Calibration parameter evaluation method, system and equipment of new energy vehicle and storage medium

Abstract

The invention relates to the technical field of new energy vehicles, and particularly provides a calibration parameter evaluation method, system and device for a new energy vehicle and a storage medium, and the method comprises the steps: collecting and preprocessing the multi-source heterogeneous operation data of the new energy vehicle; based on the preprocessed data, constructing a multi-dimensional evaluation index system covering safety, economy, dynamic property, comfort and reliability; determining the comprehensive weight of each index in the evaluation index system by adopting a subjective and objective combination weighting method; and in combination with the comprehensive weight, performing quantitative evaluation on the multi-parameter calibration performance of the vehicle through a fuzzy comprehensive evaluation method to obtain a comprehensive evaluation result. Through a multi-source heterogeneous data acquisition and preprocessing mechanism, the integrity and reliability of vehicle operation data are effectively improved, and a solid foundation is laid for performance evaluation.

Description

Technical Field

[0001] This invention belongs to the field of new energy vehicle technology, specifically relating to a method, system, equipment, and storage medium for evaluating calibration parameters of new energy vehicles. Background Technology

[0002] With the deepening development of the new energy vehicle industry, the multi-parameter calibration of electronic control units is crucial for optimizing the overall vehicle performance. Currently, industry evaluations of vehicle performance largely rely on the analysis of specific systems or single performance indicators, such as focusing only on battery energy consumption or acceleration performance. Such methods struggle to systematically reflect the complex relationships of coupling and inverse interactions among the various subsystems during vehicle operation. This results in a one-sided evaluation of the overall vehicle calibration performance, failing to provide calibration engineers with comprehensive and quantitative optimization criteria, thus hindering the collaborative optimization of overall vehicle performance and the improvement of development efficiency. Summary of the Invention

[0003] In view of the above-mentioned shortcomings of the prior art, the present invention provides a method, system, device and storage medium for evaluating calibration parameters of new energy vehicles to solve the above-mentioned technical problems.

[0004] In a first aspect, the present invention provides a method for evaluating calibration parameters of new energy vehicles, comprising: Collect and preprocess multi-source heterogeneous operation data of new energy vehicles; Based on the preprocessed data, a multi-dimensional evaluation index system covering safety, economy, power, comfort and reliability is constructed. The comprehensive weight of each indicator in the evaluation index system is determined by using a combination of subjective and objective weighting methods. Based on the aforementioned comprehensive weights, the multi-parameter calibration performance of the vehicle is quantitatively evaluated using the fuzzy comprehensive evaluation method to obtain the comprehensive evaluation result.

[0005] In one optional implementation, the multi-source heterogeneous operating data of new energy vehicles is collected and preprocessed, including: Data is collected through a distributed sensor network and includes the following: battery system parameters, including voltage, current, temperature, and SOC; motor system parameters, including speed, torque, and power; electronic control system parameters, including input current, voltage, and controller temperature; and vehicle operating parameters, including vehicle speed, acceleration, and energy consumption. Null value processing, multi-period data synchronization and normalization are performed on the collected multi-source heterogeneous operation data; The null value processing includes dynamic masking based on parameter configuration mapping table and correlation filtering based on vehicle operating status; the multi-cycle data synchronization includes using sliding window interpolation to perform linear interpolation on continuous signals and neighbor interpolation on discrete signals.

[0006] In one optional implementation, based on the preprocessed data, a multi-dimensional evaluation index system covering safety, economy, power, comfort, and reliability is constructed, including: Based on vehicle performance goals, at least one quantifiable performance indicator is configured for each of the five dimensions: safety, economy, power, comfort, and reliability. The safety dimension includes at least one of the following indicators: battery overcharge / over-discharge protection response time and power system insulation resistance value; the economy dimension includes at least one of the following indicators: energy consumption per 100 kilometers and energy recovery ratio.

[0007] In one optional implementation, the quantifiable performance metrics configured for each dimension include: The safety dimension is configured with indicators including at least one of the following: battery overcharge / overdischarge protection response time, power system insulation resistance, and motor overload protection sensitivity. The indicators configured in the economic dimension include at least one of the following: energy consumption per 100 kilometers, energy recovery ratio, and thermal management energy consumption. The performance metrics include at least one of the following: maximum torque, 0-50km / h acceleration time, and top speed. The indicators configured for the comfort dimension include at least one of the following: three-dimensional vibration amplitude of the whole vehicle and acceleration / deceleration impact. The reliability dimension is configured with at least one of the following metrics: functional failure rate and percentage of time spent in power / torque limiting state.

