Light commercial vehicle product access type inspection configuration optimization method and device

Abstract

The application provides a light commercial vehicle product access type test configuration optimization method and device, based on the applicable regulations clause analysis of type test project, based on the analysis result to construct type test configuration parameter library, fill in the product configuration parameter into the parameter library to form a plurality of configuration parameter sets, and construct the association matrix of the record parameter and the type test project, based on the configuration parameter set, the corresponding configuration selection matrix and the association matrix, the optimized configuration parameter set combination is obtained through the model, and the type test scheme table is generated based on the configuration parameter set combination. Compared with the traditional manual configuration method, the following beneficial effects are obtained: through automatic configuration and optimization, manual intervention is reduced, and the efficiency of test scheme formulation is greatly improved; through intelligent algorithm, the test scene is comprehensively covered, omission and error are avoided; the resource configuration is automatically optimized, and repeated testing and resource waste are reduced; the unified configuration process improves the comparability and reliability of the test result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle inspection, in particular to a light commercial vehicle product access type inspection configuration optimization method and device. BACKGROUND

[0002] With the rapid development of the automotive industry, product access type inspection of light commercial vehicles has become a key link to ensure that vehicles meet safety, environmental protection and performance standards. Access type inspection is a mandatory inspection that must be passed when applying for vehicle production enterprises and product announcements, mandatory product certification and environmental protection information disclosure. It usually involves multiple test items such as emissions, noise, braking performance, etc. Inspection parameters and processes need to be configured according to regulations and vehicle characteristics. At present, the common implementation scheme mainly relies on manual experience or fixed templates for inspection configuration, which has the following deficiencies: first, due to reliance on manual operation for configuration, the inspection configuration efficiency is low, prolonging the product launch cycle; second, the detection configuration accuracy is insufficient, as reliance on manual operation, it is easy to lead to detection item omission or repetition, affecting the reliability of the test results; third, due to reliance on manual experience, resource configuration is not optimal, resulting in waste of manpower, material resources and time.

[0003] At present, there is a method for generating optimal solution and feasible solution of actual measurement scheme for multiple filing declaration values. The scheme is to form an entry system by logically associating and classifying parameters according to enterprise filing information, based on the preset logic of the entry system, and combined with the binding relationship, the project configuration information can be obtained; filter the project configuration that cannot be covered by the report, integrate these configurations that need to be measured, only keep the parameters related to the measured items and the parameters with multiple declaration values in the results, form the final measured analysis configuration, and gradually combine to find the measured scheme that best meets the requirements. The limitation of this scheme is that it does not classify and integrate the equivalent judgment principles specified in the regulations, but only compares with the existing measurement reports of the enterprise; it cannot output the generated vehicle configuration matrix, only generates a type inspection scheme table, which cannot be used to guide the actual application of enterprises. SUMMARY

[0004] To solve the above problems, the application provides a light commercial vehicle product access type test configuration optimization method and device, which introduces intelligent optimization algorithm and automation process to realize efficient, accurate and adaptive test configuration, thereby reducing test cost and improving product access speed. In the application, based on the judgment logic relationship between the commercial vehicle configuration parameters and the type test items, an association relationship model of vehicle configuration parameters and test items is constructed to represent the influence relationship of different vehicle configuration parameters on the triggering and judgment results of each test item; according to the model, a test vehicle combination model meeting the type test regulation constraints is constructed, wherein the test vehicle combination model is used to describe the corresponding relationship between different vehicle configuration combinations and test item coverage; under the condition of meeting all type test item coverage requirements, the test vehicle combination model is executed to solve the multi-objective optimization, taking the test vehicle number and test item coverage as the optimization target, to generate at least one vehicle configuration combination scheme for type test.

[0005] Specifically, the application adopts the following technical solutions:

[0006] The application provides a light commercial vehicle product access type test configuration optimization method, which has the following technical features: the method comprises the following steps: step S1, based on the predetermined type test items of the light commercial vehicle product, the applicable type judgment regulation clauses are analyzed to obtain the analysis result; step S2, based on the analysis result, a type test configuration parameter library for the light commercial vehicle product is constructed, which contains multiple judgment conditions, each judgment condition contains a record parameter and a judgment logic; step S3, multiple sets of product configuration parameters of the light commercial vehicle product to be tested are obtained, each product configuration parameter is filled into the corresponding record parameter of the type test configuration parameter library to form multiple configuration parameter sets, and a configuration selection matrix is set for each configuration parameter set to select several configuration parameter sets; step S4, based on the judgment logic in the type test configuration parameter library, an association matrix of each record parameter and each type test item is constructed; step S5, based on multiple configuration parameter sets, the configuration selection matrix and the association matrix, an optimized configuration parameter set combination scheme is obtained through a configuration optimization model; step S6, a type test scheme table is generated based on the optimized configuration parameter set combination scheme.

[0007] The light commercial vehicle product access type test configuration optimization method provided by the application can also have the following technical features: in step S1, the record parameters are divided into three predetermined parameter types, including discrete parameters, continuous parameters, and composite parameters, and a unique parameter code, value range, and judgment logic are set for each parameter type. The value of the discrete parameter is a plurality of enumerable discrete values, a specific code or number is assigned to each discrete value, the value range is a set containing all discrete values, and the judgment logic includes that the discrete value of the discrete parameter of the type-expandable vehicle model is the same as the discrete value of the corresponding discrete parameter of the base vehicle model, and the expression of the judgment condition is formed based on the discrete value and the judgment logic. The value of the continuous parameter is a numerical value, the value range is a predetermined numerical value range, and the judgment logic includes that the numerical value of the continuous parameter of the type-expandable vehicle model satisfies a predetermined comparison rule with the numerical value of the corresponding continuous parameter of the base vehicle model, the comparison rule includes greater than, greater than or equal to, less than, less than or equal to, within a certain proportion range, and a combination of any two or more of the above comparison rules, and the expression of the judgment condition is formed based on the numerical value and the judgment logic. The composite parameter contains a plurality of basic parameters, the basic parameters are the discrete parameters or the continuous parameters, the expressions of the judgment conditions corresponding to the plurality of basic parameters are logically ANDed to form the expression of the composite judgment condition.

