Method and system for testing direct current charging compatibility of electric vehicle

Abstract

The invention discloses an electric vehicle direct current charging compatibility test method and system. The method comprises the steps of performing charging test on a test vehicle to obtain vehicle charging data; generating characteristic parameters based on the charging data, and loading the characteristic parameters to the virtual charging pile; and connecting the virtual charging pile and the to-be-tested vehicle to carry out charging compatibility testing. According to the invention, the charging test is carried out on the test vehicle to obtain the vehicle charging data and generate the characteristic parameters, and the communication characteristics of different charging piles are simulated by using the virtual charging pile based on the characteristic parameters, so that the problem of insufficient software communication compatibility test in the prior art is effectively solved; the method has the advantages of improving compatibility test coverage and effectiveness and reducing research and development test cost.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle charging technology, specifically relating to a method and system for testing the DC charging compatibility of electric vehicles. Background Technology

[0002] With the rapid development of my country's electric vehicle industry, users' demand for vehicle charging is becoming increasingly urgent, and charging infrastructure is also developing rapidly to meet this demand. Although the construction of charging infrastructure is constantly improving, the market is flooded with charging piles using various solutions, resulting in inconsistent quality and compatibility with electric vehicles. This leads to numerous user complaints due to charging failures, highlighting the urgent need to strengthen the compatibility between electric vehicles and DC charging piles.

[0003] DC charging for electric vehicles is achieved through the joint efforts of a DC charging station, the electric vehicle's BMS (Battery Management System), and the power battery. After the charging gun is inserted into the electric vehicle's charging port, the control guidance circuit and CAN communication circuit are activated. The charging station then sends handshake and identification signals to the vehicle via the CAN bus. Once the handshake is successful, the charging station initiates power transfer and begins charging. During the charging process, the charging station and the electric vehicle must send messages to each other at standard intervals, and power transfer will retry or stop if either party experiences an anomaly.

[0004] Currently, DC compatibility testing for electric vehicles generally focuses on physical and electrical compatibility, neglecting software communication compatibility testing. However, modern electric vehicles and charging stations possess comprehensive hardware design processes, control strategies, and fault detection and recovery mechanisms. Therefore, problems arising from improper physical operation, hardware malfunctions, or damage are gradually decreasing. In actual development and testing, most compatibility issues involving unexpected charging stoppages or charging failures are software communication faults, which are difficult to detect in advance through physical and electronic / electrical testing during the R&D process.

[0005] Since DC charging involves communication protocol interaction between vehicles and charging piles, and the specific implementation methods of the current communication protocols are diverse, it is difficult to conduct targeted testing in the early stage of R&D. Furthermore, the compatibility testing in the later stage of R&D is time-consuming and costly, which is not conducive to improving the efficiency and reducing the cost of R&D.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of the aforementioned background technology by providing a method and system for DC charging compatibility testing of electric vehicles, which has the advantages of improving the coverage and effectiveness of compatibility testing and reducing R&D testing costs.

[0008] The technical solution adopted in this invention is: a method for testing the DC charging compatibility of electric vehicles, comprising the following steps: Conduct charging tests on the test vehicles to obtain vehicle charging data; Feature parameters are generated based on charging data, and these feature parameters are then loaded into the virtual charging station. Connect the virtual charging station and the vehicle under test to conduct charging compatibility testing.

[0009] Furthermore, the step of conducting charging tests on the test vehicle to obtain vehicle charging data includes: Several electric vehicles were selected as test vehicles. For each test vehicle, the power consumption and charging operations were repeated multiple times, and the charging data for each charging process was recorded. Furthermore, the power consumption and charging operation are as follows: The test vehicle's battery is depleted to SOC0. The test vehicle is then connected to a DC charging station, and the test vehicle is charged from SOC0 to 100%, where SOC0 is less than or equal to a set low battery value.

[0010] Furthermore, the charging data includes charging pile terminal message parameters, vehicle terminal message parameters, and vehicle terminal measurement parameters.

[0011] Furthermore, the generation of feature parameters based on charging data includes: Generate parameter accuracy features: For each measurement parameter, calculate the percentage deviation of the measurement parameter in each cycle when the charging pile output reaches steady state, and cluster several deviation percentages to obtain the maximum value, minimum value and characteristic value, thus obtaining a set of parameter accuracy features.

[0012] Furthermore, the generation of feature parameters based on charging data includes: Generate periodic features: For each periodically sent message at the pile end, measure the time interval between two transmissions of the same message, cluster several time intervals to obtain the maximum value, minimum value and feature value, and obtain a set of periodic features.

[0013] Furthermore, the generation of feature parameters based on charging data includes: Generate message response characteristics: For each type of message, measure the response time between the time the pile end receives the message and the time the vehicle end sends the message, and cluster several response times to obtain the maximum value, minimum value and characteristic value, thus obtaining a set of message response characteristics.