[0008] In one optional implementation, a combined subjective and objective weighting method is used to determine the comprehensive weight of each indicator in the evaluation index system, including: Using the analytic hierarchy process, a judgment matrix is ​​constructed to reflect the relative importance of each dimension and indicator, and the subjective weight of each indicator is calculated based on this judgment matrix. The entropy weight method is used to calculate the information entropy of each evaluation index based on the preprocessed data, and the objective weight of each index is calculated based on the information entropy. The subjective weights obtained by the analytic hierarchy process (AHP) and the objective weights obtained by the entropy weight method are weighted and integrated to obtain the comprehensive weights of each evaluation index.

[0009] In an optional implementation, the multi-parameter calibration performance of the vehicle is quantitatively evaluated using a fuzzy comprehensive evaluation method, incorporating the comprehensive weights, to obtain a comprehensive evaluation result, including: The multi-dimensional evaluation index system is initialized, the evaluation level domain with {excellent, good, average, poor} as elements is determined, and based on expert scoring or historical calibration data, trapezoidal or triangular membership functions for each evaluation index in the multi-dimensional evaluation index system are defined to belong to different evaluation levels. Based on the preprocessed multi-source heterogeneous operation data and the multi-dimensional evaluation index system, the specific values ​​of each evaluation index are calculated; the specific values ​​are input into their corresponding membership functions to calculate the membership degree of each index to different evaluation levels, thereby constructing a fuzzy membership matrix. The comprehensive weight vector and the fuzzy membership matrix are subjected to fuzzy synthesis operation, and a weighted average operator is used to obtain a fuzzy comprehensive evaluation result vector that represents the distribution of the overall vehicle performance at each evaluation level. The centroid method is used to defuzzify the fuzzy comprehensive evaluation result vector, and a quantified comprehensive evaluation score is output. The comprehensive evaluation score is used as a supervision label and associated with the preprocessed multi-parameter data and comprehensive weights on which the score is generated to form a training sample set. The training sample set is then learned using a decision tree regression algorithm to establish a data-driven relationship that can directly predict the comprehensive evaluation score from the multi-parameter data, thereby accelerating or optimizing the fuzzy comprehensive evaluation process.

[0010] In an optional implementation, the training sample set is learned using a decision tree regression algorithm to establish a data-driven relationship that can directly predict the comprehensive evaluation score from multi-parameter data, including: The preprocessed multi-parameter data in the training sample set is used as input features, and the corresponding comprehensive evaluation score is used as output label to construct supervised learning samples. The sample is trained by a decision tree regression algorithm to learn the direct mapping relationship from multi-parameter data to evaluation scores, and a regression prediction model that can replace or assist the fuzzy comprehensive evaluation process is generated. The trained regression prediction model is applied to new vehicle operation data to directly output the predicted evaluation score, thereby accelerating and optimizing the original multi-step fuzzy evaluation method.

[0011] Secondly, the present invention provides a calibration parameter evaluation system for new energy vehicles, comprising: The data acquisition module is used to collect and preprocess multi-source heterogeneous operating data of new energy vehicles; The system construction module is used to build a multi-dimensional evaluation index system covering safety, economy, power, comfort and reliability based on preprocessed data; The weight determination module is used to determine the comprehensive weight of each indicator in the evaluation index system by adopting a subjective and objective combined weighting method. The result generation module is used to combine the comprehensive weights and use the fuzzy comprehensive evaluation method to quantitatively evaluate the multi-parameter calibration performance of the vehicle to obtain a comprehensive evaluation result.

[0012] Thirdly, a device is provided, comprising: The memory is used to store the calibration parameter evaluation program for new energy vehicles; The processor is configured to implement the steps of the calibration parameter evaluation method for new energy vehicles as provided in the first aspect when executing the calibration parameter evaluation program for the new energy vehicle.

[0013] Fourthly, a computer-readable storage medium is provided, on which a calibration parameter evaluation program for a new energy vehicle is stored. When the calibration parameter evaluation program for a new energy vehicle is executed by a processor, it implements the steps of the calibration parameter evaluation method for a new energy vehicle provided in the first aspect.

[0014] The beneficial effects of this invention are as follows: the calibration parameter evaluation method, system, equipment, and storage medium for new energy vehicles provided by this invention effectively improve the integrity and reliability of vehicle operation data through a multi-source heterogeneous data acquisition and preprocessing mechanism, laying a solid foundation for performance evaluation. Based on a multi-dimensional evaluation index system, it systematically covers key performance dimensions such as safety and economy, overcoming the limitations of traditional methods with their single evaluation dimension. A subjective-objective combined weighting method is used to determine index weights, taking into account both expert experience and the objective characteristics of data, making the weight allocation more scientific and reasonable. Combined with the fuzzy comprehensive evaluation method, it achieves accurate quantitative evaluation of multi-parameter calibration performance, providing clear guidance for calibration optimization. Through a dynamic feedback mechanism, a closed-loop process of "evaluation-optimization-iteration" is formed, significantly shortening the development cycle and improving calibration efficiency. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram illustrating the principle of a method according to an embodiment of the present invention.