[0008] The light commercial vehicle product access type test configuration optimization method provided by the application can also have the following technical features: in step S4, for the type test items related to the discrete parameters or the continuous parameters, a mapping relationship among the record parameters, the judgment conditions, and the type test items is established based on the direct relationship between the type test items and the set of record parameters. For the type test items related to the composite parameters, a set of basic parameters corresponding to the composite parameters is obtained, each logical node of each basic parameter is analyzed one by one, and it is judged whether the change of any basic parameter will change the trigger condition of the type test item. If the answer is yes, the corresponding basic parameter is added to the parameter set, and thus a complete set of influence parameters of the type test item is obtained. The association matrix is represented as: R = [r ij ], where when p i ∈A j , r ij = 1; when p i ∉A j , r ij = 0.

[0009] The method for optimizing the configuration of type inspection for light commercial vehicle products provided by this invention may also have the following technical features: In step S2, the constructed type inspection configuration parameter library contains multiple records, each record representing a judgment condition. The record includes a judgment condition identifier, filing parameters, and judgment logic. The record contains multiple fields. The fields corresponding to the filing parameters include parameter code, parameter name, data type, value range, and unit. The fields corresponding to the judgment logic include comparison operators, requirements for the vehicle model to be inspected, requirements for the basic vehicle model, and inspection necessity.

[0010] The configuration optimization method for type approval testing of light commercial vehicles provided by this invention may also have the following technical features: In step S3, multiple differentiated configuration parameter sets are sorted according to a predetermined sorting rule, and a configuration selection matrix is ​​set for the sorted configuration parameter sets. The configuration selection matrix is ​​a vector matrix, represented as: X = [X1, X2, ..., X...]. N ], where X i This is a binary variable whose value represents whether to select configuration parameter set i as the configuration parameters for the test sample vehicle of the light commercial vehicle product, and N is the total number of test sample vehicles.

[0011] The configuration optimization method for type testing of light commercial vehicle products provided by the present invention may also have the following technical features: in step S5, the configuration optimization model uses an improved second-generation non-dominated sorting genetic algorithm to solve multi-objective optimization problems, and obtains the optimal Pareto solution set as the optimized configuration parameter set combination scheme. The optimization objectives of the second-generation non-dominated sorting genetic algorithm include minimizing the number of configuration parameter sets to be tested and maximizing the coverage of the type testing items.

[0012] The light commercial vehicle product access type test configuration optimization method provided by the application can also have the following technical features: step S5 includes the following sub-steps: step S5-1, constructing a plurality of population individuals based on a plurality of configuration parameter sets, each population individual representing a selected state of a corresponding configuration parameter set, thereby initializing a population; step S5-2, calculating the fitness of each population individual based on a predetermined fitness function, the fitness function including two independent objectives: counting the number of selected configuration parameter sets in the population to achieve the objective of minimizing the number of configuration parameter sets for testing; calculating the proportion of type test items that the population can cover according to the correlation matrix to achieve the objective of maximizing the coverage of type test items; step S5-3, non-dominant sorting of population individuals based on fitness, dividing them into a plurality of Pareto levels according to the dominance relationship between population individuals; step S5-4, for each population individual in each Pareto level, combining the fitness and the distance correction factor based on the difference degree of the configuration parameter set to calculate the crowding distance; step S5-5, selecting part of the population individuals from the population as parent individuals based on the crowding distance and a predetermined selection rule to form a parent population, and performing crossover and mutation operations on the parent individuals to generate offspring individuals to form an offspring population, wherein the mutation probability of key parameters affecting the type determination regulation clauses is reduced, the mutation probability of other parameters except the key parameters is increased, and the global mutation rate is adjusted with the number of iterations; step S5-6, calculating the fitness of the offspring individuals based on the fitness function; step S5-7, determining whether a predetermined iteration stop condition is reached, stopping iteration if the determination is yes, and obtaining a Pareto optimal solution set; step S5-8, filtering the Pareto optimal solution set according to the preferences of an enterprise for test cost, item robustness, and / or test declaration simplification degree, thereby obtaining an optimized configuration parameter set combination scheme; step S5-9, if the determination in step S5-7 is no, merging the parent population and the offspring population as a new population, and returning to step S5-3.

[0013] The light commercial vehicle product access type test configuration optimization method provided by the application can also have the following technical features: in step S6, generating a type test sample vehicle list, a corresponding type test item list, a configuration parameter difference matrix, a deemed applicable matrix, and a type test scheme table based on the optimized configuration parameter set combination scheme.

[0014] The method for optimizing the configuration of type testing for light commercial vehicle products provided by the present invention may also have the following technical features: the method further includes: step S7, using historical type testing data to verify the generated type testing scheme table, detecting whether there are conflicts in the table entries of the type testing scheme table, and if so, adjusting the logical structure of the type configuration parameter library and / or optimizing the weights of the configuration optimization model according to the detected conflicts, and returning to step S2.