[0014] Furthermore, the generation of feature parameters based on charging data includes generating normalized features: (1) Response normative characteristics: For each request parameter, calculate the normative characteristics of the request parameter in each cycle when the charging pile output reaches a steady state, and cluster several normative characteristics to obtain the maximum value, minimum value and characteristic value, thus obtaining a set of response normative characteristics; the normative characteristics = (vehicle-end request value - charging pile-end output value) / (charging pile-end output value × battery voltage); the request parameters include voltage and current; (2) Protocol normative characteristics: For multiple stages of the interaction protocol, the types and frame proportions of non-standard messages sent at each stage are statistically analyzed to form protocol normative characteristics. Non-standard messages are non-fixed messages in the corresponding stage. (3) Multi-frame message standardization characteristics: For each type of long message, the time taken for successful transmission of the long message at the pile end, the interruption information of the transmission interruption, and the retransmission information of successful retransmission after the transmission interruption are statistically analyzed to form multi-frame message standardization characteristics. Long messages are messages that exceed 8 bytes and need to be divided into multiple frames for transmission.

[0015] Furthermore, the time taken for a successful transmission includes the time taken from the moment the first frame is sent to the moment the last frame is sent. The interruption information of the transmission interruption includes the message ID of each transmission and the sequence number of the interruption frame. The retransmission information for successful retransmission after transmission interruption includes the message ID of each transmission, the sequence number of the repeated transmission frame, and the number of times it was retransmitted.

[0016] An electric vehicle DC charging compatibility testing system includes: The test vehicle is used to connect to a DC charging station for charging and send charging data to the feature module. DC charging pile is used to connect test vehicles for charging tests and send charging data to the feature module. The feature module is used to generate feature parameters based on charging data and send them to the virtual charging station; The vehicle under test is used to connect to a virtual charging station for charging. Virtual charging stations are used to load characteristic parameters and connect to the vehicle under test for charging compatibility testing.

[0017] The beneficial effects of this invention are as follows: This invention acquires vehicle charging data and generates characteristic parameters by conducting charging tests on test vehicles. Based on these characteristic parameters, it uses virtual charging piles to simulate the communication characteristics of different charging piles, enabling systematic analysis of message parameters and measurement parameters during the charging process. This effectively solves the problem of insufficient software communication compatibility testing in existing technologies, and has the advantages of improving the coverage and effectiveness of compatibility testing and reducing R&D testing costs. Attached Figure Description

[0018] Figure 1 This is a flowchart of the testing method of the present invention.

[0019] Figure 2 This is a schematic diagram of the test system of the present invention. Detailed Implementation

[0020] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0021] In existing technologies, DC charging compatibility testing for electric vehicles mainly relies on physical connection detection and electrical performance verification, with testing methods limited to hardware-level compatibility assessment. As charging pile communication protocols become more diverse, traditional testing methods struggle to cover the differences in message interaction and parameter response among different manufacturers. Communication protocol compatibility issues that arise during R&D are often only discovered in later real-vehicle testing phases, leading to extended testing cycles and increased development costs.

[0022] To address these issues, researchers noticed a lack of targeted detection methods for communication protocol differences in existing testing systems. Analysis revealed significant variations among different charging piles in message transmission cycles, parameter response speeds, and protocol interaction logic, directly impacting the stability of the charging process. Therefore, constructing a testing environment capable of simulating the characteristics of multiple communication protocols became a key breakthrough. By studying the interaction data characteristics in real-world charging scenarios, a proposal was made to transform the communication behavior of actual charging piles into configurable parameters, thereby enabling the construction of a virtualized testing environment.

[0023] Therefore, this invention provides a method for testing the DC charging compatibility of electric vehicles, such as... Figure 1 As shown, it includes the following steps: Conduct charging tests on the test vehicles to obtain vehicle charging data; Feature parameters are generated based on charging data, and these feature parameters are then loaded into the virtual charging station. Connect the virtual charging station and the vehicle under test to conduct charging compatibility testing.

[0024] Vehicle charging data refers to the set of interactive information collected during the actual charging process. Specifically, this can be achieved by recording the timing of message transmissions, parameter values, and measurement differences between the charging pile and the vehicle, reflecting the communication characteristics in real-world charging scenarios. Feature parameters are quantitative indicators of communication behavior extracted from the raw data. These can be achieved by using clustering algorithms to statistically analyze data such as message cycles, response times, and parameter deviations, characterizing the differences in communication protocol implementations among different charging piles. Virtual charging piles are testing devices with configurable communication protocol features. Specifically, they can be implemented by dynamically loading feature parameters and simulating message interaction behavior through an embedded system, used to replace physical charging piles for compatibility testing.

[0025] Specifically, the process begins by acquiring interaction data between the vehicle and the charging station during actual charging, including message transmission and reception timestamps, parameter requests and output values, and electrical measurement values. Then, feature extraction is performed on the raw data, such as calculating metrics like message transmission cycle stability and parameter response deviation rate, and cluster analysis is used to determine the range of typical feature values. These feature parameters are configured into the virtual charging station to simulate the communication behavior patterns of a specific charging station. Finally, the vehicle under test is connected to the virtual charging station, and the vehicle's compatibility with different communication protocol implementations is evaluated by observing whether communication interruptions, parameter deviations, or protocol interaction anomalies occur during charging.