[0018] Figure 3 This is a schematic flowchart illustrating the data acquisition and preprocessing method of one embodiment of the present invention.

[0019] Figure 4 This is a schematic block diagram of a system according to an embodiment of the present invention.

[0020] Figure 5 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0023] The calibration parameter evaluation method for new energy vehicles provided in this embodiment of the invention is executed by computer equipment, and correspondingly, the calibration parameter evaluation system for new energy vehicles runs in computer equipment.

[0024] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention. Wherein, Figure 1 The implementing entity can be a calibration parameter evaluation system for new energy vehicles. Depending on different needs, the order of the steps in this flowchart can be changed, and some steps can be omitted.

[0025] like Figure 1 As shown, the method includes: S1. Collect and preprocess multi-source heterogeneous operation data of new energy vehicles; S2. Based on the preprocessed data, construct a multi-dimensional evaluation index system covering safety, economy, power, comfort and reliability; S3. Using a combination of subjective and objective weighting methods, determine the comprehensive weight of each indicator in the evaluation index system; S4. Combining the aforementioned comprehensive weights, the multi-parameter calibration performance of the vehicle is quantitatively evaluated using the fuzzy comprehensive evaluation method to obtain the comprehensive evaluation result.

[0026] The following is combined with Figure 2 The solution provided by the present invention will be described in detail below.

[0027] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0028] Please refer to Figure 3 Multi-source heterogeneous data acquisition and preprocessing: Real-time acquisition of multi-parameter data from new energy vehicles under different operating conditions via sensors, including battery parameters, motor parameters, electronic control parameters, accessory parameters, vehicle operating parameters, driver operation data, and environmental parameters. Preprocessing of the acquired data includes data cleaning, noise reduction, and normalization to remove outliers and noise, ensuring data accuracy and reliability.

[0029] Real-time vehicle operation data is collected using a distributed sensor network, including: (1) Battery system parameters: battery voltage (V), current (A), temperature (°C), SOC (%), cell temperature (°C), cell voltage (V), battery relay status; (2) Motor system parameters: motor speed (rpm), torque (N·m), power (kW), torque fluctuation (N·m / s); (3) Electrical control system parameters: input current (A), input voltage (V), motor controller temperature (°C); (4) Accessory system parameters: steering oil pump pressure (MPa), steering motor output torque (N·m), compressor power (kW), critical node temperature (°C); (5) Vehicle operating parameters: vehicle speed (km / h), acceleration (m / s²), driving range (km), energy consumption (kW·h / 100km), current vehicle weight (kg); (6) Driver operating parameters: steering wheel angle (°), steering angular velocity (° / s), hand torque (N·m), accelerator / brake pedal opening (%), accelerator / brake pedal opening change rate (% / s); (7) Environmental parameters: ambient temperature (°C), humidity (%), altitude (m), road surface adhesion coefficient (μ).

[0030] Data preprocessing: Null value handling: To address the differences in configuration across different vehicle models, a parameter configuration mapping table is established, and missing parameters are dynamically masked. For special operating conditions of the vehicle (low-voltage power-on state, high-voltage power-on state, emergency avoidance state), a state correlation filtering algorithm is used to handle null values ​​of key parameters such as battery current, vehicle speed, acceleration, and road adhesion coefficient.

[0031] Multi-cycle data synchronization: Under normal circumstances, due to the different transmission rates of various parameters, such as battery current SOC, temperature, and humidity, which do not have high real-time requirements, a period of 500ms-1s is generally used to reduce bus load. However, for vehicle power torque control data, a control cycle of 50ms-5ms is generally used. For data with different cycles, sliding window interpolation is performed. For continuous signals, such as vehicle speed, motor speed, and motor torque, linear continuous interpolation is used, and limits are applied to states where the interpolation exceeds the usable value. For discrete signals, such as fault codes and gear positions, nearest-neighbor interpolation is used.

[0032] The motor power, motor torque, and wheel-end driving force are normalized.

[0033] The data is filtered to remove discrete values, abrupt changes, and singular values.

[0034] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0035] Based on vehicle performance goals, at least one quantifiable performance indicator is configured for each of the five dimensions: safety, economy, power, comfort, and reliability. The safety dimension includes at least one of the following indicators: battery overcharge / over-discharge protection response time and power system insulation resistance value; the economy dimension includes at least one of the following indicators: energy consumption per 100 kilometers and energy recovery ratio.

