Test methods and systems for electric vehicle controllers

Abstract

This application relates to the field of electric vehicle testing technology, and discloses a testing method and system for an electric vehicle controller. The method first establishes an electrothermal impedance hysteresis model based on the physical characteristics of the load under test; it captures critical boundary values ​​by monitoring the steady-state switching action of the output signal, and superimposes a dynamic disturbance signal containing orthogonal components to construct a composite excitation signal, causing the vehicle controller to operate in a critical state; it calculates the effective thermal stress power within the sliding time window based on the voltage change rate, and uses the electrothermal impedance hysteresis model to calculate the virtual internal temperature and target equivalent impedance value in real time; it simulates the target equivalent impedance value by adjusting the physical input impedance, and monitors the convergence trend of the virtual internal temperature and the switching frequency of the control state to determine logic stability. This invention achieves accurate evaluation of the control stability of the vehicle controller under complex electrothermal coupling conditions by reconstructing the load thermal inertia, introducing orthogonal disturbances, and correcting high-frequency losses.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle testing technology, specifically to a testing method and system for an electric vehicle controller. Background Technology

[0002] As the core unit of energy management, the vehicle controller of an electric vehicle is responsible for scheduling the operation of various high-voltage accessories and power loads. Under the complex operating conditions of an electric vehicle, the physical characteristics of high-power components such as PTC heaters and motor stator coils will change significantly with the accumulation of temperature, directly affecting the stability of the control loop.

[0003] Existing testing schemes generally use general-purpose programmable electronic loads as analog terminals for the object under test. During testing, the system injects analog signals into the VCU and controls the electronic load to exhibit specific resistance or voltage characteristics according to preset values. To detect the controller's logic threshold, a linear scanning method is typically used, that is, applying a monotonically increasing or decreasing voltage signal to the input and observing the transition point of the output state. For power assessment, the instantaneous values ​​of voltage and current are often directly read, and the energy consumed by the load is monitored through conventional product operations.

[0004] However, in existing electric vehicle controller testing, the response speed of general-purpose electronic loads is extremely fast, with impedance switching completed almost instantaneously. Real physical devices are affected by thermal capacitance, resulting in impedance changes with lag. This leads to a severe mismatch in time scale, easily masking the risk of low-frequency oscillations caused by thermal hysteresis. Using a single-frequency signal for interference testing can result in the signal frequency coinciding with the controller's sampling frequency, causing metastable logic defects to be missed. Traditional average power algorithms only focus on steady-state values; in the critical region where control logic frequently switches, high-frequency switching losses are filtered out by smoothing algorithms. This calculation method underestimates actual thermal stress and makes it difficult to accurately define the physical boundaries of thermal runaway. Therefore, this invention provides a testing method and system for electric vehicle controllers to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a testing method and system for electric vehicle controllers, which solves the problems of failing to effectively simulate the coupling effect between high-frequency logic switching and low-frequency thermal response in existing electric vehicle controller testing, as well as the difficulty in accurately locating control dead zones and thermal runaway boundaries.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The first aspect of this invention provides a testing method for an electric vehicle controller, comprising the following steps:

[0008] Initialize the test environment and establish an electrothermal impedance hysteresis model for the physical characteristics of the load under test;

[0009] A continuously changing excitation signal is input to the vehicle controller. By monitoring the steady-state switching action of the output signal of the vehicle controller, the critical boundary value that triggers the logic flip is captured and recorded.

[0010] Using the critical boundary value as a static reference value, a dynamic disturbance signal containing orthogonal components is generated and superimposed on the static reference value to form a composite excitation signal, which is then input to the vehicle controller so that the vehicle controller operates in a critical state.

[0011] Collect voltage and current data from the vehicle controller and calculate the effective thermal stress power within the sliding time window based on the voltage change rate.

[0012] Run the electrothermal impedance hysteresis model, calculate the virtual internal temperature based on the effective thermal stress power, calculate the target equivalent impedance value based on the virtual internal temperature, and simulate the target equivalent impedance value by adjusting the physical input impedance;

[0013] The stability of the vehicle controller's control logic is determined by continuously monitoring the convergence trend of the virtual internal temperature and the switching frequency of the control state.

[0014] Preferably, the establishment of the electrothermal impedance hysteresis model for the physical characteristics of the load under test specifically includes:

[0015] Retrieve the pre-stored configuration file, extract the virtual thermal capacity, virtual thermal resistance, nominal impedance value and resistance temperature coefficient, and define the thermal time constant based on the product of virtual thermal capacity and virtual thermal resistance, and limit the thermal time constant to be greater than the response time of the load simulation hardware.

[0016] A thermal equilibrium differential equation is constructed, which takes the effective thermal stress power as the input variable and simulates the thermal inertial physical process through integral operation, transforming the transient power fluctuation into a smooth convergence trend of virtual internal temperature.

[0017] An impedance mapping equation is constructed, which uses the calculated virtual internal temperature as the independent variable and combines the resistance temperature coefficient and the nominal impedance value to derive the target equivalent impedance value that dynamically drifts with temperature. The electrothermal impedance hysteresis model includes a thermal balance differential equation and an impedance mapping equation.

[0018] Preferably, capturing and recording the critical boundary value that triggers the logic flip specifically includes:

[0019] A discrete excitation sequence is constructed in the time domain in a step-like increasing or decreasing manner. The excitation signal value of each step cycle is generated based on the excitation signal value of the previous moment superimposed with a preset scanning step size, and the scanning step size is set to be less than the minimum detection accuracy of the vehicle controller.

[0020] The normalized vehicle controller output signal is substituted into the state difference function for real-time calculation. The change amplitude of the output state is quantified by calculating the difference between adjacent steady-state sampling points.

