CN-121681393-B - Adaptation evaluation method and device for core requirements of control chip based on whole vehicle application scene

CN121681393BCN 121681393 BCN121681393 BCN 121681393BCN-121681393-B

Abstract

The invention discloses an adaptation evaluation method and device for core requirements of a control chip based on a whole vehicle application scene. The method comprises the steps of determining a target whole vehicle application scene of a control chip to be evaluated, generating a test case set containing a multi-dimensional function test based on the target whole vehicle application scene, executing the test case set in a test platform simulating a whole vehicle operation environment to test the control chip to be evaluated, collecting response data generated by the control chip to be evaluated in the test process, performing abnormality judgment and performance quantification based on the response data to obtain an suitability evaluation result of the control chip in the target whole vehicle application scene, and outputting chip type selection suggestions by comparing suitability evaluation results of multiple types of the control chip to be evaluated. The invention can realize the multi-dimensional, quantifiable and contrastable suitability evaluation of the control chip based on the real whole vehicle working condition, thereby providing a reliable basis for accurate and efficient chip model selection.

Inventors

ZHANG YONGCHANG
LI ZHAOLIN

Assignees

北京国家新能源汽车技术创新中心有限公司

Dates

Publication Date: 20260512
Application Date: 20260209

Claims (8)

1. An adaptation evaluation method for core requirements of a control chip based on a whole vehicle application scene is characterized by comprising the following steps: determining a target whole vehicle application scene of a control chip to be evaluated; generating a test case set containing a multi-dimensional function test based on the target whole vehicle application scene; Executing the test case set in a test platform simulating the whole vehicle running environment so as to test the control chip to be evaluated deployed in the test platform; Collecting response data generated by the control chip to be evaluated in the test process; Based on the response data, performing abnormality judgment and performance quantification to obtain an suitability evaluation result of the control chip in the target whole vehicle application scene; Outputting a chip type selection suggestion by comparing the suitability evaluation results of a plurality of types of control chips to be evaluated; the generating a test case set containing a multi-dimensional function test based on the target whole vehicle application scene comprises the following steps: performing function and load analysis on the target whole vehicle application scene to analyze the core requirements of the target whole vehicle application scene on the control chip in a plurality of different dimensions; aiming at the analyzed core requirements of each dimension, corresponding test cases are designed to form the test case set, and the plurality of different dimensions comprise: Computing capability dimension, communication capability dimension, data sampling capability dimension and driving capability dimension; performing function and load analysis on the target whole vehicle application scene, including: Based on the determined target whole vehicle application scene, task disassembly and behavior modeling are carried out on specific control function sub items of the target whole vehicle application scene, and execution logic, data flow, real-time window and resource occupation conditions are analyzed, so that core requirements of the target whole vehicle application scene on a plurality of different dimensions on a control chip are systematically resolved; Designing respective, executable test cases targeted based on the core requirements, each of the test cases intended to simulate one extreme pressure condition or typical operating mode in a corresponding dimension; Systematically integrating and arranging the test cases corresponding to each dimension to form the test case set; the core requirement of the computing capability dimension corresponds to the computing power requirement, and the design of the corresponding test case comprises the steps of designing test cases for checking single-core or multi-core computing throughput of a chip and covering effective computing power of a functional security core and encryption and decryption performance related to information security under a plurality of standard computing models; the core requirement of the communication capability dimension corresponds to the communication requirement, and the design of the corresponding test case comprises the design of the test case for evaluating the response instantaneity, the protocol processing capability and the sustainable data rate of various communication interfaces integrated by the chip under the actual high load; the core requirement of the dimension of the data sampling capability corresponds to the data sampling requirement, and the design of the corresponding test cases comprises the design of an analog/digital conversion channel of a test chip and the test cases of sampling rate, conversion precision, linearity and synchronism of digital I/O; The core requirement of the driving capability dimension corresponds to the driving requirement, and the design of the corresponding test cases comprises the design of the test cases of output current intensity, voltage level, response speed and load stability.
2. The method for evaluating the adaptation of the whole vehicle application scene to the core requirement of the control chip according to claim 1, wherein the test platform simulating the whole vehicle operation environment is a hardware-in-loop test platform, and comprises: the device comprises a hardware-in-loop simulation unit and a control chip bearing plate in signal connection with the hardware-in-loop simulation unit.
3. The method for evaluating the adaptation of the whole vehicle application scene to the core requirements of the control chip according to claim 2, wherein the control chip carrier plate adopts a modularized design and comprises a replaceable chip adaptation unit and a fixed interface and peripheral circuit unit.