[0036] The specific configuration scheme is as follows: (1) Safety Dimension: The configured indicators aim to evaluate the timeliness and effectiveness of vehicle high-voltage electrical safety and critical system protection. Specifically, these include: Battery overcharge / over-discharge protection response time: refers to the time taken from when the battery management system (BMS) detects that the cell voltage exceeds the upper limit of charging or falls below the lower limit of discharging, to when it executes protective actions such as cutting off the relay. The shorter the time, the faster the protection.

[0037] Insulation resistance of the power system: refers to the insulation resistance between the high-voltage bus and the vehicle chassis, usually expressed in Ω / V. Its value must be higher than the minimum limit specified by national standards to ensure personnel safety.

[0038] Motor overload protection sensitivity: refers to the response threshold and delay time of the motor controller triggering power reduction or shutdown protection after detecting that the output torque or current continuously exceeds the rated value.

[0039] (2) Economic dimension: The configured indicators are used to quantify the energy utilization efficiency of the vehicle. Specifically, they include: Electricity consumption per 100 kilometers: The amount of electricity consumed by a vehicle when it travels 100 kilometers under specific operating conditions (such as the China Light Vehicle Driving Condition CLTC). The unit is kW·h / 100km, which is a direct indicator of energy consumption.

[0040] Energy recovery ratio: During braking or coasting, the percentage of energy recovered by the motor and fed back to the battery is the total energy consumed during the deceleration process. The calculation formula can be (recovered energy / (recovered energy + energy consumed by mechanical braking)) × 100%.

[0041] Thermal management energy consumption: The proportion of energy consumed by the air conditioning system, liquid cooling system, etc., in the total energy consumption of the vehicle in order to maintain the battery, motor and other components in the optimal temperature range.

[0042] (3) Performance dimension: The configuration indicators are used to evaluate the vehicle's driving capability. Specifically, these include conventional indicators in this field such as maximum torque, 0-50km / h acceleration time, and top speed.

[0043] (4) Comfort dimension: The configuration indicators focus on the driving and riding experience. Specifically, this includes: Three-dimensional vibration amplitude of the whole vehicle: The effective value (RMS) of longitudinal, lateral and vertical vibration acceleration of the vehicle is measured by acceleration sensors arranged in the seat rails, steering wheel and other positions.

[0044] Acceleration / deceleration impact: refers to the rate of change of acceleration over time (jerk), i.e., m / s³, used to quantify the impact felt during acceleration or braking. The smaller the value, the better the smoothness.

[0045] (5) Reliability dimension: The configured metrics are used to evaluate the stability and durability of the system. Specifically, these include: Functional failure rate: The number or frequency of functional fault codes reported by each electronic control system (such as BMS, MCU) within a specific test cycle or mileage.

[0046] Percentage of time spent in power and torque limiting state: This refers to the cumulative time during which the motor output power or torque is actively limited by the system due to reasons such as excessively high temperature or low power, out of the total running time. The higher the value, the more frequently the system performance is limited.

[0047] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0048] S301. Calculate subjective weights using the Analytic Hierarchy Process (AHP): Based on the established hierarchical model comprising a target layer (vehicle calibration performance), a criterion layer (five dimensions), and a solution layer (specific indicators), domain experts were invited to conduct pairwise comparisons of elements at each level using a 1-9 scale. For example, at the criterion layer, if an expert considers "safety" significantly more important than "economy," a score of 3 is assigned. Through systematic comparison, a judgment matrix A is constructed:

[0049] Subsequently, the weight vector is calculated using the square root method. The geometric mean is calculated for each row of the judgment matrix A, and the resulting vector is then normalized to obtain the subjective weight vector for each criterion. The calculation process is as follows:

[0050] To ensure logical consistency in the judgments, a consistency check is required. The consistency ratio CR (CR = CI / RI) is calculated. When CR < 0.1, the consistency of the judgment matrix is ​​considered acceptable, and then... Formalized as subjective weight Otherwise, experts need to readjust the judgment matrix. This process is completed automatically by an algorithm to ensure the rationality of the weight allocation.

[0051] S302. Calculate objective weights using the entropy weight method. Information is extracted from the measured data itself to determine the objective weights of each indicator, thus compensating for potential biases in subjective judgment. First, the preprocessed multi-cycle, multi-condition vehicle operation data (i.e., indicator sample values) is used to construct an evaluation matrix. Given m samples and n evaluation indicators, the matrix is ​​X = (x... ij ) m×n .

[0052] The formula for calculating the information entropy of an indicator is:

[0053] in,

[0054] The objective weights of each indicator are calculated based on information entropy:

[0055] Entropy weights are calculated using a combination of subjective and objective weights.

[0056] For subjective entropy weight, For objective entropy weight, and The entropy weights are the coefficients. and The value can be adjusted according to the specific focus of the project, and .