[0021] Once the calculated result of the state difference function exceeds the preset state switching judgment threshold, a logic flip is determined, the current excitation signal value is immediately locked and extracted, and the excitation signal value is defined as the static DC operating point for subsequent closed-loop testing. The current excitation signal value is the critical boundary value.

[0022] Preferably, the step of generating a dynamic disturbance signal containing orthogonal components and superimposing it onto a static reference value specifically includes:

[0023] Two AC components that are coprime and have frequencies higher than the closed-loop bandwidth of the control algorithm are selected, and a digital waveform of the dynamic disturbance signal is constructed using a waveform synthesis algorithm under the condition of phase orthogonality.

[0024] A random initial phase angle is generated by a pseudo-random number generator and added to the digital waveform of the dynamic perturbation signal to construct a dynamic perturbation sequence with differentiated initial states.

[0025] The dynamic perturbation sequence is superimposed with the static reference value to obtain a digital sequence of composite excitation signal;

[0026] The digital sequence of the composite excitation signal is converted into a physical electrical signal through a digital-to-analog converter and power amplification stage. A stress field that continuously fluctuates around the state switching judgment threshold is constructed at the input of the vehicle controller to stimulate the high-frequency switching behavior of the control logic.

[0027] Preferably, the calculation of the effective thermal stress power within the sliding time window based on the voltage change rate specifically includes:

[0028] The voltage data of the measured load is subjected to first-order differential operation to obtain the voltage change rate, and the voltage change rate is compared with the voltage change rate threshold, which is determined based on the boundary between the linear amplification region and the saturation region of the power device.

[0029] Based on the comparison results, an oscillation weighting factor is generated in real time: when the voltage change rate exceeds the voltage change rate threshold, the circuit is determined to be in a high-frequency switching state, and the preset high-frequency loss coefficient is retrieved as the oscillation weighting factor; otherwise, the oscillation weighting factor is set to a unit value.

[0030] Within a set sliding time window, the product of voltage and current data is weighted and integrated using the oscillation weighting factor, thereby converting high-frequency switching losses and skin effect losses into an increase in effective thermal stress power.

[0031] Preferably, the step of running the electrothermal impedance hysteresis model, calculating the virtual internal temperature based on the effective thermal stress power, and calculating the target equivalent impedance value based on the virtual internal temperature specifically includes:

[0032] The thermal equilibrium differential equation is invoked and solved in real time using the discrete difference method. The virtual internal temperature of the previous calculation step is used as the system state variable, and the currently calculated effective thermal stress power is used as the input variable. The virtual internal temperature at the current moment is calculated through numerical integration iteration.

[0033] Substituting the virtual internal temperature into the impedance mapping equation, if the measured load is linear, the impedance drift is calculated through the linear equation, and the target equivalent impedance value is obtained based on the impedance drift and the nominal impedance value. If the measured load is nonlinear, the impedance-temperature characteristic curve data table is called by the lookup table method to reproduce the nonlinear impedance characteristics of the material near a specific temperature point and obtain the corresponding target equivalent impedance value.

[0034] The calculated target equivalent impedance value is subjected to safety limiting processing to obtain an impedance command. The final output impedance command is limited to the safe range defined by the short-circuit protection threshold and the open-circuit protection threshold of the vehicle controller.

[0035] Preferably, determining the stability of the vehicle controller's control logic further includes:

[0036] The virtual internal temperature sequence within the observation time window is extracted. The local peaks and valleys of the waveform of the virtual internal temperature sequence are identified by extreme value search to define the continuous oscillation period in the time domain. The amplitude ratio of adjacent oscillation periods is calculated to generate the oscillation decay rate.

[0037] The calculated oscillation decay rate is compared with the pass / fail threshold: if the oscillation decay rate is less than the pass / fail threshold, it indicates that the system has positive damping characteristics and is judged as thermal convergence; if the oscillation decay rate is greater than or equal to the pass / fail threshold, it indicates that the system is in a constant amplitude oscillation or divergence state, and the vehicle controller is judged to have a risk of thermal runaway and the test is terminated.

[0038] Preferably, the determination of the stability of the vehicle controller's control logic also includes dual verification based on state transition density:

[0039] The total number of logic level switches in the output signal of the vehicle controller within the observation time window is counted and divided by the duration of the observation time window to generate the state flip density.

[0040] The state flip density is compared with the logic stability threshold determined based on the physical lifetime of the execution device;

[0041] The vehicle controller is confirmed to have reached stability at both the thermodynamic macroscopic level and the logical microscopic level only when the determined oscillation decay rate is qualified and the state reversal density is simultaneously lower than the logic stability threshold. The vehicle controller is then deemed to have passed the test.

[0042] Preferably, the setting of the AC component amplitude in the generation of the dynamic disturbance signal containing orthogonal components follows the following constraints:

[0043] An upper limit constraint is added to the frequency setting of the AC component, requiring the frequency to be lower than the cutoff frequency of the hardware low-pass filter at the input of the vehicle controller, so as to ensure that the disturbance energy is not filtered out by the hardware circuit.

[0044] In the digital-to-analog conversion stage, the update rate of the digital-to-analog converter is set to be greater than ten times the highest frequency of the AC components;

[0045] In the signal driving stage, the bandwidth of the linear power amplifier circuit is configured to cover the range from DC to 1 kHz, and it has low output impedance characteristics.

[0046] A second aspect of the present invention provides a test system for implementing the test method of the electric vehicle controller, comprising:

[0047] The main control module is used to initialize the test environment and establish an electrothermal impedance hysteresis model, receive voltage data fed back by the data acquisition module, calculate the voltage change rate based on the voltage data, calculate the effective thermal stress power, run the electrothermal impedance hysteresis model to solve the virtual internal temperature and target equivalent impedance value in real time, generate and send the corresponding impedance command to the load simulation module, and determine the control logic stability of the electric vehicle controller based on the convergence trend of the virtual internal temperature and the switching frequency of the control state.