[0026] This method abstracts communication characteristics into configurable parameters using virtualization technology, enabling the simulation of multiple communication protocol scenarios on a single test platform, significantly improving test coverage and efficiency. This invention achieves systematic detection of compatibility issues caused by differences in communication protocols. Through dynamic parameter configuration of virtual charging piles, it can identify vehicle compatibility issues with different protocol implementations in advance, shortening the R&D testing cycle and reducing real-vehicle verification costs.

[0027] In one embodiment, the present invention further proposes to obtain vehicle charging data by conducting charging tests on test vehicles, including selecting several electric vehicles as test vehicles, repeatedly performing power consumption and charging operations on each test vehicle, and recording the charging data during each charging process.

[0028] Among these, "several electric vehicles" refers to at least two different brands or models of vehicles. Specifically, mainstream electric vehicle models on the market can be used as test samples, for example, models with a range difference exceeding 100 kilometers. This feature is used to eliminate the bias in communication protocol compatibility assessments based on test data from a single model. Repeatedly performing power consumption and charging operations means that each test vehicle must complete at least three full charge-discharge cycles, preferably 3-5 cycles. Specifically, this can be achieved through an automated control system that discharges the vehicle from a fully charged state to a set low charge level and then recharges it to a fully charged state. This feature captures occasional message transmission anomalies by forcibly triggering communication interactions at different charging stages.

[0029] Specifically, during the vehicle selection phase, priority is given to models equipped with different battery management system versions, such as three vehicles each with battery capacities of 50kWh and 80kWh; alternatively, different types of vehicles can be selected, such as BEVs, PHEVs, and REVs. During charging and discharging operations, each discharge cycle must deplete the vehicle's battery from 100% to a set low battery level. The set low battery level can differ for different vehicle types. For example, for BEVs, the set low battery level is 0%, while for PHEVs and REVs, it is 20%. During charging, a CAN bus recorder synchronously collects all interaction messages between the charging station and the vehicle. For example, during the charging start-up phase, the timestamp and data content of the handshake message are recorded, and during the charging maintenance phase, voltage and current measurements are collected every 10 milliseconds.

[0030] This invention, through multi-vehicle combination testing, can identify the compatibility defects of charging piles with specific vehicle model communication protocols. During the testing process, the differentiated communication characteristics of different vehicle models are fully recorded. For example, a certain vehicle model sends a charging request message with a special identifier when the battery is low; this message was only detected as an abnormal response three times in repeated tests. Simultaneously, multiple charging and discharging operations can trigger communication anomalies in the charging pile during continuous operation. For instance, in one test, the charging pile exhibited a message verification error during the fifth charging attempt, a problem that was not detected in a single test.

[0031] In one embodiment, the present invention further proposes that the power consumption and charging operation be as follows: the power of the test vehicle is consumed to SOC0, the test vehicle and the DC charging pile are connected, and the test vehicle is charged from SOC0 to 100%, where SOC0 is less than or equal to a set low power value.

[0032] Here, SOC0 refers to the battery's state of charge after the vehicle's battery power has been depleted. Setting a low charge value is a benchmark parameter used to define the critical charging state. Specifically, it can be achieved by using the minimum allowable charge value corresponding to the extreme operating conditions specified in industry standards, such as setting the low charge value to 5% of the battery's rated capacity. Charging the test vehicle from SOC0 to 100% means performing a complete charging cycle, covering the entire operating range of the battery management system from the low charge protection state to the full charge state.

[0033] Specifically, the forced discharge of the test vehicle to SOC0 simulates the extreme condition of deep battery discharge in actual user use. At this point, the battery management system activates the low-charge protection mechanism, triggering a specific communication protocol interaction mode. After connecting to a DC charging station, the charging station needs to accurately identify this critical state and start the charging process. During this process, the timing of message interaction and parameter response accuracy between the vehicle and the charging station will be rigorously tested. By limiting SOC0 to no more than a set low charge value, all test vehicles of the same type are ensured to be under uniform baseline conditions, eliminating evaluation bias caused by differences in the initial test state. Executing a complete charging process from SOC0 to 100% can comprehensively detect the dynamic response characteristics of the battery management system to the charging station output parameters in different state of charge ranges, especially compatibility issues such as changes in message interaction frequency and abnormal parameter requests in the low-charge range.

[0034] This invention, by constructing standardized extreme test scenarios, achieves for the first time a quantitative assessment of the communication protocol interaction capabilities of charging piles under deep battery discharge conditions. It effectively detects the accuracy of message parsing by charging piles in low-battery protection mode and exposes anomalies in parameter requests by the vehicle's battery management system under critical conditions. This solution can verify the charging pile's response speed to low-battery vehicles and ensure the reliable execution of the charging protocol under extreme conditions. Through standardized testing procedures, the compatibility performance of vehicles and charging piles throughout a complete charging cycle can be systematically evaluated, particularly the differences in protocol interaction between low-battery and high-battery ranges.

[0035] In one embodiment, the present invention further proposes that the charging data includes charging pile terminal message parameters, vehicle terminal message parameters, and vehicle terminal measurement parameters. The charging pile terminal message parameters include message type and corresponding ID, message reception time, message transmission time, and specific message value, etc. The vehicle terminal message parameters include message type and corresponding ID, message reception time, message transmission time, etc. The vehicle terminal measurement parameters include voltage, current, power, etc.