[0057] In one embodiment of the present invention, based on step S4, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0058] S401. Initialize the multi-dimensional evaluation index system, determine the evaluation level domain with {excellent, good, average, poor} as elements, and define trapezoidal or triangular membership functions for each evaluation index in the multi-dimensional evaluation index system to belong to different evaluation levels based on expert scoring or historical calibration data. First, the evaluation level domain is defined as four levels, namely V = {Excellent, Good, Average, Poor}, thus forming the fuzzy evaluation level set. Next, for each underlying quantifiable indicator in the multi-dimensional evaluation index system (such as energy consumption per 100 kilometers, 0-50 km / h acceleration time, etc.), a membership function is defined to belong to the above four evaluation levels. This embodiment preferably uses trapezoidal or triangular membership functions, which are computationally simple and conform to human judgment patterns. The parameters of the function are determined in two ways: first, by inviting domain experts to directly define the critical values ​​between different levels for each indicator based on historical experience and performance targets; second, by automatically determining the threshold range of each level based on a large amount of historical calibration data through statistical methods such as cluster analysis. For example, for the "energy consumption per 100 kilometers" indicator, below 15 kWh can be set as "Excellent", 15 to 18 kWh as "Good", 18 to 22 kWh as "Average", and above 22 kWh as "Poor", and a triangular membership function is used to describe its fuzzy transition interval.

[0059] S402. Based on the preprocessed multi-source heterogeneous operation data and the multi-dimensional evaluation index system, calculate the specific values ​​of each evaluation index; input the specific values ​​into their corresponding membership functions to calculate the membership degree of each index to different evaluation levels, thereby constructing a fuzzy membership matrix.

[0060] Based on real-time collected and preprocessed multi-source heterogeneous operational data, the specific values ​​of each evaluation index under the current state are calculated. The calculation method is not a single approach, but rather, based on the definition and characteristics of the indicators, it is mainly divided into the following three categories: 1. Direct reading or simple conversion The values ​​of some evaluation indicators can be obtained directly from the corresponding signal messages on the vehicle network (such as the CAN bus), or only require simple unit conversion or ratio calculation.

[0061] Examples: battery temperature, motor speed, vehicle speed Calculation method: These are physical quantities directly measured by basic sensors. After preprocessing steps such as null value handling, multi-cycle data synchronization and filtering and elimination mentioned in the document, their values ​​can be directly used as input values ​​for the corresponding evaluation indicators.

[0062] II. Calculation based on models and formulas More key performance indicators need to be calculated by establishing mathematical models or physical formulas, taking into account one or more directly measured parameters.

[0063] Example: Energy consumption per 100 kilometers Calculation method: This is a typical cumulative and average calculation indicator. It is not an instantaneous value, but is calculated over a certain period of time or under specific operating conditions. The formula is usually: Energy consumption per 100 km (kW·h / 100km) = (Total energy consumption over a certain period of time (kW·h) / Distance traveled over that period of time (km)) * 100. Wherein, the "total energy consumption" needs to be calculated by integration, that is, by integrating the instantaneous power (voltage × current) of the high-voltage battery over time.

[0064] Example: Acceleration / deceleration impact Calculation method: This is a kinetic index, defined as the rate of change of acceleration (Jerk). The calculation formula is: Impact force (m / s²) 3 ) = d(a) / dt, where a is the acceleration value measured by the longitudinal acceleration sensor. In actual calculations, it is usually approximated by taking the first-order difference of the continuous acceleration signal (i.e., calculating the difference in acceleration between two consecutive sampling points and dividing by the sampling time interval).

[0065] III. Based on event statistics and logical judgment Some metrics, especially reliability and security metrics, are calculated based on statistical and logical judgments about the frequency of occurrence of specific events or the duration of system states.

[0066] Example: Battery overcharge / over-discharge protection response time Calculation Method: This metric requires monitoring relevant events. The system continuously monitors the voltage of individual battery cells. When any cell voltage exceeds the charging protection threshold (triggers an "overcharge event") or falls below the discharging protection threshold (triggers an "over-discharge event"), the precise timestamp t_start of the event's start is recorded. Subsequently, when the Battery Management System (BMS) issues a command to disconnect the charging and discharging relays, the precise timestamp t_end of the action's completion is recorded. The response time of this protection item is then t_end - t_start.

[0067] Example: Percentage of runtime under power and torque limiting conditions Calculation method: The system continuously monitors the status word of the motor controller to determine whether it is in a power or torque limiting state. Over the entire operating time T_total, the sum of all time periods in the limiting state, T_derate, is accumulated. The value of this indicator is then: Percentage = (T_derate / T_total) * 100%.