[0048] The simulation module is used to run the electric vehicle dynamics model and interact with the vehicle controller through the electric vehicle communication bus. It sends wake-up messages and simulated vehicle speed signals to the vehicle controller to ensure that the functional loops of the vehicle controller under test are in an active and ready state.

[0049] The excitation injection module is used to respond to the instructions of the main control module. During the stage of acquiring the critical boundary value, it outputs a discrete excitation sequence that changes in a step-like manner. During the closed-loop test stage, it generates a composite excitation signal containing orthogonal components and superimposed on the static reference value to drive the vehicle controller to operate in the critical state.

[0050] The data acquisition module is used to monitor the output signal status of the vehicle controller in real time to identify logic flipping actions, and synchronously acquire the voltage and current data applied by the vehicle controller to the two ends of the load simulation module and transmit them to the main control module.

[0051] The load simulation module is used to receive impedance commands sent by the main control module and dynamically adjust its own physical input impedance characteristics to simulate the calculated target equivalent impedance value.

[0052] This invention provides a testing method and system for the vehicle controller of an electric vehicle. It has the following beneficial effects:

[0053] 1. This invention establishes an electrothermal impedance hysteresis model for the physical characteristics of the load under test. This model uses thermal balance differential equations to transform transient power fluctuations into a smooth convergence trend of virtual internal temperature. It also calculates the target equivalent impedance value in real time through impedance mapping equations and controls the execution of the load simulation module. This enables the test system to accurately reproduce the temperature rise hysteresis characteristics and nonlinear impedance drift process of high-power physical devices under long-term operation when using electronic loads with extremely fast response speeds. Thus, the closed-loop stability verification of the vehicle controller under complex thermo-electric coupling conditions is completed without actually heating the physical devices or damaging the test equipment.

[0054] 2. This invention employs a dynamic disturbance signal generation and injection technology based on orthogonal components. By superimposing two mutually prime and orthogonal AC components on the captured critical logic flip boundary, a composite stress field with non-periodic ergodic characteristics is constructed. This enables the sampling and judgment logic of the vehicle controller to repeatedly traverse near the threshold, thereby fully stimulating potential metastable behavior and avoiding the phenomenon of fixed sampling points and missed fault detection caused by the sampling frequency of the vehicle controller being a multiple of the signal frequency.

[0055] 3. This invention proposes a detection mechanism for calculating effective thermal stress power based on voltage change rate. By differentially monitoring the voltage signal, it can identify high-frequency switching actions in real time and automatically apply an oscillation weighting factor, converting switching losses and skin effect losses into an increment of effective thermal stress power. Combined with dual judgment indicators based on oscillation decay rate and state reversal density, it can identify the thermal runaway risk of the system from a macroscopic thermodynamic level and quantify the operating frequency of the actuator from a microscopic logic level, effectively preventing the physical life loss of the actuator caused by the accumulation of small jitters in the control logic. Attached Figure Description

[0056] Figure 1 This is a test system architecture diagram for implementing the test method of the electric vehicle controller of the present invention;

[0057] Figure 2 This is a flowchart of the test method for the electric vehicle controller of the present invention;

[0058] Figure 3 This is a flowchart illustrating the system stability evaluation logic of the present invention.

[0059] Among them, 100 is the main control module; 200 is the simulation module; 300 is the excitation injection module; 400 is the data acquisition module; and 500 is the load simulation module. Detailed Implementation

[0060] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] See attached document Figure 1 , Figure 1 This is a system architecture diagram of one embodiment of the present invention. The present invention provides a test system for implementing the test method of the electric vehicle controller, which mainly includes a main control module 100, a simulation module 200, an excitation injection module 300, a data acquisition module 400, a load simulation module 500, and an interface module. The various modules are connected via a data bus and signal harness.

[0062] The main control module 100 is the core control unit of the test system, used to run the test management software and mathematical model algorithms. The main control module 100 integrates a logic boundary search unit, a disturbance generation unit, and a thermal impedance reconstruction unit. The main control module 100 is responsible for sending control commands to other modules, processing the feedback monitoring data, and performing timing scheduling and numerical calculations.

[0063] The simulation module 200 is communicatively connected to the main control module 100 and is used to run the electric vehicle dynamics model. The simulation module 200 simulates the physical parameters of the electric vehicle during operation and interacts with the vehicle controller in real time through the electric vehicle communication bus.

[0064] The excitation injection module 300 is connected to the main control module 100 and the signal input terminal of the vehicle controller. The excitation injection module 300 is equipped with a digital-to-analog converter circuit, which is used to convert the control commands generated by the main control module 100 into physical electrical signals and output analog voltage signals or analog resistance signals to the vehicle controller.

[0065] The data acquisition module 400 is connected to the power output terminal of the vehicle controller and the main control module 100. The data acquisition module 400 is equipped with a high-speed analog-to-digital converter to acquire the drive voltage waveform and loop current waveform output by the vehicle controller and transmit the waveform data to the main control module 100.

[0066] The load simulation module 500 is connected to the power drive interface of the vehicle controller. The load simulation module 500 consists of a programmable electronic load with a programmable communication interface, which is used to consume electrical energy and adjust the equivalent impedance characteristics of the input port according to the received control commands.

[0067] The interface module includes a custom wiring harness and a signal adapter box, which are used to realize the physical electrical connection between the vehicle controller and the above modules.

