among them, each association is attached with an initial weight value.

[0015] Furthermore, dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening of the associated test items, including:

[0016] Statistically analyze the association frequency between each test item and the target market based on historical test data;

[0017] Dynamically update the weight of each test item by combining the target market priority coefficient and the timeliness factor;

[0018] After obtaining the matching test items, give priority to recommending key test items with high weight values, and automatically eliminate unnecessary test items with a repetition rate greater than the set value according to the historical data of vehicle enterprises to obtain a test list.

[0019] Furthermore, obtain the real-time feedback data of each test item in the test list, and retrieve the test parameter configuration strategy corresponding to the target area, including:

[0020] When the target area is the first area, dynamically adjust and control to generate configuration parameters that meet the regulatory requirements of the first area by combining the real-time feedback data and the fuzzy PID control algorithm;

[0021] When the target areas are the second area and the third area, dynamically adjust the configuration of the test parameters by combining the real-time feedback data and the collaborative optimization strategy of test efficiency and measurement accuracy.

[0022] Furthermore, when the target areas are the second area and the third area, perform optimization based on the NSGA-II algorithm, specifically including:

[0023] Define a multi-objective optimization function, including two items. The first item is to minimize the test time, and the second item is that the measurement error ≤ 1%; define the decision variable: test parameter combination;

[0024] Perform population initialization, crossover and mutation, and non-dominated sorting, and finally output the Pareto optimal solution set, and select the corresponding test parameter configuration scheme according to the requirements of test efficiency and measurement accuracy.

[0025] Further, the process of constructing the compliance risk prediction model includes:

[0026] Input the preprocessed historical test data into a 3-layer LSTM unit to capture long-term dependencies, input the output of the LSTM unit into the Attention layer, calculate the attention scores of each time step, and extract key abnormal points.

[0027] Further, the hierarchical storage architecture adopts a three-level logical framework of test category → test item → threshold parameter.

[0028] The second aspect of the present invention provides an intelligent vehicle certification test system based on the coordination of multiple regulations, including:

[0029] A multi-regulation database construction module, which is used to parse and obtain key test items and parameter constraint data in the obtained multiple regulation documents, and store the parsed data according to the hierarchical storage architecture to obtain a multi-regulation database;

[0030] A test item screening module, which is used to combine the multi-regulation database to construct an association knowledge graph of vehicle parameters and regulation test items, obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle to be tested, and traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle to be tested; dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening associated with the test item;

[0031] A test parameter configuration module, which is used to obtain the real-time feedback data of each test item in the test list, retrieve the test parameter configuration strategy corresponding to the target area, and adjust the configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy;

[0032] A boundary condition generation module, which is used to train the constructed compliance risk prediction model based on historical test data, predict the configured test parameters based on the trained compliance risk prediction model, and generate critical test conditions that meet the regulatory boundary conditions.

[0033] The third aspect of the present invention provides a computer-readable storage medium.

[0034] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps in the above-mentioned intelligent vehicle certification test method based on the coordination of multiple regulations.

[0035] The fourth aspect of the present invention provides a computer device.

[0036] A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in the intelligent vehicle certification test method based on the coordination of multiple regulations as described above.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] 1. By constructing a database of over-regulation, combining multiple regulation databases, dynamically adjusting the weight of each test item according to the priority of the target market, screening out important test items, configuring test parameters based on the feedback data of the test items, predicting the configured test parameters, and generating critical test conditions that meet the regulatory boundary conditions, the present invention effectively overcomes the technical limitations of the traditional threshold warning method in predicting implicit risks and breaks through the inherent defects of the experience-driven detection mode.

[0039] 2. The present invention retrieves the test parameter configuration strategy corresponding to the target area according to the regulations of different target areas, solving the problems of long time consumption and cumulative system errors caused by repeated parameter configuration due to regulation switching in traditional test equipment.