[0036] The message type and corresponding ID in the charging pile terminal message parameters refer to the type of communication protocol message sent or received by the charging pile and its unique identifier. This can be implemented using a CAN bus data packet parsing module, which obtains the message type and ID by parsing the data frame header. The message reception time and message transmission time refer to the time node when the charging pile processes the message. This can be implemented using a high-precision clock chip in conjunction with a timestamp recording module to mark the communication interaction sequence. The specific numerical value of the message refers to the instruction parameter value carried in the message data field, which can be extracted and stored using a data parsing algorithm. The message type and corresponding ID in the vehicle terminal message parameters refer to the type of communication protocol message sent or received by the vehicle and its identifier. This can be implemented using an on-board CAN controller in conjunction with a message filtering module, which intercepts and categorizes communication bus data in real time. The message reception time and message transmission time refer to the time node when the vehicle processes the message. This can be implemented using a synchronization clock module to ensure consistency with the charging pile terminal time reference. The voltage and current parameters measured at the vehicle end refer to the real-time electrical parameters of the power battery, which can be achieved by using Hall sensors in conjunction with analog-to-digital conversion circuits; the charge refers to the battery's state of charge, which can be achieved by integrating a coulomb counter with a battery management system module; the power refers to the energy transfer rate during the charging process, which can be calculated in real time by multiplying the voltage and current.

[0037] Specifically, during charging testing, the charging pile's message parameters record the message type, ID, and timestamp of each message sent, thus fully reconstructing the charging pile's communication behavior pattern. For example, the identification message ID sent by the charging pile during the handshake phase is 0x100, and the difference between its sending time and the vehicle's receiving time reflects communication latency. The vehicle's message parameters synchronously collect the type and timestamp of the vehicle's response messages. For example, the charging parameter configuration message ID sent by the vehicle during the charging preparation phase is 0x101, and the deviation between its sending time and the charging pile's receiving time can detect response timeliness. Vehicle-side measurement parameters monitor battery voltage and current in real time and compare them with the voltage and current commands output by the charging pile. For example, if the charging pile outputs a current command of 200A, and the measured current on the vehicle consistently deviates from 190A, then parameter matching is considered abnormal. By cross-analyzing the message timing characteristics and electrical parameter consistency, software communication defects such as charging pile message sending frequency deviation from standards, vehicle-side response timeouts, and mismatches between electrical output and communication commands can be identified.

[0038] This invention establishes a correlation analysis model between communication behavior and electrical output by synchronously acquiring the message interaction trajectory and electrical parameters between the charging pile and the vehicle, thereby accurately identifying hidden faults caused by differences in protocol implementation. Through this technical solution, the invention can solve the problem of insufficient compatibility testing between electric vehicles and charging piles due to differences in communication protocol implementation, particularly addressing testing blind spots at the software communication level such as abnormal message interaction timing and insufficient parameter matching. By simultaneously collecting communication protocol interaction data and electrical parameters, cross-verification of communication behavior and electrical output is achieved, effectively detecting defects such as abnormal charging pile message sending frequency, excessive vehicle-side response delay, and mismatch between electrical parameters and communication commands, providing multi-dimensional data support for loading test parameters for virtual charging piles.

[0039] In one embodiment, the present invention further proposes to generate feature parameters based on charging data, including generating parameter accuracy features: for each measurement parameter, calculate the percentage deviation of the measurement parameter in each cycle when the charging pile output reaches a steady state, and cluster several deviation percentages to obtain the maximum value, minimum value and feature value, thereby obtaining a set of parameter accuracy features; deviation percentage = (pile end output value - vehicle end measured value) / vehicle end measured value.

[0040] Among these features, parameter accuracy refers to quantifying the difference between the charging pile's output parameters and the vehicle's received values ​​through a percentage deviation. Specifically, K-means clustering can be used to group multi-period deviation data and extract feature values ​​representing the parameter fluctuation range to construct an evaluation system for parameter transmission accuracy. This feature reflects the stability and consistency of the charging pile's output parameters. The percentage deviation refers to the relative error between the charging pile's output value and the vehicle's measured value. This can be calculated by real-time acquisition of the charging pile's output signal and the vehicle's measurement signal to dynamically assess the accuracy of parameter transmission. This indicator can eliminate error interference caused by absolute numerical differences. Clustering refers to grouping and statistically analyzing multi-period deviation data. Specifically, density-based clustering methods can be used to identify concentrated areas of deviation distribution, extracting maximum, minimum, and typical feature values ​​to describe the overall distribution pattern of parameter deviations. This method can effectively distinguish between occasional and systematic deviations.

[0041] Specifically, after the charging pile output reaches a steady state, the output value at the pile end and the measured value at the vehicle end, such as voltage or current parameters, are acquired synchronously in each measurement cycle. The relative error between the two is calculated using a percentage deviation formula, forming time series data. Cluster analysis is performed on the percentage deviation data for multiple consecutive cycles, for example, dividing the data into three categories: high deviation, medium deviation, and low deviation. The maximum value, minimum value, and the feature value with the highest distribution density for each category are extracted. The resulting set of parameter accuracy features can reflect the fluctuation characteristics of the charging pile output parameters in the dynamic process, providing data support for the virtual charging pile to simulate the response characteristics of the real charging pile.