[0068] See attached document Figure 2 , Figure 2 This is a flowchart of the testing method for the electric vehicle controller of the present invention. The testing method for the electric vehicle controller provided by the present invention includes the following steps:

[0069] S10, the main control module 100 initializes the test environment and establishes the electrothermal impedance hysteresis model of the load under test;

[0070] Before performing the S20 scan, the main control module 100 controls the simulation module 200 to send necessary wake-up messages and simulated vehicle speed signals to the vehicle controller via the CAN bus to ensure that the tested functional circuit is in an active standby state. The main control module 100 controls the excitation injection module 300 to input continuously changing excitation signals to the vehicle controller and monitors the output signals of the vehicle controller in real time. When the output signal is detected to switch from one steady state to another, the excitation signal value at the moment of switching is recorded as the critical boundary value.

[0071] S30, the main control module 100 selects the critical boundary value recorded in step S20 as the static reference value of the excitation signal, calls the preset waveform generation algorithm to generate a dynamic disturbance signal containing sine and cosine components, superimposes the dynamic disturbance signal onto the static reference value to form a composite excitation signal and inputs it into the vehicle controller, so that the vehicle controller operates in the critical state.

[0072] S40, the data acquisition module 400 synchronously acquires the voltage and current data applied by the vehicle controller to both ends of the load simulation module 500, and the main control module 100 receives the voltage and current data and calculates the effective thermal stress power within the sliding time window.

[0073] S50, the main control module 100 uses the electrothermal impedance hysteresis model to calculate the virtual internal temperature of the simulated load based on the effective thermal stress power, and calculates the target equivalent impedance value that the load simulation module 500 should present based on the virtual internal temperature, and sends the target equivalent impedance value to the load simulation module 500.

[0074] S60, the load simulation module 500 adjusts the electrical characteristics of its physical input impedance to simulate the target equivalent impedance value. The main control module 100 continuously runs the electrothermal impedance hysteresis model and monitors the virtual internal temperature sequence and control state sequence in real time. The stability of the vehicle controller's control logic is determined based on whether the dynamic characteristics of the virtual internal temperature sequence and control state sequence converge.

[0075] In the initialization phase of the test method in step S10, the main control module 100 retrieves the pre-stored configuration file and establishes an electrothermal impedance hysteresis model for the physical characteristics of the load under test. This electrothermal impedance hysteresis model is an energy mapping mechanism based on time integration, used to solve the frequency domain mismatch problem between the high-frequency logic switching of the vehicle controller and the low-speed response of the physical load. By converting transient power fluctuations into a convergent trend of virtual internal temperature with thermal inertia, this electrothermal impedance hysteresis model is mathematically equivalent to a low-pass filter, which can smooth the spikes in the high-frequency drive signal and retain only the energy components that cause substantial thermal damage to physical devices.

[0076] The main control module 100 loads the physical property parameters of the load under test. These physical property parameters correspond to the inherent thermodynamic and electrical characteristics of actual on-board high-power accessories (such as PTC heaters and motor stator coils). Specific physical property parameters include the virtual heat capacity C. th Virtual thermal resistance R th Nominal impedance value Z base And the temperature coefficient of resistance α.

[0077] To ensure that the rate of change of the target equivalent impedance value output by the electrothermal impedance hysteresis model can be effectively executed by the subsequent load simulation module 500, the physical property parameters must satisfy the time constant constraint. Define the thermal time constant τ = R. th C th The thermal time constant τ must be set to a value greater than the hardware response time constant of the load simulation module 500 (usually set to greater than 100 milliseconds). The virtual heat capacity C... th The degree of hysteresis in the temperature response is determined by C. th The value range of is determined based on the physical mass and specific heat volume of the simulated device; virtual thermal resistance R th This determines the heat dissipation equilibrium point under steady-state conditions of the system.

[0078] Based on the above physical property parameters, the main control module 100 establishes a thermal balance differential equation. This thermal balance differential equation describes the dynamic evolution of the virtual internal temperature of the measured load with respect to the effective thermal stress power input. The thermal balance differential equation is defined as follows:

[0079] C th (dT virt (t)) / dt=P stress(t)-(T virt (t)-T amb ) / R th ;

[0080] In the formula, d represents the differential operation symbol; dt represents the differential increment of the continuous-time variable t; T virt (t) represents the virtual internal temperature of the load being measured at time t; T amb This represents the ambient temperature constant, set as the constant reference temperature of the testing laboratory, typically taken as 298.15K (i.e., 25℃); P stress (t) represents the input effective thermal stress power, in watts (W). This effective thermal stress power is the input variable of the electrothermal impedance hysteresis model and is calculated in real time by the weighted integral of the driving waveform in subsequent steps; t is a continuous time variable.

[0081] After obtaining the real-time calculated value of the virtual internal temperature, the main control module 100 establishes an impedance mapping equation based on the resistivity characteristics of the measured load material. This impedance mapping equation transforms the state variables in the thermal domain into control objectives in the electrical domain. The impedance mapping equation is defined as follows:

[0082] Z target (t)=Z base [1+α(T virt (t)-T ref )];

[0083] In the formula, Z target (t) represents the target equivalent impedance value calculated at time t; Z base This indicates that the measured load is at the reference temperature T. ref The initial nominal impedance value is obtained by measuring the cold-state resistance; T ref T represents the reference temperature at which the nominal impedance value is defined. ref The value of T amb To maintain consistency; α represents the temperature coefficient of resistance of the load material being measured. For loads with copper coils, α is set to the range of 0.0039 to 0.0040. For loads with PTC thermistors, α is set to the corresponding nonlinear piecewise function or lookup table mapping value.