[0015] This invention provides a configuration optimization device for type inspection of light commercial vehicle products, characterized by the following technical features: A regulatory judgment condition analysis module, used to analyze applicable type inspection regulatory clauses based on predetermined type inspection items for light commercial vehicle products, obtaining analysis results; a parameter library construction module, used to construct a type inspection configuration parameter library for the light commercial vehicle products based on the analysis results, containing multiple judgment conditions, each judgment condition including filing parameters and judgment logic; and a vehicle configuration input module, used to acquire multiple sets of product configuration parameters for the light commercial vehicle products to be inspected, and to input the various product configuration parameters... The system fills parameters into the filing parameters corresponding to the type test configuration parameter library, forming multiple configuration parameter sets, and sets a configuration selection matrix for selecting several of the configuration parameter sets; a configuration difference identification module is used to construct an association matrix between each filing parameter and each type test item based on the judgment logic in the type test configuration parameter library; a configuration optimization model is used to obtain an optimized configuration parameter set combination scheme based on the multiple configuration parameter sets, the configuration selection matrix, and the association matrix; and a test scheme generation module is used to generate a type test scheme table based on the optimized configuration parameter set combination scheme.

[0016] The role and effect of invention

[0017] The method and apparatus for optimizing configuration for type testing of light commercial vehicles provided by this invention analyzes applicable type determination regulations based on type testing items, constructs a type testing configuration parameter library based on the analysis results, fills the configuration parameters of the product to be tested into this parameter library to form multiple configuration parameter sets, and constructs an association matrix between the filing parameters and type testing items. Based on the configuration parameter sets, the corresponding configuration selection matrix, and the association matrix, an optimized configuration parameter set combination scheme is obtained through a model, and a type testing scheme table is generated based on the configuration parameter set combination scheme. Compared with traditional manual configuration methods, it has the following advantages: 1. Improved testing efficiency: Through automated configuration and optimization, manual intervention is reduced, significantly improving the efficiency of test scheme formulation; 2. Improved configuration accuracy: Intelligent algorithms can comprehensively cover testing scenarios, avoiding omissions and errors; 3. Reduced testing costs: It can automatically optimize resource allocation, reducing duplicate testing and resource waste; 4. Promoted standardization: Unified configuration process, improving the comparability and reliability of test results. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the configuration optimization device for type approval testing of light commercial vehicles in an embodiment of the present invention;

[0019] Figure 2 This is a flowchart of the configuration optimization method for type approval testing of light commercial vehicles in this embodiment of the invention;

[0020] Figure 3 This is a flowchart of step S5 in an embodiment of the present invention;

[0021] Figure 4 This is an example diagram of a data table in the type testing configuration parameter library in an embodiment of the present invention;

[0022] Figure 5 This is a comparative diagram of the actual test vehicles required for the two methods under different configuration complexities in the embodiments of the present invention.

[0023] Figure label:

[0024] The system includes: a configuration optimization device for type approval testing of light commercial vehicles; a regulatory judgment condition analysis module; a parameter library construction module; a vehicle configuration input module; a configuration difference identification module; a configuration optimization model; an inspection plan generation module; an inspection plan verification module; and a process control module. Detailed Implementation

[0025] To make the technical means, creative features, objectives and effects of this invention easy to understand, the invention will be specifically described below in conjunction with embodiments and accompanying drawings.

[0026] Example

[0027] Figure 1 This is a schematic diagram of the configuration optimization device for type approval testing of light commercial vehicles in this embodiment.

[0028] like Figure 1 As shown, the configuration optimization device 10 for type approval inspection of light commercial vehicle products in this embodiment includes: a regulatory judgment condition analysis module 11, a parameter library construction module 12, a vehicle configuration input module 13, a configuration difference identification module 14, a configuration optimization model 15, an inspection plan generation module 16, an inspection plan verification module 17, and a process control module 18.

[0029] The regulatory determination condition analysis module 11 is used to analyze the applicable type determination regulatory clauses based on the type inspection items of the predetermined light commercial vehicle products in order to obtain the analysis results.

[0030] The parameter library construction module 12 is used to construct a type test configuration parameter library based on the analysis results output by the regulatory judgment condition analysis module 11.

[0031] The vehicle configuration input module 13 is used to obtain multiple sets of vehicle configuration parameters of the light commercial vehicle product to be processed, and fill the vehicle configuration parameters into the type test configuration parameter library constructed by the parameter library construction module 12 to form multiple configuration parameter sets, and set a configuration selection matrix for them to select and enable several of the configuration parameter sets.

[0032] The configuration difference identification module 14 is used to construct the association matrix between each configuration parameter set and the type test items based on the relevant judgment logic in the type test configuration parameter library.

[0033] Configuration optimization model 15 is used to generate optimized configuration parameter set combination schemes based on multiple configuration parameter sets, configuration selection matrices, and association matrices.

[0034] The test plan generation module 16 is used to generate a type test prototype list, a corresponding type test item list, a configuration parameter difference matrix, a deemed applicable matrix, and a type test plan table based on the optimized configuration parameter set combination scheme.

[0035] The inspection plan verification module 17 is used to verify the generated type inspection plan table using historical type inspection data, detect whether there are conflicts in the table entries, and adjust the logical structure of the type inspection configuration parameter library and / or optimize the weights of the configuration optimization model based on the detected conflicts when a conflict is found.

[0036] The process control module 18 is used to control the corresponding modules and models to work in sequence according to the predetermined workflow order and branch judgment logic, thereby realizing automated type inspection configuration optimization and generation.

[0037] The specific algorithms involved in each of the above modules will be explained in detail in the methods section below.

[0038] Figure 2 This is a flowchart of the configuration optimization method for type approval testing of light commercial vehicle products in this embodiment.

[0039] like Figure 2 As shown, based on the above-mentioned light commercial vehicle product access type testing configuration optimization device 10, the corresponding method includes the following steps:

[0040] Step S1: Analyze the applicable judgment regulations based on the predetermined type inspection items for light commercial vehicle products to obtain the analysis results.