[0042] This invention uses cluster analysis of multi-period deviation data to identify the distribution patterns and extreme fluctuations of parameter deviations. For example, it can discover periodic high deviation phenomena that occur in charging piles under specific operating conditions, thereby more comprehensively reproducing the response characteristics of real charging piles in compatibility testing.

[0043] Through the above technical solution, this invention can systematically identify the dynamic differences between the output parameters of a charging pile and the values ​​received by the vehicle, and quantify the accuracy defects of parameter transmission through statistical modeling. For example, in voltage parameter testing, a charging pile may exhibit a periodic deviation of 5% to 8% during the steady-state output phase. After this characteristic is loaded into a virtual charging pile, it can simulate the abnormal output characteristics of the charging pile, thereby verifying the vehicle's compatibility with such deviations in advance during the R&D phase.

[0044] In one embodiment, the present invention further proposes a method for generating periodic features, which involves measuring the time interval between two transmissions of the same message for each periodically sent message at the pile end, clustering several time intervals to obtain the maximum value, minimum value and feature value, and obtaining a set of periodic features.

[0045] Periodically transmitted messages refer to communication data sent by the charging pile to the vehicle at fixed time intervals. This can be implemented using message types defined in the CAN bus protocol, such as charging status messages and voltage / current parameter messages. The time interval refers to the time difference between two consecutive transmissions of the same message. This can be achieved by recording the message transmission times and calculating the difference between adjacent times, for example, using a high-precision timer or system clock to record timestamps. Clustering refers to grouping multiple time interval data according to their similarity. This can be implemented using K-means or hierarchical clustering algorithms, extracting typical feature values ​​by analyzing the data distribution.

[0046] Specifically, during charging testing, the time intervals of the periodically sent messages by the charging pile are monitored and recorded in real time. For example, for voltage parameter messages sent by the charging pile every 100 milliseconds, their transmission timestamps are continuously collected and adjacent intervals are calculated. After collecting multiple interval data, a clustering algorithm is used to identify the maximum interval, minimum interval, and the most frequently occurring characteristic interval. The maximum value reflects the maximum possible delay of the charging pile, such as a message transmission delay of 150 milliseconds due to excessive system load; the minimum value reflects the charging pile's response capability under ideal conditions, such as a stable message interval of 95 milliseconds; and the characteristic value represents the typical interval in most cases, such as 98 milliseconds. These characteristic values ​​are loaded into a virtual charging pile to simulate charging piles with different periodic characteristics and to test the vehicle's communication compatibility under different message interval scenarios.

[0047] This invention, by extracting periodic features, can identify fluctuations or anomalies in the message transmission intervals of charging piles. For example, excessively long intervals may cause vehicles to time out and interrupt charging, while excessively short intervals may cause the vehicle communication module to overload. This multi-dimensional feature analysis covers anomalies in the time dimension of periodic messages, overcoming the shortcomings of traditional testing methods.

[0048] Through the above technical solution, this invention can effectively detect communication compatibility issues caused by the unstable periodic message transmission of charging piles, such as vehicles stopping charging due to the protection mechanism being triggered by excessive message intervals. By simulating different interval characteristics using virtual charging piles, the vehicle's adaptability to message periodic fluctuations can be verified in advance during the R&D stage, reducing charging failures caused by abnormal protocol interactions.

[0049] In one embodiment, the present invention further proposes to generate feature parameters based on charging data, including generating message response features. For each type of message, the response time between the time when the pile receives the message and the time when the vehicle sends the message is measured is measured. Several response times are clustered to obtain the maximum value, minimum value and feature value, thus obtaining a set of message response features.

[0050] Among them, message response characteristics refer to the evaluation index constructed by quantifying the communication interaction delay between the vehicle end and the pile end. Specifically, it can be implemented by using a timestamp recording device, deploying high-precision clock modules in the vehicle end controller and the pile end controller respectively to accurately record the message sending and receiving times.

[0051] Response time refers to the time difference between when the vehicle sends a message and when the pile receives the message. Specifically, a CAN bus analyzer can be used to capture time nodes in the message transmission path, and the single response time data can be obtained by calculating the difference between the two time nodes. Cluster analysis involves extracting features from the response time dataset obtained from multiple tests. Specifically, the K-means algorithm can be used to divide the data into typical categories, thereby extracting feature parameters representing extreme latency, normal latency, and optimal latency.

[0052] Specifically, during the charging compatibility test, the time difference between the vehicle-side sending time and the charging pile-side receiving time was recorded for different types of messages during the charging handshake phase, parameter configuration phase, and charging phase. For example, during the charging handshake phase, the vehicle identification message was tested 50 times consecutively, and the time taken for the message to travel from the vehicle-side controller to the charging pile-side controller was recorded for each test. These 50 sets of time-consuming data were input into a clustering algorithm to classify the message into three feature categories: maximum latency, minimum latency, and typical latency. These three feature values ​​were then loaded into the virtual charging pile to simulate the characteristics of slow response, fast response, and normal response, respectively, thereby reproducing compatibility issues such as charging interruptions and charging start-up failures caused by abnormal message responses during testing.