[0084] To achieve real-time computer calculation, the main control module 100 uses a numerical integration algorithm to discretize the aforementioned thermal equilibrium differential equation, setting the discretization step size h to be equal to the control cycle of the main control module 100 (e.g., 10 ms). Within each step cycle, the main control module 100 uses the virtual internal temperature from the previous moment and the effective thermal stress power input at the current moment to iteratively calculate the increment of the virtual internal temperature at the current moment using either the fourth-order Runge-Kutta method or the forward Euler method. The specific code implementation of the numerical integration algorithm is a well-known technique in the field of computational mathematics and is known to those skilled in the art; therefore, it will not be elaborated upon here.

[0085] For step S20, after constructing the electrothermal impedance hysteresis model, the system executes an automatic threshold capture program. This automatic capture program is used to locate the metastable operating point of the vehicle controller's control algorithm, that is, the region where the control output is most sensitive to the rate of change of the input variable. Within this region, small physical characteristic drifts are most likely to induce logic oscillations. Therefore, accurately locking this threshold value is a prerequisite for the effectiveness of subsequent stability testing.

[0086] The main control module 100 configures the excitation injection module 300 to operate in a mode that applies a monotonically changing excitation signal to the signal input port of the vehicle controller. The physical type of this excitation signal matches the input characteristics of the functional circuit under test: for voltage protection logic, it outputs an analog high-voltage source signal; for temperature protection logic, it outputs an analog variable resistance signal. The main control module 100 generates a discrete excitation sequence that presents a stepped structure in the time domain according to a preset linear scanning strategy.

[0087] To prevent boundary positioning errors caused by a delayed response from the vehicle controller due to excessively fast scanning speed, the update frequency of the excitation signal must be strictly controlled. Within each scanning step cycle, the current excitation signal value u(t) is incremented based on the previous excitation signal value u(t-Δt). The formula for generating the excitation signal is defined as follows:

[0088] u(t) = u(t - Δt) + δsgn(U end -U start );

[0089] In the formula, u(t) represents the value of the excitation signal applied to the vehicle controller at the current moment;

[0090] Δt represents the scanning step cycle. The value of this parameter must be greater than the sum of the "internal filtering time of the vehicle controller" and the "communication bus transmission delay". It is usually set to 50ms to 200ms to ensure that the vehicle controller can achieve a steady-state response under each excitation step.

[0091] δ represents the scan step size, which determines the resolution of the boundary search. The value must be less than the minimum detection accuracy specified in the vehicle controller specification (for example, if the specification accuracy is 0.5V, then δ is set to 0.1V).

[0092] U start and U end U represents the start and end values ​​of the preset scan interval, respectively. start and U end The dynamic range of the test was defined;

[0093] sgn() is a sign function used to automatically adapt to incremental or decremental scan modes.

[0094] While applying the excitation signal, the main control module 100 acquires the output signal of the vehicle controller in real time through the data acquisition module 400 or the electric vehicle bus interface. In order to uniformly process the output signals of different physical dimensions (such as analog voltage values, PWM duty cycle percentage, and CAN message status words), the main control module 100 first normalizes the acquired raw data and then maps it to the [0,1] interval to obtain the dimensionless value y(t).

[0095] Subsequently, the main control module 100 calculates the amplitude of the output signal change between adjacent steady-state sampling points to identify logic flip actions. The determination of logic flip is based on the state difference function D. stat Whether (t) exceeds the preset state transition threshold, the state determination logic formula is as follows:

[0096] D stat (t)=|y(t)-y(t-Δt)|;

[0097] The judgment condition is: if D stat (t)>Y th If so, then a logical flip has occurred;

[0098] In the formula, y(t) represents the normalized output response value of the vehicle controller at the current moment; y(t-Δt) represents the normalized output response value at the previous steady-state moment; Y th This represents the state transition determination threshold, which is set based on the signal-to-noise ratio (SNR): for discrete switch outputs (such as relay drive states), Y... th Set to 0.5; for continuous analog outputs (such as PWM duty cycle), Y th Set it to 0.05 to 0.1 (i.e., a change of 5% to 10%) to be able to distinguish between signal noise and actual logic adjustment.

[0099] Once D is detected statIf the above judgment condition is met, the main control module 100 immediately interrupts the scanning process, extracts and stores the excitation signal value u(t) at the moment of triggering the logic flip as the critical boundary value U. crit The critical boundary value U crit The static DC operating point (static reference value) is marked by the system for subsequent closed-loop testing. If the scan process traverses to U... end If the judgment condition is still not triggered, the system will output an exception message indicating that the boundary was not found and suspend the current test process for manual inspection.

[0100] After locking the critical boundary value in step S30, the test process enters the dynamic stability verification stage, which mainly involves constructing a "stress field" around the critical operating point. A composite excitation signal is generated by superimposing a specific frequency combination of small AC signals onto the static reference value to simulate input signal jitter caused by power supply ripple, electromagnetic interference, or poor contact in actual operating conditions. This composite excitation signal forces the instantaneous voltage data of the vehicle controller to repeatedly cross both sides of the state switching judgment threshold, thereby verifying the effectiveness of its hysteresis comparison algorithm and the stability of the control loop under critical conditions.

[0101] The main control module 100 retrieves the critical boundary value U recorded in the previous stage. crit This is set as the DC bias component of the composite excitation signal. Based on this, the main control module 100 calls the waveform synthesis algorithm of the multi-dimensional disturbance generation unit to superimpose two AC components with different frequencies and orthogonal phases. The purpose of using orthogonal sine wave superposition is to eliminate the test blind zone: if only a single-frequency sine wave is used, when this single frequency is an integer multiple of the sampling frequency of the vehicle controller, a "flicker effect" may occur where each sampling point falls on the zero-crossing point or peak point of the waveform, leading to test failure. The orthogonal dual-frequency signal ensures that the input waveform exhibits non-periodic ergodic characteristics in the time domain.