4. The method for evaluating the adaptation of a complete vehicle application scene to the core requirements of a control chip according to claim 1, wherein the anomaly determination comprises at least one of the following types of detection: detecting abnormal running state of the control chip; detecting the functional logic correctness of the control chip; Detecting the communication robustness of the control chip; and detecting electrical parameter abnormality of the control chip.
5. The method for evaluating the suitability of the control chip core requirements based on the whole vehicle application scene as claimed in claim 1, wherein the suitability evaluation result comprises: an abnormality determination result indicating whether abnormality and abnormality type of the control chip to be evaluated occur in a test; the performance quantification result comprises one or more performance indexes of the control chip to be evaluated in the target whole vehicle application scene; and generating an suitability conclusion based on the abnormality determination result and the performance quantification result.
6. The method for evaluating the adaptation of the whole vehicle application scene to the core requirement of the control chip according to claim 5, wherein the performance index comprises: Characterizing the actual highest computational throughput of the computational capability; characterizing an error-free maximum communication bandwidth of the communication capability; characterizing a highest sampling rate under guaranteed accuracy of the data sampling capability; The stable output maximum driving capability of the characteristic driving capability.
7. An adaptation evaluation device based on whole car application scene is to control chip core demand, which is characterized by comprising: The determining module is used for determining a target whole vehicle application scene of the control chip to be evaluated; The generating module is used for generating a test case set containing a multi-dimensional function test based on the target whole vehicle application scene; The test platform is used for executing the test case set in the test platform simulating the whole vehicle running environment so as to test the control chip to be evaluated which is deployed in the test platform; The test platform comprises a hardware-in-loop simulation unit and a control chip test board connected with the hardware-in-loop simulation unit; The hardware-in-loop simulation unit comprises: The signal sending unit is used for sending a test excitation signal corresponding to the test case set to the control chip test board so as to simulate a real and comprehensive whole vehicle signal and load; the data acquisition unit is used for acquiring response data generated by the control chip to be evaluated in the test process, and carrying out abnormality judgment and performance quantification based on the response data so as to obtain an suitability evaluation result of the control chip in the target whole vehicle application scene; the result output module is used for outputting chip type selection suggestions by comparing the suitability evaluation results of the plurality of types of control chips to be evaluated; The generating a test case set including a multi-dimensional function test based on the target whole vehicle application scene includes: performing function and load analysis on the target whole vehicle application scene to analyze the core requirements of the target whole vehicle application scene on the control chip in a plurality of different dimensions; aiming at the analyzed core requirements of each dimension, designing corresponding test cases to form the test case set; The plurality of different dimensions includes: Computing capability dimension, communication capability dimension, data sampling capability dimension and driving capability dimension; performing function and load analysis on the target whole vehicle application scene, including: Based on the determined target whole vehicle application scene, task disassembly and behavior modeling are carried out on specific control function sub items of the target whole vehicle application scene, and execution logic, data flow, real-time window and resource occupation conditions are analyzed, so that core requirements of the target whole vehicle application scene on a plurality of different dimensions on a control chip are systematically resolved; Designing respective, executable test cases targeted based on the core requirements, each of the test cases intended to simulate one extreme pressure condition or typical operating mode in a corresponding dimension; Systematically integrating and arranging the test cases corresponding to each dimension to form the test case set; the core requirement of the computing capability dimension corresponds to the computing power requirement, and the design of the corresponding test case comprises the steps of designing test cases for checking single-core or multi-core computing throughput of a chip and covering effective computing power of a functional security core and encryption and decryption performance related to information security under a plurality of standard computing models; the core requirement of the communication capability dimension corresponds to the communication requirement, and the design of the corresponding test case comprises the design of the test case for evaluating the response instantaneity, the protocol processing capability and the sustainable data rate of various communication interfaces integrated by the chip under the actual high load; the core requirement of the dimension of the data sampling capability corresponds to the data sampling requirement, and the design of the corresponding test cases comprises the design of an analog/digital conversion channel of a test chip and the test cases of sampling rate, conversion precision, linearity and synchronism of digital I/O; The core requirement of the driving capability dimension corresponds to the driving requirement, and the design of the corresponding test cases comprises the design of the test cases of output current intensity, voltage level, response speed and load stability.
8. The device for adapting and evaluating the core requirements of the control chip based on the whole vehicle application scene as claimed in claim 7, wherein the control chip test board adopts a structure of a replaceable control chip daughter board and a fixed interface mother board.