[0041] In this step, the regulatory judgment condition parsing module 11 retrieves the text of the applicable type judgment regulatory clauses based on the text of the type test item and decomposes it to extract the text of multiple judgment conditions involved in the light commercial vehicle product. The judgment conditions include filing parameters and judgment logic, and the text of the judgment conditions (i.e., the judgment conditions described in natural language) is converted into corresponding Boolean expressions or mathematical inequalities to obtain the parsing results.

[0042] More specifically, in this step, the filing parameters are divided into three predetermined types, and a unique parameter code, value range, and judgment logic are set for each type of filing parameter, thereby constructing a computable type detection configuration parameter library. Specifically, regarding the classification and coding methods of the filing parameters, based on the characteristics of the judgment conditions, the filing parameters can be divided into three main categories and subjected to corresponding parameterization processing: discrete parameters, continuous parameters, and composite parameters.

[0043] The values ​​of discrete parameters are finite, enumerable options. The judgment logic is qualitative, and the result is "yes" or "no". Specifically, the corresponding parameterization method is as follows: A unique parameter code is generated for each discrete parameter according to a preset encoding rule. The unique parameter code can contain letters and / or numbers, and a specific code or number is assigned to each discrete value of the discrete parameter (i.e., each option). The value range of the discrete parameter is the set containing all its discrete values. The judgment logic for discrete parameters is qualitative, and the judgment logic can include: the discrete value of the discrete parameter of the type-extendable vehicle is the same as the discrete value of the corresponding discrete parameter of the base vehicle. The discrete values ​​and the corresponding judgment logic can form an expression for the judgment condition.

[0044] For example, the name of a discrete filing parameter is: suspension type, and the unique parameter code is: QB018; the data type is defined as enumeration type (Categorical); the value range (optional value range) is: 1, 2, 3, 4, where 1 represents leaf spring, 2 represents air spring, 3 represents non-independent suspension, and 4 represents independent suspension; the judgment logic is: the discrete value of this filing parameter (suspension type) of the type-extendable model must be exactly the same as the discrete value of this filing parameter of the base model. For example, if the discrete value of both is 1, it means that the suspension type of both is leaf spring.

[0045] For example, another type of discrete filing parameter is named: Service Braking System Type; the unique parameter code is: QE012; the data type is defined as enumeration type; the value range is: 1, 2, where 1 represents hydraulic braking and 2 represents air braking; the judgment logic is: the discrete value of this filing parameter (service braking system type) of the type-extendable vehicle model must be exactly the same as the discrete value of this filing parameter of the base vehicle model. For example, if the discrete values ​​of both are 1, it means that the service braking system type of both is hydraulic braking.

[0046] Continuous parameters are defined by specific numerical values. The judgment logic is a quantitative comparison, requiring comparison of the magnitude of the values ​​or calculation of the rate of change. Specifically, the corresponding parameterization method is as follows: A unique parameter code is generated for each continuous parameter according to preset coding rules. The parameter code can contain letters and / or numbers and records the specific numerical value of the continuous parameter. The value range of the continuous parameter can be a preset numerical range or a range defined by the maximum and minimum values ​​among all continuous parameters. Discrete parameters are judged using quantitative judgment logic. This logic can include: the specific numerical value of the parameter in the type-extendable vehicle model and the specific numerical value of the parameter in the base vehicle model satisfy a predetermined comparison rule. The comparison rule can include: greater than (>), greater than or equal to (≥), equal to (=), less than (<), less than or equal to (≤), within a certain percentage range (±%), or any combination of two or more of the above rules. An expression is formed based on the specific numerical value and the comparison rule to determine the judgment condition.

[0047] For example, the name of the continuous filing parameter is: engine power; the parameter code is: QC0011; the data type is defined as floating point (Float), the unit is kW; the value range is: 80kW-150kW; the judgment logic is: if the value of the filing parameter of the expandable model is ≥ the value of the filing parameter of the base model, it means that the engine power of the expandable model must be greater than or equal to the engine power of the base model.

[0048] A composite parameter comprises multiple basic parameters, which can be discrete or continuous. The decision logic can be a logical combination judgment, meaning that certain decision conditions require multiple parameters to simultaneously satisfy the corresponding decision rules, forming an "AND" logical relationship. Specifically, the corresponding parameterization method is as follows: each composite parameter is decomposed into multiple basic parameters, and each basic parameter is parameterized according to the parameterization method for discrete or continuous parameters described above, resulting in multiple expressions. These expressions are then subjected to a logical "AND" operation to generate a composite decision condition expression.

[0049] For example, a type test item requires that the braking system of the type-extendable vehicle model is the same as that of the base vehicle model. The judgment regulations corresponding to this type test item can be broken down into multiple judgment conditions. Accordingly, the parameters involved are composite parameters, including four basic parameters: service braking system type (QE012), emergency braking system type (QE013), parking braking system type (QE014), and auxiliary braking system type (QE015), all of which are discrete parameters. The decision logic for each discrete parameter is the same as above. After performing a logical AND operation on the conditional expression of each discrete parameter, the resulting composite conditional expression is: (QE012_new == QE012_base) AND (QE013_new == QE013_base) AND (QE014_new == QE014_base) AND (QE015_new == QE015_base). In this expression, QE012_new represents the discrete value of the service braking system type of the expandable vehicle model, QE012_base represents the discrete value of the service braking system type of the base vehicle model, and so on.

[0050] Step S2: Construct a type inspection configuration parameter library for light commercial vehicle products based on the analysis results. The library contains multiple judgment conditions, each of which includes filing parameters and judgment logic.

[0051] By classifying the above three parameter types and performing corresponding parameterization processing in step S1, calculable parameterization results are obtained. In this step, the parameter library construction module 12 constructs a type verification configuration parameter library based on these parameterization results.