[0068] The specific value of each indicator is input into its membership function defined in step S401, and the membership degree of that value to the four levels of "Excellent," "Good," "Medium," and "Poor" is calculated respectively. For example, if the calculated energy consumption per 100 kilometers is 16.5 kWh, according to its membership function, its membership degree to "Excellent" might be 0.2, its membership degree to "Good" might be 0.8, and its membership degree to "Medium" and "Poor" might both be 0. This process is repeated for all indicators, and finally the membership degree results of all indicators are combined to form a fuzzy relation matrix with multiple rows (corresponding to each indicator) and four columns (corresponding to each evaluation level), namely the fuzzy membership degree matrix R.

[0069] S403. Perform a fuzzy synthesis operation on the comprehensive weight vector and the fuzzy membership matrix, and use a weighted average operator to obtain a fuzzy comprehensive evaluation result vector that characterizes the distribution of the overall vehicle performance at each evaluation level.

[0070] The comprehensive weight vector W (row vector) of each index obtained in step S303 is combined with the fuzzy membership matrix R constructed in step S402 to perform fuzzy synthesis. This embodiment uses an operator suitable for the weighted average concept (such as the M(•,+) operator) for calculation. The calculation formula can be expressed as follows: ,in" " is the composition operator. The operation results in a comprehensive evaluation result vector B=(b1,b2,b3,b4), which represents the degree to which the overall performance of the vehicle belongs to the four levels of "excellent, good, medium and poor".

[0071] S404. The centroid method is used to defuzzify the fuzzy comprehensive evaluation result vector, and a quantified comprehensive evaluation score is output.

[0072] To obtain an intuitive and comparable quantitative score, the fuzzy comprehensive evaluation result vector B needs to be defuzzified. This embodiment uses the centroid method. Assuming the four evaluation levels "Excellent, Good, Average, Poor" are assigned values ​​of 95, 80, 65, and 50 respectively (which can be adjusted according to actual conditions), the final comprehensive evaluation score S can be calculated using the following formula: S = (b1×95 + b2×80 + b3×65 + b4×50) / (b1 + b2 + b3 + b4). This outputs a scalar score between 0 and 100, intuitively reflecting the vehicle's overall performance level under the current calibration parameters.

[0073] S405. The comprehensive evaluation score is used as a supervision label and associated with the preprocessed multi-parameter data and comprehensive weights on which the score is generated to form a training sample set. The training sample set is then learned using a decision tree regression algorithm to establish a data-driven relationship that can directly predict the comprehensive evaluation score from the multi-parameter data, thereby accelerating or optimizing the fuzzy comprehensive evaluation process.

[0074] To improve evaluation efficiency and achieve model self-optimization, this invention introduces a supervised learning mechanism. Specifically, historical data generated during multiple evaluation processes are used as training samples: the preprocessed multi-parameter data (input features) used for each evaluation are associated with the comprehensive evaluation score S (output label or supervision label) calculated in step S404, and the comprehensive weight W used in this evaluation is recorded, forming a training sample together.

[0075] After accumulating sufficient samples, a training sample set is constructed. This sample set is then trained using a decision tree regression algorithm. The decision tree regression model automatically learns the nonlinear mapping relationship between complex multi-parameter input features and a single comprehensive evaluation score. After training, a regression prediction model is generated. In practical applications, for new vehicle operation data, this trained decision tree regression model can be directly input, and the model can quickly predict an evaluation score. Compared to the complete fuzzy evaluation process, this method eliminates steps such as constructing membership matrices and fuzzy synthesis operations, significantly improving evaluation speed. It is particularly suitable for online evaluations requiring rapid response or large-scale parameter optimization scenarios, thus effectively accelerating and optimizing the original method. This data-driven model can also run in parallel with the fuzzy evaluation method for mutual verification, ensuring the reliability of the evaluation results.

[0076] Based on the above embodiments, in one embodiment, the comprehensive evaluation results can be analyzed in depth to identify problems and deficiencies in the multi-parameter calibration of new energy vehicles. According to the analysis results, targeted suggestions and solutions are provided for parameter optimization, and the optimized parameters are fed back to the vehicle's control system to achieve dynamic adjustment and optimization of multiple parameters of the new energy vehicle, thereby improving the overall performance of the vehicle.

[0077] A reinforcement learning-based model is used to traverse simulation adjustments to optimize the overall direction of each variable. Different variable calibration strategies are formulated, and further experimental tests are conducted on different calibration strategies. The test results are then evaluated and iterated.

[0078] For special indicators such as economy and safety, expert experience is used to make adjustments, an expert adjustment model is established, and the beneficial adjustment points in the expert adjustment are used as the boundary function and optimization guidance direction of the reinforcement learning model to improve optimization efficiency.