[0040] Advantages of additional aspects of the present invention will be partially given in the following description, partially become apparent from the following description, or be understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments and descriptions thereof of the present invention are used to explain the present invention and do not constitute an improper limitation to the present invention.

[0042] Figure 1 is a flowchart of an intelligent vehicle certification test method based on the coordination of multiple regulations provided by an embodiment of the present invention;

[0043] Figure 2 is a flowchart of the operation of an intelligent matching engine provided by an embodiment of the present invention;

[0044] Figure 3 is an algorithm logic diagram of a data analysis module provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0046] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0047] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0048] Embodiment 1

[0049] As Figure 1 and Figure 2 shown, this embodiment provides an intelligent vehicle certification test method based on the coordination of multiple certification standards, including the following steps:

[0050] Step 1: Analyze and obtain the key test items and parameter constraint data in the obtained multiple regulatory documents, and store the parsed data in a hierarchical storage architecture to obtain a multi-regulatory database;

[0051] In this embodiment, when analyzing and obtaining multiple regulatory documents, the built-in Natural Language Processing (NLP) technology is used to analyze the regulatory documents. For example, pre-trained models such as BERT can be used to obtain the structured data;

[0052] Taking the implementation of the ECE R100 certification test as an example, the BERT model is used to perform semantic segmentation on the regulatory text to identify keyword fields such as "test item" and "parameter threshold";

[0053] In this embodiment, the hierarchical storage architecture adopts a three-level logical framework of "test category → test item → threshold parameter", and supports a dynamic update mechanism in XML / JSON format;

[0054] For example: Extract "Overcharge protection trigger time ≤ 5 seconds" from ECE R100 and map it to the structured fields in the database: Test_Item: Overcharge_Protection; Threshold: 5s.

[0055] Through this technical solution, the problems of lagging update and data deviation existing in traditional manual entry are effectively solved. After actual measurement, a significant 60% improvement in the regulatory analysis efficiency can be achieved, and the technical ability to synchronize the latest regulatory version in real time is available.

[0056] Step 2: Construct an associated knowledge graph of vehicle parameters and regulatory test items, obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle model to be tested, and traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle model to be tested; Dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening associated with the test item;

[0057] As Figure 3 shown, it specifically includes the following steps:

[0058] Step 201: Construct an associated knowledge graph of vehicle parameters and regulatory test items;

[0059] Specifically, it includes the following steps:

[0060] Define the nodes of the knowledge graph, and construct three types of core nodes: vehicle components (such as batteries, vehicle bodies), regulatory test items (such as overcharge protection tests), and constraint rules (such as trigger time ≤ 5 seconds)

[0061] Define the relationships of the knowledge graph, and establish logical relationships based on a multi-regulatory database with a hierarchical storage architecture. For example, "ternary lithium battery → overcharge protection test needs to be performed → trigger time threshold ≤ 5 seconds".

[0062] Import the nodes and relationships into the Neo4j graph database to form a dynamically scalable network structure; Among them, each association is attached with an initial weight value, and the initial weight is obtained from the statistics of historical compliance data;

[0063] Step 202: Obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle model to be tested;

[0064] Specifically, existing VIN code parsing technologies can be used to extract production information (such as model year, power type) in the vehicle unique identifier, and associate with the enterprise database to obtain key parameters such as battery capacity and body material;

[0065] Step 203: Traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle model to be tested;

[0066] For example, call the VIN code parsing interface to obtain the battery type (such as NCM ternary lithium), and retrieve the associated "overcharge test" and "pinprick test" in the graph.

[0067] Step 204: After obtaining the matching test items, dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening associated with the test item;

[0068] In this embodiment, dynamically adjusting the weight of each test item according to the priority of the target market specifically includes:

[0069] Step 2041: Based on historical test data, statistically calculate the association frequency between each test item and the target market. For example, under the EU market, the initial weight of the "crash test" is 0.8, and the "EMC test" is 0.6.