[0053] This invention constructs a test environment that can simulate dynamic response characteristics by collecting real interaction data and extracting multi-dimensional time delay features, which can effectively detect the vehicle's adaptability to charging piles with different response characteristics.

[0054] Through the above technical solution, this invention can accurately identify response timeout fault tolerance defects in vehicles during message interaction. For example, when simulating the maximum delay characteristic at the pile end, it can verify whether the vehicle initiates the timeout retransmission mechanism within a preset time; when simulating the minimum delay characteristic, it can detect whether the vehicle controller has insufficient high-frequency message processing capabilities. This test environment, built based on real data characteristics, significantly improves the coverage and fault detection rate of compatibility testing.

[0055] In one embodiment, the present invention further proposes to introduce normative features when generating feature parameters based on charging data, specifically including response normative features, protocol normative features, and multi-frame message normative features.

[0056] Among them, the response standardization feature refers to the quantification of the charging pile's response accuracy to vehicle dynamic requests by calculating the standardized deviation between the vehicle-side request value and the charging pile-side output value for each request parameter. Specifically, it can be achieved using the formula (vehicle-side request value - charging pile-side output value) / (charging pile-side output value × battery voltage). This formula eliminates the influence of different voltage levels on parameter evaluation by correlating the deviation with the battery voltage, thereby more objectively reflecting the standardization of the charging pile's output parameters.

[0057] Protocol standardization features refer to the types and frame ratios of non-fixed messages in each stage of the statistical interaction protocol. Specifically, this can be achieved by parsing the message types in the communication logs for each stage, filtering out messages that do not conform to the protocol's fixed requirements, and calculating their proportion. This feature can identify compatibility anomalies caused by message type confusion or redundancy during protocol execution. Taking the GB / T 27930-2015 protocol as an example, in addition to the physical connection completion and low-voltage auxiliary power-on stages, it can be divided into the charging handshake stage, the charging parameter configuration stage, the charging stage, and the charging end stage. In each stage, except for event-type messages, the charging pile should only send messages within the corresponding stage. By analyzing the data collected at the charging pile during the charging process, the types and frame ratios of non-standard messages sent by the pile are recorded, forming the protocol standardization features. The standardized characteristics of multi-frame messages refer to recording the transmission time of long messages, interruption frame information, and retransmission information. Since a set of CAN messages can only transmit 8 bytes of data, while actual communication protocols such as BRM and BCS messages exceed 8 bytes, the transmission of these long messages requires a multi-frame request protocol. The multi-frame protocol includes four stages: transmission request, transmission confirmation, data transmission, and transmission completion. The standardized characteristics of multi-frame messages are formed by analyzing the time taken to send the request at the charging station, transmission interruption situations, and successful retransmissions after interruptions. During a charging process, long messages are also sent periodically, but not every cycle is successful. There are cases of successful transmission, cases of unsuccessful retransmissions after interruptions, and cases of successful retransmissions after interruptions. Therefore, it is necessary to monitor and record these situations. Specifically, this can be achieved by monitoring the transmission process of long messages exceeding 8 bytes, recording the total time from the first frame to the last frame, the message ID and frame sequence number corresponding to the interruption, and the number of repetitions during retransmission. This characteristic can expose transmission efficiency defects or insufficient interruption recovery capabilities in the charging station's multi-frame message processing mechanism.

[0058] Specifically, when the charging pile output reaches a steady state, the response normative characteristics of the voltage and current request parameters for each cycle are calculated. Cluster analysis is then used to extract the maximum, minimum, and typical characteristic values, forming an evaluation index reflecting the charging pile's dynamic response capability. Simultaneously, during the charging handshake, parameter configuration, charging phase, and termination phase, the frequency and type distribution of non-fixed messages in each phase are statistically analyzed to identify deviations in protocol execution. For long messages exceeding 8 bytes, a quantitative index of multi-frame message transmission reliability is constructed by recording transmission time, interruption frame location, and retransmission count. These features are collectively used for anomaly simulation of the virtual charging pile, exposing compatibility issues caused by parameter deviations, protocol confusion, or transmission interruptions in a virtual environment for the vehicle under test.

[0059] This invention quantifies the accuracy of dynamic parameter matching through response normative features, identifies protocol execution deviations through protocol normative features, and evaluates the stability of long message transmission through multi-frame message normative features. It is the first to construct a compatibility testing system covering multiple dimensions of parameters, protocols, and transmission. This invention can effectively identify parameter output deviations of charging piles when responding to vehicle dynamic requests, detect communication anomalies caused by non-standard messages during protocol interaction, and expose interruptions and recovery defects in multi-frame message transmission. This allows for the discovery of potential compatibility issues in the early stages of development, reducing charging failures caused by differences in communication protocol implementation.

[0060] In one embodiment, the present invention further proposes that the time taken for successful transmission includes the time taken from the moment the first frame is sent to the moment the last frame is sent; the interruption information for transmission interruption includes the message ID of each transmission and the sequence number of the interrupted frame; and the retransmission information for successful retransmission after transmission interruption includes the message ID of each transmission, the sequence number of the repeated transmission frame, and the number of times.