[0052] Figure 4 This is an example diagram of the data table in the type testing configuration parameter library in this embodiment.

[0053] like Figure 4As illustrated, the type testing configuration parameter library can, for example, contain parameter information in the form of a table. Each record in the data table represents a judgment condition, and each record contains a judgment condition identifier (ID), the corresponding filing parameter, and judgment logic information. Each record contains multiple fields. Fields related to the filing parameter include parameter code, parameter name, data type, value range, and unit. Fields related to the judgment logic include comparison operators, target vehicle model (i.e., the vehicle model of the light commercial vehicle product to be tested) requirements, basic vehicle model requirements, and test necessity (whether the judgment condition is mandatory). A computable expression for a judgment condition can be generated based on the information in each record. In other words, the type testing configuration parameter library is equivalent to a template for configurable parameters, used to define and standardize the input data required for testing.

[0054] Step S3: Obtain multiple sets of product configuration parameters for the light commercial vehicle product to be tested, fill each product configuration parameter into the corresponding filing parameters in the type test configuration parameter library to form multiple configuration parameter sets, and set a configuration selection matrix for them to select and enable several configuration parameter sets.

[0055] In this step, the vehicle configuration input module 13 obtains basic information about multiple light commercial vehicle models to be processed. This basic information includes product configuration parameters defined by the enterprise. Following the logical structure of the type testing configuration parameter library, each parameter in each set of product configuration parameters is entered as a value into the corresponding field, forming a standardized configuration parameter set. The light commercial vehicle products to be tested can have multiple models, each corresponding to a different configuration parameter set. Multiple differentiated configuration parameter sets can be automatically sorted according to a predetermined sorting rule, and a configuration selection matrix is ​​set for the sorted sets to select and activate several of them. The activated configuration parameter sets then serve as the input source for subsequent steps.

[0056] The configuration selection matrix can be a vector matrix, represented as: X = [X1, X2, ..., X...]. N ], where X i This is a binary variable, taking the value 0 or 1. The value represents whether the sorted configuration parameter set i is selected as the configuration parameters for a test vehicle of the light commercial vehicle product. X i = 1 indicates that the configuration parameter set i is selected as the configuration parameter of the test sample vehicle; otherwise, its value is 0. N is the total number of test sample vehicles.

[0057] Step S4: Based on the judgment logic in the type test configuration parameter library, construct the association matrix between each filing parameter and each type test item.

[0058] The correlation matrix can be represented as: R = [r ij ], where r ij Indicates the filing parameter p i The influence relationship on type test item j.

[0059] In the above steps, for each type test item j, the text of the corresponding type determination regulation clause was retrieved, and the determination conditions triggered by that type test item were extracted. These determination conditions include filing parameters (such as mass, axle moment, suspension type, etc.), determination logic (such as "consistent," "not greater than," etc.), and condition association methods (i.e., AND, OR, NOT, etc.). This transforms the natural language descriptions of the determination conditions into computable expressions, yielding the condition for type test item j in relation to the parameter set P. j The direct relationship between the parameter set P and the parameter set P j It contains multiple sets of configuration parameters.

[0060] In this step, for type test item j involving discrete or continuous parameters, the difference identification module 14 is configured to analyze the parameter set P according to the type test item j. j The direct relationship between filing parameters, judgment conditions, and type test items is established. For type test item j involving composite parameters, the following steps are also performed: obtain the set of basic parameters (i.e., its sub-parameter set) corresponding to the composite parameter, parse the logical nodes of each basic parameter one by one, and determine whether any change in a basic parameter will change the triggering condition of the type test item. If the determination is yes, the corresponding basic parameter is added to the parameter set P. j Thus, the complete set of influencing parameters A for type test item j is obtained. j That is, the dependency list of all filing parameters that type test item j depends on.

[0061] Complete set of influencing parameters A j It can be represented as: A j = {p i1 , p i2 , …, p ik}, where p ik This refers to the k-th filing parameter upon which type testing item j depends. Construct the association matrix R = [r] ij ], for all filing parameters p i And type test item j, fill matrix element r according to the following rules ij When p i ∈A j When setting r ij = 1; when p i ∉A j When setting r ij= 0. The final correlation matrix is ​​an M×N binary matrix, where M is the number of filing parameters and N is the number of type test items.

[0062] Step S5: Based on multiple configuration parameter sets, configuration selection matrix, and association matrix, an optimized configuration parameter set combination scheme is obtained through a configuration optimization model.

[0063] In this step, the configuration optimization model 15 employs an improved second-generation non-dominated sorting (NSGA-II) genetic algorithm for multi-objective optimization, obtaining the optimal Pareto solution set as the optimized configuration parameter set combination scheme. The optimization objectives of this genetic algorithm include minimizing the number of configuration parameter sets to be inspected and maximizing the coverage of type inspection items. That is, it simultaneously satisfies the optimization requirements of "minimum number of inspected vehicles" and "maximum inspection item coverage."

[0064] Figure 3 This is a flowchart of step S5 in this embodiment.

[0065] like Figure 3 As shown, specifically, step S5 includes the following sub-steps:

[0066] Step S5-1: Construct multiple population individuals based on multiple configuration parameter sets, with each population individual representing a selection state of a corresponding configuration parameter set, thereby initializing the population.

[0067] In this step, population individuals for genetic evolution are constructed based on the differential parameter encoding rules of each configuration parameter set (equivalent to multiple vehicle models with different configuration parameters). Each population individual represents the selection state of different configuration parameter sets in the form of a structured binary vector, so that the combination scheme of multiple different configuration parameter sets can be directly mapped to an operable genetic code.

[0068] Step S5-2: Calculate the fitness of each individual in the population based on a predetermined fitness function, which is used to evaluate the quality of each individual in the population.