[0070] Step 2042: Dynamically update the weights of each test item by combining the target market priority coefficient and the timeliness factor;

[0071] Specifically, the weight update formula is:

[0072] Final weight = Basic weight × (1 + Market priority coefficient) × Timeliness factor,

[0073] For example, market priority: If the proportion of recent EU orders of a car company > 70%, the weights of EU-related test items are automatically multiplied by 1.2;

[0074] Timeliness factor: The weights of test items for newly released regulations are increased by 50% within 3 months. For example, the overcharge test weight of the 2023 version of ECE R100 is increased from 0.7 to 1.05;

[0075] Step 2043: After obtaining the matching test items, give priority to recommending key test items with high weight values. According to the historical data of the car company, automatically eliminate non-essential test items with a duplication rate > 90%;

[0076] After each test, reverse-correct the weights according to the result confidence level (such as a test pass rate of 95%). The weights of test items with an error rate > 5% are reduced by 10%.

[0077] Step 3: Obtain the real-time feedback data of each test item in the test list, retrieve the test parameter configuration strategy corresponding to the target area, and adjust the configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy;

[0078] Specifically, it includes the following steps:

[0079] Step 301: Obtain the real-time feedback data of each test item in the test list, such as the dynamic response curve of the hydraulic loading force of the brake test bench;

[0080] Step 302: Retrieve the test parameter configuration strategy corresponding to the target area;

[0081] Step 3021: When the target area is the first area, dynamically regulate and generate configuration parameters that meet the regulatory requirements of the first area by combining the real-time feedback data and the fuzzy PID control algorithm;

[0082] In this embodiment, the regulatory requirements of the first area can be the EU ECE R100 certification; for example, deploy the fuzzy PID control algorithm to dynamically regulate the overcharge test current waveform, and ensure that the timeliness of power cut-off accurately meets the regulatory requirement of "completing circuit interruption within 5 seconds" through the real-time feedback mechanism.

[0083] Specifically, the fuzzy PID control algorithm can achieve dynamic regulation by using the existing fuzzy PID control algorithm.

[0084] Step 3022: When the target area is the second area and the third area, the configuration of the test parameters is dynamically adjusted by combining the real-time feedback data and the collaborative optimization strategy of test efficiency and measurement accuracy.

[0085] Specifically, it includes the following steps:

[0086] Step 30221: Define a multi-objective optimization function based on the NSGA-II algorithm.

[0087] The objective function includes two terms. The first term is to minimize the test time (efficiency), and the second term is that the measurement error ≤ 1% (accuracy).

[0088] Step 30222: Define decision variables: test parameter combinations (such as braking pressure, temperature range).

[0089] Step 30223: Population initialization: Randomly generate 50 groups of parameter combinations to cover the legal value ranges of FMVSS 305 and GB 18384.

[0090] Crossover and mutation: Use simulated binary crossover (SBX) and polynomial mutation to retain the high-quality solutions on the Pareto front.

[0091] Non-dominated sorting: Select 20% of the elite solutions in each round of iteration. For example, a certain group of parameters (braking pressure 80 bar, temperature 25 °C) simultaneously meets the test time of 35 hours and the error of 0.8%.

[0092] Step 30224: Output the Pareto optimal solution set, and vehicle manufacturers can select the "efficiency first" or "accuracy first" scheme based on their needs.

[0093] For example, if it is necessary to simultaneously meet the parallel test requirements of the EU Euro VI E stage and the Chinese National VI b emission standards, multi-objective optimization is performed on the collision acceleration thresholds of FMVSS 305 and GB 18384 based on the multi-objective genetic algorithm (NSGA-II) to generate the global optimal test parameter combination. Using the same collision test bench, the dynamic switching of the control strategy is realized through a programmable logic controller, and the full-frontal collision condition of 50 km / h in the US standard and the 40% offset collision test condition in the national standard are accurately simulated respectively.

[0094] Through the above method, the problems of long time consumption and cumulative system errors caused by the need to repeatedly configure parameters due to regulation switching in traditional test equipment can be effectively solved.