[0061] The successful transmission time refers to the time span from the start to the end of message transmission. This can be achieved by using a high-precision timer to record the difference between the time the first frame is sent and the time the last frame is received and acknowledged, thus quantifying the efficiency and stability of the charging pile during long message transmission. The interruption information refers to the identification data generated during transmission interruptions. This can be obtained by parsing the CAN bus log to obtain the message ID and the sequence number of the interrupted frame, used to pinpoint the specific message type and transmission stage where the interruption occurred. The successful retransmission information refers to the data records generated when transmission resumes after an interruption. This can be achieved by comparing the frame sequence numbers and retransmission counts during the original transmission and the retransmission process, used to evaluate whether the charging pile's retransmission mechanism complies with the protocol specifications.

[0062] Specifically, during long message transmission, by recording the time taken from the first frame to the last frame, it is possible to identify whether the charging pile experiences abnormal transmission efficiency due to excessively long frame intervals or response delays. When transmission is unexpectedly interrupted, by collecting the message ID and interrupted frame sequence number at the time of the interruption, the protocol interaction stage and message type at which the interruption occurred can be accurately traced. For example, it can be determined whether the communication anomaly is caused by the charging pile's incorrect handling of multi-frame fragmentation. During retransmission, by statistically analyzing the sequence number and repetition count of retransmitted frames, it is possible to verify whether the charging pile retransmits only the lost frame or completely repeats the entire message sequence as required by the protocol, thereby detecting the standardization of the retransmission logic.

[0063] In some specific implementations, transmission time can be calculated by embedding a timestamp module in the virtual charging station to record the transmission and reception times of each frame at the millisecond level and generate a time difference curve. Interruption information can be collected by monitoring the CAN bus status in real time through a protocol parser; when a frame sequence interruption is detected, the current message ID and the number of transmitted frames are automatically recorded. Retransmission information can be statistically analyzed using a combination of frame counters and hash tables to match and count the sequence numbers of repeatedly transmitted frames and generate a retransmission log.

[0064] This invention achieves protocol compliance assessment of the entire long message transmission process by quantifying transmission time, accurately recording interruption locations, and statistically analyzing retransmission behavior. Through the above technical solution, this invention can effectively identify compatibility issues in charging piles caused by excessive transmission time, faulty interruption recovery mechanisms, or non-standard retransmission logic during long message transmission. For example, when a certain model of charging pile transmits a battery status message exceeding 8 bytes, if its transmission time is significantly higher than the characteristic value of cluster analysis, it can be determined that the pile has a transmission efficiency defect; if the interruption information shows multiple transmission failures at specific frame sequence numbers, the frame segmentation algorithm can be optimized accordingly; if the retransmission information indicates that the charging pile frequently retransmits successfully transmitted frames, its retransmission mechanism can be determined to violate protocol specifications.

[0065] This invention further proposes an electric vehicle DC charging compatibility testing system for implementing the above-mentioned testing method, such as... Figure 2 As shown, the system includes a test vehicle, a DC charging pile, a feature module, a vehicle under test (DUT), and a virtual charging pile. The test vehicle is used to charge the DC charging pile and send charging data to the feature module; the DC charging pile is used to connect to the test vehicle for charging tests and also sends charging data to the feature module; the feature module generates feature parameters based on the charging data and sends them to the virtual charging pile; the DUT is used to connect to the virtual charging pile for charging; after loading the feature parameters, the virtual charging pile connects to the DUT for charging compatibility testing.

[0066] The test vehicle refers to an electric vehicle equipped with a power battery and charging interface, which can be a mass-produced model or a customized test platform, used to collect vehicle-side messages and measurement parameters in actual charging scenarios. The DC charging pile refers to a physical charging device that meets national standards, specifically a commercially available charging pile with data acquisition capabilities, used to acquire the timing of pile-side messages and output parameters during actual charging. The feature module refers to a data processing unit, specifically implemented using edge computing devices or cloud servers, which extracts features from charging data, such as message response time and parameter deviation, through clustering algorithms. The virtual charging pile refers to a protocol simulation device, specifically implemented using an embedded system or virtualization platform, which reconstructs the communication behavior patterns of different charging piles by loading feature parameters. The vehicle under test refers to a vehicle connected to a virtual charging pile with loaded feature parameters, used to test communication compatibility by monitoring data and messages during the charging process of the vehicle under test. The vehicle under test and the test vehicle can be the same or different. Preferably, different typical vehicles are selected as the test vehicle and the vehicle under test respectively according to actual needs to better test line-of-sight charging compatibility. Alternatively, the vehicle under test can be partially or completely the same as the test vehicle. When connecting to a virtual charging pile for charging compatibility testing, the charging data when connecting to a DC charging pile can be compared to reflect the test results.