[0102] Composite excitation signal u in The generation of (t) follows the following time-domain synthesis formula:

[0103] u in (t)=U crit +A amp sin(2πf1t)+B amp cos(2πf2t+φ rand );

[0104] In the formula:

[0105] u in (t) represents the instantaneous value of the composite excitation signal input to the vehicle controller at time t, which is generated by the main control module 100 in the form of a high-precision floating-point sequence;

[0106] Ucrit This represents the critical boundary value, which serves as the static DC operating point.

[0107] A amp and B amp This represents the amplitude of two orthogonal perturbation components. A amp and B amp The value of A must satisfy: amp B amp >Δ ADC , where Δ ADC This is the smallest quantization unit (1 LSB) for the analog-to-digital converter of the vehicle controller. The amplitude is typically set to U. crit The range is 0.5% to 1.5% to ensure that the disturbance amplitude can be recognized by the vehicle controller without deviating significantly from the current operating range;

[0108] f1 and f2 represent A respectively amp and B amp The frequencies of f1 and f2 must adhere to two constraints: first, f1 and f2 must be coprime to avoid low-frequency beats; second, f1 and f2 must be lower than the cutoff frequency of the low-pass filter at the vehicle controller input (typically 100Hz to 200Hz) and higher than the closed-loop bandwidth of the control algorithm. Typical settings are: f1 = 37Hz, f2 = 73Hz;

[0109] φ rand To represent a random initial phase angle, we introduce a random initial phase angle φ. rand This is to ensure that the initial disturbance state is different in each round of repeated testing, thereby more comprehensively evaluating the robustness of the vehicle controller under different initial conditions. At the beginning of each test, φ rand A value in the range [0, 2π] is generated by the pseudo-random number generator in the main control module 100.

[0110] To ensure that the physical output signal is not distorted, the update rate of the digital-to-analog converter is set to more than 10 times the highest disturbance frequency (i.e., >1kHz). The linear power amplifier circuit is responsible for impedance matching and signal driving, and its -3dB bandwidth must cover the range from DC to 1kHz, and it must have low output impedance characteristics (<0.1Ω).

[0111] Finally, the amplified composite excitation signal is applied to the monitoring port of the vehicle controller. At this time, the input of the vehicle controller will receive a continuous signal fluctuating around the critical boundary value. This continuous signal will continuously excite the control logic to perform high-frequency state switching around the "state switching determination threshold".

[0112] For step S40, under the condition that the vehicle controller is excited by orthogonal dynamic disturbance signals and generates high-frequency logic switching action, this embodiment adopts an effective thermal stress calculation method based on oscillation weighted integral: that is, by introducing dynamic weighting coefficients into the time domain integral, the switching loss caused by high-frequency switching action and the skin effect loss generated by high-frequency harmonics are equivalently converted into the increment of DC thermal power.

[0113] The data acquisition module 400 is equipped with a wideband high-speed analog-to-digital converter, responsible for synchronously sampling the waveforms of voltage and current data across the load under test. To fully reconstruct the transient edge characteristics of the power device during rapid switching, the sampling frequency f... sample The setting must satisfy the engineering sampling theorem. Specifically, f sample Set it to more than 20 times the PWM carrier frequency of the vehicle controller, or set it to a fixed value of 200kHz to 500kHz, to ensure that at least 5-10 valid data points are collected on each switching edge.

[0114] The instantaneous voltage data v collected load (τ) and instantaneous current data i load (τ) is transferred to the main control module 100 for real-time processing via direct memory access (DMA).

[0115] The main control module 100 sets a sliding time window of length W, which defines the integration interval for the conversion of instantaneous power into thermal power. The window length W is determined based on the thermal diffusion time constant of the measured load, typically ranging from 10 ms to 50 ms. This window length smooths out random Gaussian white noise during the measurement process while preserving the power fluctuation trend caused by logic oscillations. Within each sliding time window, the system executes the effective thermal stress power P. stress Integral operation of (t).

[0116] The calculation formula is as follows:

[0117] ;

[0118] In the formula, P stress (t) represents the effective thermal stress power calculated at time t, which represents the equivalent heat density applied to the load under test; W represents the window length of the sliding time window; τ is the integration variable, representing any time within the sliding time window; v load (τ) represents the instantaneous voltage data across the measured load; i load (τ) represents the instantaneous current flowing through the measured load; dτ represents the differential increment of the integral variable τ; K osc(τ) represents the oscillation weighting factor, which is a dimensionless dynamic coefficient used to quantify the additional heat loss caused by high-frequency switching action (such as the heat conducted to the load by MOSFET switching loss, the skin effect loss of the coil winding, etc.).

[0119] Oscillation weighting factor K osc The value of (τ) is not a fixed constant, but is determined in real time based on the rate of change of voltage data. The main control module 100 pairs instantaneous voltage data v load Perform first-order differential calculation to obtain the voltage change rate |dv load / dt|. The decision logic is as follows:

[0120] ;

[0121] In the formula:

[0122] K th This represents the voltage change rate threshold, measured in volts per microsecond (V / μs). The physical significance of this threshold lies in defining the linear amplification region and saturation region of the power device. Its value is determined based on the hardware characteristics of the vehicle controller's drive circuit, specifically set to 30%–50% of the nominal slew rate specified in the datasheet of the drive device (such as a MOSFET). For example, if the measured rise slope of the drive signal is 5V / μs, then K... th Set to 1.5V / μs to 2.5V / μs;

[0123] γ represents the high-frequency loss factor, ranging from 1.05 to 1.30. This factor is the AC resistance R. ac With DC resistance R dc Engineering approximations for the ratio. For resistive loads such as PTC heaters, γ is set to 1.05–1.10 due to the weak skin effect; for inductive loads such as motor windings or solenoid valves, γ is set to 1.20–1.30 considering the significant increase in eddy current losses and hysteresis losses at high frequencies.