[0095] Step 4: Train the constructed compliance risk prediction model based on historical test data, and predict the configured test parameters based on the trained compliance risk prediction model to generate critical test conditions that meet the regulatory boundary conditions;

[0096] Specifically, it includes the following steps:

[0097] Step 401: Preprocess the obtained historical test data;

[0098] Specifically, extract time series features from the historical test data, including voltage, temperature, internal resistance, etc., slice them into 500-dimensional vectors at 10-second intervals; use cubic spline interpolation to fill in the missing values;

[0099] Step 402: Construct a compliance risk prediction model;

[0100] In this embodiment, the compliance risk prediction model adopts the LSTM-Attention model,

[0101] Input the preprocessed historical test data into a 3-layer LSTM unit (128 neurons) to capture long-term dependencies, input the output of the LSTM unit into the Attention layer, calculate the attention scores at each time step, and extract key abnormal points, such as the weight in the stage of sudden temperature rise is increased to 0.9.

[0102] Step 403: Train the compliance risk prediction model;

[0103] For example, input: battery data for 24 consecutive hours; output: the probability of thermal runaway in the next 1 hour (in the range of 0-1); loss function: weighted cross-entropy, assign 3 times the loss weight to high-risk samples (probability > 0.7).

[0104] Step 404: Simulate extreme scenarios to verify the early warning ability of the model under extreme conditions;

[0105] In this embodiment, use GAN to generate critical condition data (such as temperature 45°C + voltage sudden drop of 10%) to verify the early warning ability of the model under extreme conditions.

[0106] For example, when the model detects that the temperature of a certain battery cell rises by 4°C within 30 minutes and the volatility > 15%, trigger a secondary warning and start the cooling system.

[0107] By constructing the above-mentioned extreme scenario simulation engine driven by the generative adversarial network (GAN), generate critical test conditions that meet the regulatory boundary conditions, complete the system robustness verification, effectively overcome the technical limitations of traditional threshold warning methods in implicit risk prediction, and break through the inherent defects of the experience-driven detection mode.

[0108] The empirical study takes the ECE R100 regulation as the analysis object:

[0109] The traditional manual parsing mode requires engineers to read through 50 pages of regulation text and manually mark key test items, with an average of 8 hours consumed for a single regulation; the NLP-based automated parsing system fine-tunes the BERT model (the training data is from a corpus of 1,000 annotated regulation samples), and the time consumed to complete the same regulation parsing work is reduced to 3.2 hours. Based on this calculation:

[0110] The efficiency improvement rate = (8 - 3.2) ÷ 8 × 100% = 60%,

[0111] Verification plan: Through comparative experiments on 10 typical regulation documents (including ECE R100, FMVSS 305, etc.), the cumulative time consumed by the manual parsing group is 80 hours, and the cumulative time consumed by the NLP parsing group is 32 hours, with the error controlled within ±5%.

[0112] Taking a cross-border sales vehicle model (VIN code: LDYECS32*R0037667) as an example, it is required to simultaneously meet the regulatory requirements of China, the United States, and Europe. The traditional manual screening method takes an average of 4 hours per vehicle model, while the intelligent generation system based on the Neo4j graph database only needs 0.8 hours per vehicle model.

[0113] Quantitative formula for efficiency improvement: The time reduction rate = (4 - 0.8) / 4 × 100% = 80%,

[0114] Verification methodology: Select 100 test vehicle model samples for a control experiment. The average time consumed by the manual group is 4.2 hours (σ = 0.28), and the average time consumed by the system group is 0.85 hours (σ = 0.25). The data dispersion is lower than the 0.3-hour threshold.

[0115] Effectiveness verification: The comprehensive efficiency improvement of cross-regulation testing reaches 40%.

[0116] Empirical analysis: Taking a double-standard certification project of a multinational automobile enterprise as an example, the single-regulation test cycle of the traditional single-regulation test is 50 hours; after optimization, the adaptive multi-regulation test cycle is shortened to 30 hours.