[0079] For example, safety dimensions include indicators such as energy consumption per 100 kilometers and charging efficiency; power dimensions include indicators such as maximum torque, top speed, and acceleration time; comfort dimensions include indicators such as in-vehicle noise, seat comfort, and air conditioning system performance; and reliability dimensions include indicators such as battery cycle life, motor failure rate, and system mean time between failures.

[0080] This invention establishes an architecture for performance evaluation during the calibration process of new energy vehicles, and clearly proposes a comprehensive evaluation method for various systems and performance indicators in the calibration process. It realizes the entire chain of work, including vehicle information collection and filtering, information aggregation, multi-dimensional analysis, feedback evaluation, and iterative upgrading. The evaluation system provided by this invention can achieve qualitative and quantitative evaluation of calibration parameters, which is beneficial to improving performance calibration efficiency and shortening the development cycle.

[0081] In some embodiments, the calibration parameter evaluation system for new energy vehicles may include multiple functional modules composed of computer program segments. The computer programs for each program segment in the calibration parameter evaluation system for new energy vehicles may be stored in the memory of a computer device and executed by at least one processor to perform (see details). Figure 1 (Description) Function for evaluating calibration parameters of new energy vehicles.

[0082] In this embodiment, the calibration parameter evaluation system for new energy vehicles can be divided into multiple functional modules based on the functions it performs, such as... Figure 4 As shown. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and is stored in memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0083] The data acquisition module is used to collect and preprocess multi-source heterogeneous operating data of new energy vehicles; The system construction module is used to build a multi-dimensional evaluation index system covering safety, economy, power, comfort and reliability based on preprocessed data; The weight determination module is used to determine the comprehensive weight of each indicator in the evaluation index system by adopting a subjective and objective combined weighting method. The result generation module is used to combine the comprehensive weights and use the fuzzy comprehensive evaluation method to quantitatively evaluate the multi-parameter calibration performance of the vehicle to obtain a comprehensive evaluation result.

[0084] Figure 5The calibration parameter evaluation method for new energy vehicles provided in this application embodiment can be applied to equipment. Those skilled in the art will understand that the equipment structure involved in the embodiments of this invention does not constitute a limitation on the equipment. The equipment may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the equipment includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The equipment may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0085] The device 500 may include a processor 510, a memory 520, and a communication unit 530. These components communicate via one or more buses. Those skilled in the art will understand that the server structure shown in the figure does not constitute a limitation of the present invention. It may be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0086] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, which, when executed, may include some or all of the steps provided in the embodiments of the present invention. The storage medium may be a magnetic disk, an optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0087] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other medium capable of storing program code. It includes several instructions to cause a computer device (which may be a personal computer, a server, or a second device, network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0088] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

[0089] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or modules may be electrical, mechanical, or other forms.

[0090] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0091] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0092] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. A method for evaluating calibration parameters of new energy vehicles, characterized in that, include: Collect and preprocess multi-source heterogeneous operation data of new energy vehicles; Based on the preprocessed data, a multi-dimensional evaluation index system covering safety, economy, power, comfort and reliability is constructed. The comprehensive weight of each indicator in the evaluation index system is determined by using a combination of subjective and objective weighting methods. Based on the aforementioned comprehensive weights, the multi-parameter calibration performance of the vehicle is quantitatively evaluated using the fuzzy comprehensive evaluation method to obtain the comprehensive evaluation result.

2. The method according to claim 1, characterized in that, Collect and preprocess multi-source heterogeneous operational data of new energy vehicles, including: Data is collected through a distributed sensor network and includes the following: battery system parameters, including voltage, current, temperature, and SOC; motor system parameters, including speed, torque, and power; electronic control system parameters, including input current, voltage, and controller temperature; and vehicle operating parameters, including vehicle speed, acceleration, and energy consumption. Null value processing, multi-period data synchronization and normalization are performed on the collected multi-source heterogeneous operation data; The null value processing includes dynamic masking based on parameter configuration mapping table and correlation filtering based on vehicle operating status; the multi-cycle data synchronization includes using sliding window interpolation to perform linear interpolation on continuous signals and neighbor interpolation on discrete signals.

3. The method according to claim 1, characterized in that, Based on the preprocessed data, a multi-dimensional evaluation index system covering safety, economy, power, comfort, and reliability is constructed, including: Based on vehicle performance objectives, at least one quantifiable performance indicator is configured for each of the five dimensions: safety, economy, power, comfort, and reliability.