[0124] Through the above mechanism, when the control logic of the vehicle controller oscillates at high frequency near the critical boundary value, the voltage change rate |dv load / dt| will frequently exceed the threshold K th This causes the energy weight during the integration process to be amplified by the high-frequency loss coefficient γ. Therefore, the calculated P... stress (t) will be higher than the conventional power calculated by multiplying voltage and current alone, thus numerically reflecting the additional heat accumulation effect caused by high-frequency oscillations on the physical load.

[0125] For step S50, based on the effective thermal stress power calculated in step S40, the main control module 100 enters the dynamic reconstruction calculation stage of the virtual load impedance. This embodiment uses numerical calculation to simulate in real time the physical process of impedance drift of physical devices due to temperature rise. Through this step, the constant resistance characteristic of the electronic load is corrected to a dynamic impedance characteristic that changes with the accumulation of thermal energy, thereby introducing a real physical coupling effect into the drive circuit of the vehicle controller.

[0126] The main control module 100 first performs iterative calculations of the virtual internal temperature. Since the thermodynamic process is continuous in time, the system uses discrete difference equations to solve the thermal equilibrium differential model in real time. The main control module 100 reads the effective thermal stress power P of the current step cycle. stress (k), and combined with the virtual internal temperature of the previous step cycle, calculate the virtual internal temperature at the current moment.

[0127] To ensure the convergence of the numerical solution, a first-order forward Euler method is used for discretization. The temperature iteration formula is defined as follows:

[0128] ;

[0129] In the formula, T virt (k) represents the virtual internal temperature at the current k-th calculation step; T virt (k-1) represents the virtual internal temperature at the previous calculation step (the (k-1)th calculation step); h represents the discrete calculation step size. To ensure the discrete model accurately reproduces the continuous thermophysical process, the value of the discrete calculation step size h must satisfy the sampling theorem constraint, i.e., h must be less than one-tenth of the system's thermal time constant τ (h ≤ 0.1R). th C th h is usually set to 10ms; C th and R th These are the preset virtual heat capacity and virtual thermal resistance, respectively; P stress (k) represents the effective thermal stress power input at the current moment, which includes the additional thermal contribution from high-frequency oscillations.

[0130] Obtain the virtual internal temperature T at the current moment. virt (k) After that, the main control module 100 calculates the target equivalent impedance value based on the physical properties of the load material under test. This embodiment provides two impedance mapping modes for different types of loads under test:

[0131] Linear temperature drift mode (applicable to motor coils and solenoid valves) uses linear equations to describe impedance changes:

[0132] Z target (k)=Z base [1+α lin (Tvirt (k)-T ref )];

[0133] In the formula, Z target (k) represents the target equivalent impedance value at the k-th calculation step; k represents the index of the discrete-time series; Z base This indicates the nominal impedance value of the load under test at the reference temperature; T ref Indicates the measured nominal impedance value Z base The corresponding reference temperature; α lin It is a constant type of temperature coefficient of resistance (e.g., 0.00393 / K for copper wire).

[0134] Nonlinear mapping mode (applicable to PTC heaters and NTC sensors) uses lookup tables or higher-order polynomials to describe impedance changes:

[0135] Z target (k)=Map R-T (T virt (k));

[0136] In the formula, Map R-T The table is a pre-stored impedance-temperature characteristic curve data sheet, which is generated by importing RT characteristic curves from component datasheets or laboratory measured data. It can accurately reproduce the steep impedance rise characteristic of PTC materials near the Curie temperature.

[0137] The calculated target equivalent impedance value Z target (k) Subsequently, a safety limiting process is applied. The limiting logic ensures that the simulated impedance value remains within the normal operating range of the vehicle controller, preventing false triggering of "open circuit" or "short circuit" fault diagnosis logic. The final output impedance command Z... cmd (k) is defined as:

[0138] Z cmd (k)=min(max(Z target (k),Z min ),Z max );

[0139] In the formula, Z min Set to 110% of the vehicle controller short-circuit protection threshold (e.g., 0.5Ω), Z max Set to 90% of the open-circuit protection threshold (e.g., 100Ω).

[0140] The verified impedance command Z cmd(k) The impedance command is sent to the load simulation module 500 via a high-speed communication interface (such as EtherCAT or CANFD). The load simulation module 500 adjusts the equivalent resistance of the power stage circuit in real time according to the impedance command, so that the load characteristics of the vehicle controller drive end dynamically evolve with the change of virtual internal temperature, thus completing closed-loop control.

[0141] See attached document Figure 3 For step S60, after the virtual load impedance dynamic reconstruction loop has been running for a preset observation time window, the main control module 100 executes the convergence determination program. The physical basis of the convergence determination program is to evaluate the dual-domain stability of the coupled system: on the one hand, it evaluates whether the macroscopic thermodynamic state tends to equilibrium to prevent thermal runaway; on the other hand, it evaluates whether the microscopic control logic is in a deterministic state to prevent high-frequency invalid oscillations.

[0142] The main control module 100 extracts the virtual internal temperature sequence T within the observation time window from memory. virt and the control state sequence S of the vehicle controller ctrl The length T of the observation time window obs Set the observation time window to 5 to 10 times the thermal time constant τ of the load under test (for example, for motor systems with slow thermal response, set the window length to 30 to 60 seconds) to ensure that the system has enough time to complete the transient response and exhibit its steady-state trend.

[0143] For the virtual internal temperature sequence, the main control module 100 uses a sliding window extreme value search algorithm to identify the local peaks and valleys of the waveform, thereby defining several continuous oscillation periods. Subsequently, the oscillation decay rate is calculated, which reflects the system's ability to dampen and suppress disturbances.