[0117] Efficiency improvement calculation formula: (Benchmark cycle 50h - Optimized cycle 30h) ÷ Benchmark cycle 50h × 100% = 40%

[0118] Experimental verification: Through comparative experiments on 10 groups of double-standard test cases, the average test duration of the traditional test method is 51.5 hours (σ = 2.1), and the average test duration of the optimized system is reduced to 30.2 hours (σ = 1.8).

[0119] The empirical results show that the accuracy rate of risk early warning can reach 95% (95% confidence interval [92.1%, 97.3%]), showing a significant improvement compared with traditional methods.

[0120] Typical case analysis: During the testing process of a certain type of battery pack, the traditional threshold mechanism missed 3 cases of thermal runaway risks (miss rate 15%), while the LSTM-Attention model successfully early warned 38 effective risk events (40 samples).

[0121] Accuracy calculation: 38 / 40 × 100% = 95%

[0122] Experimental verification: The model was cross-validated using the publicly available battery aging dataset NASA. This dataset covers the performance changes of batteries under different usage conditions, simulating real aging scenarios. The test shows that the comprehensive performance of the model in distinguishing normal and risky states of the battery is 0.97, significantly better than 0.82 of traditional methods.

[0123] Example 2

[0124] This example provides an intelligent vehicle certification test system based on the coordination of multiple regulations, including:

[0125] A multi-regulation database construction module, which is used to parse and obtain key test items and parameter constraint data in the obtained multiple regulation documents, and store the parsed data according to a hierarchical storage architecture to obtain a multi-regulation database;

[0126] A test item screening module, which is used to combine the multi-regulation database to construct an association knowledge graph of vehicle parameters and regulation test items, obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle to be tested, and traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle to be tested; dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening associated with the test items;

[0127] A test parameter configuration module, which is used to obtain the real-time feedback data of each test item in the test list, retrieve the test parameter configuration strategy corresponding to the target area, and adjust the configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy;

[0128] A boundary condition generation module, which is used to train the constructed compliance risk prediction model based on historical test data, predict the configured test parameters based on the trained compliance risk prediction model, and generate critical test conditions that meet the regulatory boundary conditions.

[0129] It should be noted that the specific implementation of the intelligent vehicle certification test system based on the coordination of multiple regulations in the embodiments of the present invention is similar to that of the intelligent vehicle certification test method based on the coordination of multiple regulations in the embodiments of the present invention. For specific details, please refer to the description in the method section. To avoid redundancy, it will not be elaborated here.

[0130] Embodiment III

[0131] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps in the intelligent vehicle certification test based on the coordination of multiple regulations as described above.

[0132] Embodiment IV

[0133] This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in the intelligent vehicle certification test method based on the coordination of multiple regulations as described above.

[0134] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0135] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0136] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1The functions specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in Figure 1 One process or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes.

[0138] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above various methods. Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0139] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent vehicle certification test method based on the coordination of multiple regulations, characterized in that, Including the following steps: Parse the key test items and parameter constraint data in the obtained multiple regulation documents, and store the parsed data according to the hierarchical storage architecture to obtain a multi-regulation database; Combine the multi-regulation database to construct an association knowledge graph of vehicle parameters and regulation test items, obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle to be tested, and traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle to be tested; Dynamically adjust the weight of each test item according to the priority of the target market, and filter according to the association of the test items to obtain a test list; Obtain the real-time feedback data of each test item in the test list, retrieve the test parameter configuration strategy corresponding to the target area, and adjust the configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy; Train the constructed compliance risk prediction model based on historical test data, predict the configured test parameters based on the trained compliance risk prediction model, and generate critical test conditions that meet the regulatory boundary conditions.