[0069] In this step, the fitness function includes two independent objectives: first, to count the number of configuration parameter sets selected in the current population (i.e., the combination scheme of multiple configuration parameter sets) to achieve the goal of minimizing the number of vehicles; second, to calculate the proportion of the number of type test items that the current population can cover to the total number of type test items based on the correlation matrix to achieve the goal of maximizing the coverage of type test items.

[0070] Step S5-3: Perform non-dominated ranking of individuals in the population based on fitness, and divide the population into multiple Pareto levels according to the dominance relationship between individuals.

[0071] Step S5-4: For each individual in the population at each Pareto level, calculate the crowding distance by combining fitness and a distance correction factor based on the dissimilarity of the configuration parameter set, in order to measure the sparsity of the population in the target space.

[0072] In this step, a distance correction factor based on the difference in configuration parameter sets is introduced on the basis of the traditional congestion distance calculation method. This is to strengthen the impact of the difference in configuration parameter sets of different vehicle models on the diversity of the overall scheme, so that the evolution process of the genetic algorithm can more fully explore the combination structure of representative configuration parameter sets, thereby enhancing the stability and adaptability of the optimization results.

[0073] Step S5-5: Based on the crowding distance and predetermined selection rules, select some individuals from the population as parent individuals to form the parent population, and perform crossover and mutation operations on the parent individuals to generate offspring individuals to form the offspring population.

[0074] In this step, the selection rule can be, for example, prioritizing non-dominant levels and optimizing for those of the same level based on the crowding distance, and selecting a predetermined number of parent individuals according to this selection rule.

[0075] During the crossover and mutation phase, a dynamic adaptive mutation rate control strategy based on parameter sensitivity is adopted. The mutation probability is reduced for key parameters that affect the regulatory determination results, and the mutation probability is increased for parameters with low influence (i.e., other parameters besides the key parameters mentioned above). The global mutation rate is automatically adjusted with the number of iterations. For example, the global mutation rate increases with the number of iterations to improve evolutionary efficiency and search effect.

[0076] Steps S5-6: Calculate the fitness of each offspring individual based on the predetermined fitness function.

[0077] Step S5-7: Determine whether the predetermined iteration stopping condition has been met. If it is determined that the iteration is stopped, the Pareto optimal solution set is obtained, which is the optimized combination of configuration parameters.

[0078] In this step, the iteration stopping condition can be, for example, reaching a predetermined maximum number of iterations or reaching a predetermined convergence condition.

[0079] Steps S5-8 introduce a firm preference strategy into the Pareto optimal solution set. Based on the firm's pre-defined preferences for inspection costs, project robustness, and / or the simplification of inspection declaration (i.e., according to different focuses), the Pareto optimal solution set is further filtered to output the optimal combination of configuration parameters that can be directly used in actual type inspection declaration scenarios.

[0080] Step S5-9: If the determination in step S5-7 is negative, merge the parent population and the offspring population into a new population, and return to step S5-3. That is, repeat the following process: perform non-dominated sorting of the individuals in the new population based on fitness, calculate crowding distance, perform crossover and mutation to generate the offspring population, and merge the parent population and the offspring population to generate a new population, until the predetermined iteration stopping condition is reached to obtain the Pareto optimal solution set.

[0081] Through the above step S5, not only is the regulatory consistency and executability of the optimization results guaranteed, but also the efficiency of multi-objective optimization and the practical value of the final optimal configuration parameter set combination scheme are significantly improved through improvements such as difference correction, penalty mechanism and adaptive mutation strategy.

[0082] Step S6: Generate a type test prototype list, a corresponding type test item list, a configuration parameter difference matrix, a deemed applicable matrix, and a type test scheme table based on the optimized configuration parameter set combination scheme.

[0083] In this step, the inspection scheme generation module 16 generates, based on the optimized configuration parameter set combination scheme and according to a predetermined template or algorithm, a combination of type inspection vehicles (type inspection prototype vehicle list), a corresponding list of type inspection items, a configuration parameter difference matrix between multiple prototype vehicles, and a deemed applicability matrix, and combines these lists and matrices into a type inspection scheme table. This step can use relevant templates and algorithms in the prior art.

[0084] Step S7: Verify the generated type test scheme table using historical type test data, check for conflicts in the table entries, and if a conflict is found, adjust the logical structure of the type test configuration parameter library and / or optimize the weights of the configuration optimization model based on the detected conflict, and return to step S2 to regenerate the configuration parameter set combination scheme; if a conflict is found, use the current configuration parameter set combination scheme as the final scheme.

[0085] In this step, when a conflict is detected, the verification module 17 automatically adjusts the logical structure of the type test configuration parameter library and / or optimizes the model weights so that the conflict no longer occurs in the regenerated configuration parameter set combination scheme, and returns to step S2, that is, the type test configuration is re-optimized and generated using the adjusted parameter library and / or model, thereby forming a closed-loop update scheme.

[0086] The process sequence and branch judgment in the above method steps are controlled by the process control module 18.

[0087] To further verify the technical effectiveness of the method in this embodiment, the inventors constructed a typical configuration space for light commercial vehicles containing different numbers of configurable parameters and conducted a simulation comparison analysis between the method in this embodiment and the traditional manual type inspection configuration method.

[0088] In the simulation, the number of type test items is kept constant, and the number of variable configuration parameters is increased to simulate the change in vehicle configuration complexity. The number of test vehicles required to meet the requirements of all type test items is also counted.