4. The method according to claim 3, characterized in that, The quantifiable performance metrics configured for each dimension include: The safety dimension is configured with indicators including at least one of the following: battery overcharge / overdischarge protection response time, power system insulation resistance, and motor overload protection sensitivity. The indicators configured in the economic dimension include at least one of the following: energy consumption per 100 kilometers, energy recovery ratio, and thermal management energy consumption. The performance metrics include at least one of the following: maximum torque, 0-50km / h acceleration time, and top speed. The indicators configured for the comfort dimension include at least one of the following: three-dimensional vibration amplitude of the whole vehicle and acceleration / deceleration impact. The reliability dimension is configured with at least one of the following metrics: functional failure rate and percentage of time spent in power / torque limiting state.

5. The method according to claim 1, characterized in that, The comprehensive weight of each indicator in the evaluation index system is determined by a combination of subjective and objective weighting methods, including: Using the analytic hierarchy process, a judgment matrix is ​​constructed to reflect the relative importance of each dimension and indicator, and the subjective weight of each indicator is calculated based on this judgment matrix. The entropy weight method is used to calculate the information entropy of each evaluation index based on the preprocessed data, and the objective weight of each index is calculated based on the information entropy. The subjective weights obtained by the analytic hierarchy process (AHP) and the objective weights obtained by the entropy weight method are weighted and integrated to obtain the comprehensive weights of each evaluation index.

6. The method according to claim 1, characterized in that, Combining the aforementioned comprehensive weights, the multi-parameter calibration performance of the vehicle is quantitatively evaluated using the fuzzy comprehensive evaluation method to obtain a comprehensive evaluation result, including: The multi-dimensional evaluation index system is initialized, the evaluation level domain with {excellent, good, average, poor} as elements is determined, and based on expert scoring or historical calibration data, trapezoidal or triangular membership functions for each evaluation index in the multi-dimensional evaluation index system are defined to belong to different evaluation levels. Based on the preprocessed multi-source heterogeneous operation data and the multi-dimensional evaluation index system, the specific values ​​of each evaluation index are calculated; the specific values ​​are input into their corresponding membership functions to calculate the membership degree of each index to different evaluation levels, thereby constructing a fuzzy membership matrix. The comprehensive weight vector and the fuzzy membership matrix are subjected to fuzzy synthesis operation, and a weighted average operator is used to obtain a fuzzy comprehensive evaluation result vector that represents the distribution of the overall vehicle performance at each evaluation level. The centroid method is used to defuzzify the fuzzy comprehensive evaluation result vector, and a quantified comprehensive evaluation score is output. The comprehensive evaluation score is used as a supervision label and associated with the preprocessed multi-parameter data and comprehensive weights on which the score is generated to form a training sample set. The training sample set is then learned using a decision tree regression algorithm to establish a data-driven relationship that can directly predict the comprehensive evaluation score from the multi-parameter data, thereby accelerating or optimizing the fuzzy comprehensive evaluation process.

7. The method according to claim 6, characterized in that, The training sample set is learned using a decision tree regression algorithm to establish a data-driven relationship that can directly predict the comprehensive evaluation score from multi-parameter data, including: The preprocessed multi-parameter data in the training sample set is used as input features, and the corresponding comprehensive evaluation score is used as output label to construct supervised learning samples. The sample is trained by a decision tree regression algorithm to learn the direct mapping relationship from multi-parameter data to evaluation scores, and a regression prediction model that can replace or assist the fuzzy comprehensive evaluation process is generated. The trained regression prediction model is applied to new vehicle operation data to directly output the predicted evaluation score, thereby accelerating and optimizing the original multi-step fuzzy evaluation method.

8. A calibration parameter evaluation system for new energy vehicles, characterized in that, include: The data acquisition module is used to collect and preprocess multi-source heterogeneous operating data of new energy vehicles; The system construction module is used to build a multi-dimensional evaluation index system covering safety, economy, power, comfort and reliability based on preprocessed data; The weight determination module is used to determine the comprehensive weight of each indicator in the evaluation index system by adopting a subjective and objective combined weighting method. The result generation module is used to combine the comprehensive weights and use the fuzzy comprehensive evaluation method to quantitatively evaluate the multi-parameter calibration performance of the vehicle to obtain a comprehensive evaluation result.

9. A calibration parameter evaluation device for new energy vehicles, characterized in that, include: The memory is used to store the calibration parameter evaluation program for new energy vehicles; The processor is configured to implement the steps of the calibration parameter evaluation method for new energy vehicles as described in any one of claims 1-7 when executing the calibration parameter evaluation program for the new energy vehicle.

10. A computer-readable storage medium storing a computer program, characterized in that, The readable storage medium stores a calibration parameter evaluation program for new energy vehicles. When the calibration parameter evaluation program for new energy vehicles is executed by the processor, it implements the steps of the calibration parameter evaluation method for new energy vehicles as described in any one of claims 1-7.

Images