2. The intelligent vehicle certification test method based on the collaboration of multiple regulations according to claim 1, wherein Construct an association knowledge graph of vehicle parameters and regulation test items, including: Define the nodes of the knowledge graph, and construct three types of core nodes: vehicle components, regulation test items, and constraint rules; Define the relationships of the knowledge graph, and establish logical relationships based on the multi-regulation database with a hierarchical storage architecture; Import the nodes and relationships into the Neo4j graph database to form a dynamically expandable network structure; among them, each association is attached with an initial weight value.

3. The intelligent vehicle certification test method based on the coordination of multiple regulations according to claim 1, characterized in that The dynamically adjusting the weight of each test item according to the priority of the target market and filtering according to the association of the test items to obtain a test list includes: Statistically analyze the association frequency of each test item and the target market based on historical test data; Dynamically update the weight of each test item by combining the priority coefficient of the target market and the time effect factor; After obtaining the matching test items, give priority to recommending key test items with high weight values, automatically eliminate unnecessary test items with a repetition rate greater than the set value according to the historical data of vehicle enterprises, and filter to obtain a test list.

4. The intelligent vehicle certification test method based on the collaboration of multiple regulations according to claim 1, characterized in that The obtaining the real-time feedback data of each test item in the test list and retrieving the test parameter configuration strategy corresponding to the target area includes: When the target area is the first area, dynamically adjust and generate configuration parameters that meet the regulatory requirements of the first area by combining the real-time feedback data and the fuzzy PID control algorithm; When the target areas are the second area and the third area, dynamically adjust the configuration of the test parameters by combining the real-time feedback data and the collaborative optimization strategy of test efficiency and measurement accuracy.

5. The intelligent vehicle certification test method based on the collaboration of multiple regulations according to claim 4, characterized in that When the target areas are the second area and the third area, perform optimization based on the NSGA-II algorithm, specifically including: Define a multi-objective optimization function, including two items. The first item is to minimize the test time, and the second item is that the measurement error ≤ 1%; define the decision variable: test parameter combination; Perform population initialization, crossover and mutation, and non-dominated sorting, and finally output the Pareto optimal solution set, and select the corresponding test parameter configuration scheme according to the requirements of test efficiency and measurement accuracy.

6. The intelligent vehicle certification test method based on the collaboration of multiple regulations according to claim 1, characterized in that The process of constructing the compliance risk prediction model includes: The preprocessed historical test data is input into a 3-layer LSTM unit to capture long-term dependencies. The output of the LSTM unit is input into the Attention layer to calculate the attention scores at each time step and extract key abnormal points.

7. The intelligent vehicle certification test method based on the collaboration of multiple regulations according to claim 1, wherein The hierarchical storage architecture adopts a three-level logical framework of test category → test item → threshold parameter.

8. An intelligent vehicle certification test system based on the coordination of multiple regulations, characterized in that, It includes: A multi-regulatory database construction module, which is used to parse the key test items and parameter constraint data in the obtained multiple regulatory documents, and store the parsed data according to the hierarchical storage architecture to obtain a multi-regulatory database; A test item screening module, which is used to combine the multi-regulatory database to construct an association knowledge graph of vehicle parameters and regulatory test items, obtain the core attribute parameters of the vehicle to be tested according to the VIN code of the vehicle to be tested, and traverse the test items in the knowledge graph that match the core attribute parameters of the vehicle to be tested; dynamically adjust the weight of each test item according to the priority of the target market, and obtain a test list according to the screening associated with the test item; A test parameter configuration module, which is used to obtain the real-time feedback data of each test item in the test list, retrieve the test parameter configuration strategy corresponding to the target area, and adjust the configuration parameters based on the real-time feedback data and the corresponding test parameter configuration strategy; A boundary condition generation module, which is used to train the constructed compliance risk prediction model based on historical test data, predict the configured test parameters based on the trained compliance risk prediction model, and generate critical test conditions that meet the regulatory boundary conditions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps in the intelligent vehicle certification test method based on the coordination of multiple regulations described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the intelligent vehicle certification test method based on the coordination of multiple regulations described in any one of claims 1-7.

Images