[0089] Figure 5 This is a comparative diagram of the number of test vehicles required for the two methods under different configuration complexities in this embodiment. In this diagram, the horizontal axis represents the number of variable configurable parameters, and the vertical axis represents the number of test vehicles required for configuration. The solid line represents a line graph showing the number of test vehicles required for configuration using the traditional manual type inspection configuration method as a function of configurable parameters, while the dashed line represents a line graph showing the number of test vehicles required for configuration using the method in this embodiment as a function of configurable parameters.

[0090] from Figure 5 The comparison shows that as the number of variable configuration parameters increases, the number of test vehicles required by the traditional manual method increases significantly. However, the method in this embodiment can still maintain a low number of test vehicles even with increased configuration complexity. This indicates that the method in this embodiment significantly reduces the number of test vehicles required for type testing while ensuring that the coverage of type testing items is not reduced.

[0091] The role and effect of the embodiments

[0092] According to the method and apparatus for optimizing configuration for type testing of light commercial vehicles provided in this embodiment, the applicable type determination regulations are analyzed based on the type testing items, and a type testing configuration parameter library is constructed based on the analysis results. The configuration parameters of the product to be tested are filled into this parameter library to form multiple configuration parameter sets, and a correlation matrix between the filing parameters and the type testing items is constructed. Based on the configuration parameter sets, the corresponding configuration selection matrix, and the correlation matrix, an optimized configuration parameter set combination scheme is obtained through a model, and a type testing scheme table is generated based on the configuration parameter set combination scheme. Compared with traditional manual configuration methods, it has the following advantages: 1. Improved testing efficiency: Through automated configuration and optimization, manual intervention is reduced, significantly improving the efficiency of test scheme formulation; 2. Improved configuration accuracy: Intelligent algorithms can comprehensively cover testing scenarios, avoiding omissions and errors; 3. Reduced testing costs: Resource allocation can be automatically optimized, reducing duplicate testing and resource waste; 4. Promoted standardization: Unified configuration process, improving the comparability and reliability of test results.

[0093] In this embodiment, the configuration optimization model uses an improved NSGA-II genetic algorithm to solve multi-objective optimization problems and obtain an optimized configuration parameter set combination scheme. In this genetic algorithm, the evolution efficiency and search effect can be improved by adopting a dynamic adaptive mutation rate control strategy based on parameter sensitivity in the crossover and mutation stages. Furthermore, by introducing an enterprise preference strategy into the optimal solution set, the optimal type inspection vehicle combination scheme that can be directly used in actual inspection and declaration scenarios can be output.

[0094] Furthermore, since the type test scheme table is checked using historical type test data after it is generated to detect whether there are any conflicts in the table entries, and if a conflict is found, the logical structure of the parameter library is automatically adjusted or the model weights are optimized, a more intelligent type test configuration scheme with closed-loop updates can be formed.

[0095] The above embodiments are merely illustrative of specific implementations of the present invention, and the present invention is not limited to the scope of the description of the above embodiments. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are only for illustrating the principles of the present invention. Various changes and modifications can be made to the present invention without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing the configuration of type approval testing for light commercial vehicles, characterized in that, Includes the following steps: Step S1: Based on the predetermined type test items for light commercial vehicle products, analyze the applicable type determination regulations and obtain the analysis results; Step S2: Based on the analysis results, construct a type inspection configuration parameter library for the light commercial vehicle product, which includes multiple judgment conditions, each of which includes filing parameters and judgment logic; Step S3: Obtain multiple sets of product configuration parameters for the light commercial vehicle product to be inspected, fill each of the product configuration parameters into the filing parameters corresponding to the type inspection configuration parameter library to form multiple configuration parameter sets, and set a configuration selection matrix for them to select several of the configuration parameter sets; Step S4: Based on the judgment logic in the type test configuration parameter library, construct the association matrix between each of the filing parameters and each of the type test items; Step S5: Based on the multiple configuration parameter sets, the configuration selection matrix, and the association matrix, an optimized configuration parameter set combination scheme is obtained through a configuration optimization model; Step S6: Generate a type testing scheme table based on the optimized configuration parameter set combination scheme. In step S1, the text of the applicable type-determination regulations is retrieved based on the text of the type test item, and the text of the determination conditions related to the light commercial vehicle product is extracted from it. The filing parameters are divided into three predetermined parameter types, including discrete parameters, continuous parameters, and composite parameters. A unique parameter code, value range, and determination logic are set for each parameter type. The composite parameter includes multiple basic parameters, which can be either the discrete parameter or the continuous parameter. In step S4, for the type test items involving the discrete parameters or the continuous parameters, a mapping relationship is established between the registration parameters, the judgment conditions, and the type test items based on the direct relationship between the type test items and the registration parameter set. For the type test item involving the composite parameter, obtain the set of basic parameters corresponding to the composite parameter, parse the logical nodes of each basic parameter one by one, and determine whether a change in any basic parameter will change the triggering condition of the type test item. If the determination is yes, add the corresponding basic parameter to the parameter set, thereby obtaining the complete set of influencing parameters for the type test item. The correlation matrix is ​​represented as: R = [r ij ], where, when p i ∈A j At that time, r ij = 1; when p i ∉A j At that time, r ij = 0.

2. The method for optimizing the configuration of type approval testing for light commercial vehicles according to claim 1, characterized in that: in, The discrete parameter has multiple enumerable discrete values, each assigned a specific code or number. Its value range is a set containing all discrete values. The determination logic includes: the discrete value of the discrete parameter of the type-extendable vehicle model is the same as the discrete value of the corresponding discrete parameter of the base vehicle model; and an expression for the determination condition is constructed based on the discrete value and the determination logic. The value of the continuous parameter is a numerical value, and its value range is a preset numerical range. The determination logic includes: the value of the continuous parameter of the expandable vehicle model and the value of the corresponding continuous parameter of the base vehicle model satisfy a predetermined comparison rule. The comparison rule includes greater than, greater than or equal to, less than, less than or equal to, within a certain proportion range, and any combination of two or more of the above comparison rules. Based on the numerical value and the determination logic, an expression for the determination condition is constructed. For the composite parameter, the expressions of the determination conditions corresponding to multiple basic parameters are logically ANDed to form the expression of the composite determination condition.