[0067] Specifically, after the test vehicle completes the actual charging test with the DC charging pile, the feature module extracts key indicators such as message response time and parameter deviation percentage from the charging data, and generates maximum, minimum, and typical values ​​reflecting the communication characteristics of the charging pile through a clustering algorithm. After receiving the feature parameters, the virtual charging pile reconstructs the communication timing based on the message ID and time interval, simulating the protocol interaction mode of a specific charging pile. When the vehicle under test connects to the virtual charging pile, the system monitors the vehicle's response status to simulated messages to determine the compatibility of its protocol implementation. For example, if the virtual charging pile loads the message response time characteristics of a certain brand of charging pile, and the vehicle under test fails to complete the protocol handshake within a preset time, a communication adaptation defect is determined to exist.

[0068] This invention utilizes message timing and parameter deviation data extracted by the feature module to reproduce the abnormal response behavior of actual charging piles in a laboratory environment, such as typical fault modes like message retransmission timeout and parameter output fluctuation.

[0069] Through the above technical solution, this invention achieves comprehensive verification of vehicle communication protocol adaptability during the R&D phase, enabling multi-scenario testing without waiting for physical charging pile deployment. The closed-loop testing process between the vehicle under test and the virtual charging pile reduces the frequency of physical charging pile calls and shortens the testing cycle delay caused by equipment scheduling. By configurable feature parameters, it is possible to generate communication feature combinations from different charging pile manufacturers in batches, effectively covering compatibility boundary conditions caused by differences in protocol implementation.

[0070] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. 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 be included within the scope of protection of the present invention. Contents not described in detail in this specification belong to prior art known to those skilled in the art.

Claims

1. An electric vehicle DC charging compatibility test method, characterized in that: charging test is performed on the test vehicle to obtain vehicle charging data; feature parameters are generated based on the charging data, and the feature parameters are loaded into a virtual charging pile; the virtual charging pile and the vehicle to be tested are connected for charging compatibility test.

2. The electric vehicle DC charging compatibility test method of claim 1, wherein, The charging test is performed on the test vehicle to obtain vehicle charging data, including: selecting several electric vehicles as test vehicles, repeatedly performing power consumption and charging operation on each test vehicle, and recording the charging data in each charging process.

3. The electric vehicle DC charging compatibility test method of claim 2, wherein, The power consumption and charging operation is: the power consumption of the test vehicle is consumed to SOC0, the test vehicle is connected to the DC charging pile, and the test vehicle is charged from SOC0 to 100%, and the SOC0 is less than or equal to the set low power value.

4. The electric vehicle DC charging compatibility test method of claim 1, wherein: The charging data includes pile end message parameters, vehicle end message parameters, and vehicle end measurement parameters.

5. The electric vehicle DC charging compatibility test method of claim 1, wherein, The feature parameters are generated based on the charging data, including: generate parameter accuracy feature: for each measurement parameter, calculate the deviation percentage of each cycle measurement parameter when the charging pile output reaches steady state, cluster a number of deviation percentages to obtain maximum, minimum and characteristic value, and obtain a group of parameter accuracy features.

6. The electric vehicle DC charging compatibility test method of claim 1, wherein, The feature parameters are generated based on the charging data, including: generate periodicity feature: for each periodic message sent by the pile end, measure the time interval between the sending of the same message twice, cluster a number of time intervals to obtain maximum, minimum and characteristic value, and obtain a group of periodicity features.

7. The electric vehicle DC charging compatibility test method of claim 1, wherein, The feature parameters are generated based on the charging data, including: generate message response feature: for each type of message, measure the response time between the time when the pile end receives the message and the time when the vehicle end sends the message, cluster a number of response times to obtain maximum, minimum and characteristic value, and obtain a group of message response features.

8. The electric vehicle DC charging compatibility test method of claim 1, wherein, The feature parameters are generated based on the charging data, including generating specification features: (1) response specification feature: for each request parameter, calculate the specification feature of each cycle request parameter when the charging pile output reaches steady state, cluster a number of specification features to obtain maximum, minimum and characteristic value, and obtain a group of response specification features; (2) protocol specification feature: for multiple stages of the interaction protocol, count the type and frame number proportion of the non-standard messages sent by the pile end in each stage to form the protocol specification feature; (3) multi-frame message specification feature: for each long message, count the time-consuming time of successful transmission of long messages by the pile end, the interruption information of transmission interruption, and the retransmission information of successful retransmission after transmission interruption to form the multi-frame message specification feature.

9. The electric vehicle DC charging compatibility test method of claim 8, characterized in that: the time-consuming time of successful transmission includes the time-consuming time between the time when the first frame starts to send and the time when the last frame is sent to completion for each transmission; the interruption information of transmission interruption includes the message ID and the sequence number of the interrupted frame for each transmission; the retransmission information of successful retransmission after transmission interruption includes the message ID, the sequence number and the number of repeated transmission frames for each transmission.

10. An electric vehicle DC charging compatibility test system, characterized in that: ​ Test vehicle, for connecting with DC charging pile for charging, sending charging data to feature module; DC charging pile, for connecting test vehicle for charging test, sending charging data to feature module; Feature module, for generating feature parameters based on charging data and sending to virtual charging pile; Test vehicle, for connecting with virtual charging pile for charging; Virtual charging pile, for connecting test vehicle for charging compatibility test after loading feature parameters.

Images