Description

Adaptation evaluation method and device for core requirements of control chip based on whole vehicle application scene Technical Field The invention belongs to the technical field of automobile electronics, and particularly relates to an adaptation evaluation method and device for core requirements of a control chip based on a whole automobile application scene. Background As the automotive industry evolves to an electromotive, intelligent, networked depth, its electronic and electrical architecture is undergoing a rapid evolution from distributed control to domain controllers, to centralized architecture for central computing platforms. In the process, the control chip is used as the core of each Electronic Control Unit (ECU) of the whole vehicle, and the performance, reliability and suitability with specific application scenes of the control chip directly determine the functions, safety, efficiency and cost of the whole vehicle system. In different whole car application scenarios such as Battery Management System (BMS), motor controller, advanced Driving Assistance System (ADAS), intelligent cabin, the computational load that control chip faced, real-time requirement, communication data throughput, signal sampling precision and power drive demand are huge. Therefore, how to quickly and accurately screen out the chips with optimal performance and optimal cost from the control chips with various models for specific application scenes becomes a key technical challenge facing the whole vehicle factories and spare part suppliers in the early stage of research and development. At present, the main control chip model selection and evaluation methods in the industry mainly comprise the following two types: The first is a static parameter comparison method based on a chip data manual. The method relies on specification parameters provided by chip manufacturers, such as main frequency, kernel number, memory capacity, peripheral interface types and number, etc., to carry out transverse comparison and screening. However, the method has the remarkable limitations that firstly, the specification parameters are usually theoretical limit values of the chip under ideal and single working conditions, the actual performance of the chip under the real vehicle-mounted environment (such as wide temperature range, electromagnetic interference, multi-task concurrency, complex load and the like) cannot be reflected, and secondly, the method only focuses on isolated functional parameters and lacks comprehensive evaluation on the core capacity cooperative work of the chip under specific vehicle control logic and functional scenes. Therefore, the problem of 'enough parameters but insufficient performance' of a chip after actual loading is usually caused by static type selection, such as control delay caused by calculation bottleneck, packet loss caused by communication congestion, insufficient driving capability and the like, and the stability and the safety of a system are seriously affected. The second category is test verification methods based on Automated Test Equipment (ATE) or hardware-in-loop (HIL). The general ATE test focuses on the electrical characteristics and basic function test of the chip production and manufacturing links, and cannot simulate a complex whole vehicle application scene. While the HIL test can construct a high-fidelity whole vehicle simulation environment, the current application mode is mainly focused on control strategy verification, function test and fault injection of the designed ECU finished product. The test case is designed based on the control logic and the functional requirements of the whole vehicle, and aims to verify the correctness of the controller software, but is not specially used for evaluating and mining the physical performance boundary of the control chip under the limit scene. The method cannot carry out multi-dimensional quantitative performance assessment and transverse comparison on bare chips or different types of chips under uniform and comparable whole vehicle scene pressure at the chip model selection stage. In summary, the prior art scheme has inherent disadvantages of scene disjoint, single evaluation dimension, lack of extreme pressure test, poor transverse comparability and the like in early model selection evaluation of the control chip. The chip model selection process often depends on the experience of engineers, has larger uncertainty, prolongs the development period, and increases the later-stage correction risk and cost caused by improper model selection. Therefore, a new method and apparatus for systematically, quantitatively, and highly-fidelity evaluating the core capability of a control chip in tight combination with a specific whole vehicle application scenario are needed to support scientific, efficient, and accurate chip type selection decision. Disclosure of Invention The invention aims to provide an adaptation evaluation method and device for core requir