3. The method for optimizing the configuration of light commercial vehicle product access type testing according to claim 1, characterized in that: in, In step S2, the constructed type test configuration parameter library contains multiple records, each of which represents a judgment condition. The record includes a judgment condition identifier, filing parameters, and judgment logic. The record contains multiple fields. The fields corresponding to the filing parameters include parameter code, parameter name, data type, value range and its unit. The fields corresponding to the judgment logic include comparison operator, requirements for the vehicle model to be inspected, basic vehicle model requirements and inspection necessity.

4. The method for optimizing the configuration of type approval testing for light commercial vehicles according to claim 1, characterized in that: in, In step S3, the multiple differentiated configuration parameter sets are sorted according to a predetermined sorting rule, and the configuration selection matrix is ​​set for the sorted configuration parameter sets. The configuration selection matrix is ​​a vector matrix, represented as: X = [X1, X2, ..., X...]. N ], where X i This is a binary variable whose value represents whether to select configuration parameter set i as the configuration parameters for the test sample vehicle of the light commercial vehicle product, and N is the total number of test sample vehicles.

5. The method for optimizing the configuration of light commercial vehicle product access type testing according to claim 1, characterized in that: in, In step S5, the configuration optimization model uses an improved second-generation non-dominated sorting genetic algorithm to solve for multi-objective optimization, obtaining the optimal Pareto solution set as the optimized configuration parameter set combination scheme. The optimization objectives of the second-generation non-dominated sorting genetic algorithm include minimizing the number of configuration parameter sets to be tested and maximizing the coverage of the type test items.

6. The method for optimizing the configuration of type approval testing for light commercial vehicles according to claim 5. Its features are: Step S5 includes the following sub-steps: Step S5-1: Construct multiple population individuals based on multiple configuration parameter sets, where each population individual represents a selection state of a corresponding configuration parameter set, thereby initializing the population; Step S5-2: Calculate the fitness of each individual in the population based on a predetermined fitness function. The fitness function includes two independent objectives: to count the number of configuration parameter sets selected in the population to minimize the number of configuration parameter sets to be tested; and to calculate the proportion of type test items that the population can cover based on the correlation matrix to maximize the coverage of type test items. Step S5-3: Based on the fitness, perform non-dominated ranking on the individuals in the population, and divide them into multiple Pareto levels according to the dominance relationship between the individuals in the population; Step S5-4: For each individual in the population at each Pareto level, calculate the crowding distance by combining the fitness and the distance correction factor based on the dissimilarity of the configuration parameter set. Step S5-5: Based on the crowding distance and predetermined selection rules, select a portion of the individuals in the population as parent individuals to form a parent population, and perform crossover and mutation operations on the parent individuals to generate offspring individuals to form an offspring population. In this process, reduce the mutation probability of key parameters that affect the type determination regulations, increase the mutation probability of other parameters besides the key parameters, and adjust the global mutation rate with the number of iterations. Steps S5-6: Calculate the fitness of the offspring individuals based on the fitness function; Step S5-7: Determine whether the predetermined iteration stopping condition has been met. If the condition is met, stop the iteration and obtain the Pareto optimal solution set. Steps S5-8: Based on the predetermined enterprise's preferences for inspection costs, project robustness, and / or the simplification of inspection declarations, the Pareto optimal solution set is screened to obtain the optimized configuration parameter set combination scheme. Step S5-9: If the determination in step S5-7 is negative, merge the parent population and the offspring population to form a new population, and return to step S5-3.

7. The method for optimizing the configuration of light commercial vehicle product access type testing according to claim 1, characterized in that: in, In step S6, a type test prototype list, a corresponding type test item list, a configuration parameter difference matrix, a deemed applicable matrix, and a type test scheme table are generated based on the optimized configuration parameter set combination scheme.

8. The method for optimizing the configuration of type approval testing for light commercial vehicles according to claim 1, characterized in that, Also includes: Step S7: Verify the generated type test scheme table using historical type test data, detect whether there are conflicts in the table entries, and if so, adjust the logical structure of the type test configuration parameter library and / or optimize the weights of the configuration optimization model based on the detected conflicts, and return to step S2.

9. A configuration optimization device for type approval testing of light commercial vehicles, employing the configuration optimization method for type approval testing of light commercial vehicles as described in any one of claims 1-8, characterized in that, include: The regulatory determination condition analysis module is used to analyze the applicable type determination regulatory clauses based on the predetermined type inspection items of light commercial vehicle products, and obtain the analysis results. The parameter library construction module is used to construct a type inspection configuration parameter library for the light commercial vehicle product based on the parsing results. It contains multiple judgment conditions, and each judgment condition includes filing parameters and judgment logic. The vehicle configuration input module is used to obtain multiple sets of product configuration parameters of the light commercial vehicle product to be inspected, fill each of the product configuration parameters into the filing parameters corresponding to the type inspection configuration parameter library, form multiple configuration parameter sets, and set a configuration selection matrix for them to select several of the configuration parameter sets. A configuration difference identification module is configured to construct an association matrix between each of the filing parameters and each of the type inspection items based on the judgment logic in the type inspection configuration parameter library. A configuration optimization model is used to obtain an optimized combination scheme of configuration parameter sets based on multiple configuration parameter sets, the configuration selection matrix, and the correlation matrix. as well as The test plan generation module is used to generate a type test plan table based on the optimized configuration parameter set combination scheme.

Images