[0144] The formula for calculating the oscillation decay rate is defined as follows:

[0145] λ osc =(T max (n)-T min (n)) / (T max (n-1)-T min (n-1));

[0146] In the formula, λ osc λ represents the oscillation decay rate. osc The value is dimensionless; n represents the nth complete oscillation period identified within the observation window; T max (n) and T min (n) represent the local peak and local trough values ​​of the waveform of the virtual internal temperature sequence within the nth oscillation period, respectively; T max (n-1) and T min(n-1) represent the local peak and local valley values ​​of the waveform of the virtual internal temperature sequence within the (n-1)th oscillation period, respectively.

[0147] Based on the calculated λ osc The main control module 100 performs a convergence determination:

[0148] If λ osc <1.0 indicates that the amplitude of the subsequent period is smaller than that of the preceding period, the system has positive damping characteristics, and is judged to be thermally convergent;

[0149] If λ osc A value ≥1.0 indicates that the amplitude remains constant or continues to increase, and the system is in a state of constant amplitude oscillation or divergence, which is determined to be thermal runaway.

[0150] In engineering applications, to ensure the system has sufficient stability margin, the pass / fail threshold is set to 0.85, i.e., λ. osc When the value is less than 0.85, it indicates that the system oscillation is converging and has an engineering-acceptable stability margin. This qualification threshold requires that the energy of each oscillation cycle decays by at least 15%.

[0151] In addition to thermodynamic stability, the main control module 100 synchronously calculates the state flip density to quantify the jitter of the control logic, thereby capturing the high-frequency switching behavior of the actuator and preventing relay contact burn-out or power tube overheating.

[0152] The formula for calculating state-flip density is defined as follows:

[0153] ;

[0154] In the formula, D trans T represents the state-flipping density; obs S is the window duration for the observation time window. ctrl (t) represents the control state value of the vehicle controller at time t, which is quantized into integer logic levels (e.g., 0 for open state, 1 for closed state; or discrete step values ​​of PWM duty cycle); t0 represents the start time of the observation time window; t0+T obs This indicates the end time of the observation time window; Δt is the sampling interval, and the value of Δt must be less than the minimum control cycle of the vehicle controller (e.g., 5ms). This is an indicator function; the function outputs 1 when the logical condition within the parentheses is true (i.e., the current state is not equal to the previous state), and 0 otherwise. ctrl (t-Δt) represents the control state value of the vehicle controller at the previous sampling time (i.e., time t minus the sampling interval Δt).

[0155] The main control module 100 will calculate the D transWith the preset logic stability threshold D th Perform a comparison. Logical stability threshold D th The setting is determined based on the physical switching life and thermal characteristics of the actuator:

[0156] For mechanical contact loads (such as relays), D th The frequency is set to 0.1Hz (meaning that the average number of actions allowed is no more than once every 10 seconds) to prevent mechanical fatigue of the contacts.

[0157] For semiconductor power devices (such as MOSFETs / IGBTs), D th The frequency limit is set to 5Hz to 10Hz. This frequency limit is determined based on the dynamic switching loss curve of the device to ensure that the junction temperature of the device does not exceed the limit when heat dissipation conditions are limited.

[0158] If the system simultaneously satisfies λ osc ≤0.85 and D trans ≤D th If the main control module 100 determines that the test has passed, it will generate a test report containing the steady-state temperature value and convergence time; otherwise, it will determine that the system has a risk of coupled oscillation and the test will automatically terminate.

[0159] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A test method for an electric vehicle controller, characterized in that, Includes the following steps: Initialize the test environment and establish an electrothermal impedance hysteresis model for the physical characteristics of the load under test; A continuously changing excitation signal is input to the vehicle controller. By monitoring the steady-state switching action of the output signal of the vehicle controller, the critical boundary value that triggers the logic flip is captured and recorded. Using the critical boundary value as a static reference value, a dynamic disturbance signal containing orthogonal components is generated and superimposed on the static reference value to form a composite excitation signal, which is then input to the vehicle controller so that the vehicle controller operates in a critical state. Collect voltage and current data from the vehicle controller and calculate the effective thermal stress power within the sliding time window based on the voltage change rate. Run the electrothermal impedance hysteresis model, calculate the virtual internal temperature based on the effective thermal stress power, calculate the target equivalent impedance value based on the virtual internal temperature, and simulate the target equivalent impedance value by adjusting the physical input impedance; The stability of the vehicle controller's control logic is determined by continuously monitoring the convergence trend of the virtual internal temperature and the switching frequency of the control state. The establishment of the electrothermal impedance hysteresis model for the physical characteristics of the load under test specifically includes: Retrieve the pre-stored configuration file, extract the virtual thermal capacity, virtual thermal resistance, nominal impedance value and resistance temperature coefficient, and define the thermal time constant based on the product of virtual thermal capacity and virtual thermal resistance, and limit the thermal time constant to be greater than the response time of the load simulation hardware. A thermal equilibrium differential equation is constructed, which takes the effective thermal stress power as the input variable and simulates the thermal inertial physical process through integral operation, transforming the transient power fluctuation into a smooth convergence trend of virtual internal temperature. An impedance mapping equation is constructed, which uses the calculated virtual internal temperature as the independent variable and combines the resistance temperature coefficient and the nominal impedance value to derive the target equivalent impedance value that dynamically drifts with temperature. The electrothermal impedance hysteresis model includes a thermal balance differential equation and an impedance mapping equation.

2. The test method for the electric vehicle controller according to claim 1, characterized in that, The process of capturing and recording the critical boundary value that triggers the logic flip specifically includes: A discrete excitation sequence is constructed in the time domain in a step-like increasing or decreasing manner. The excitation signal value of each step cycle is generated based on the excitation signal value of the previous moment superimposed with a preset scanning step size, and the scanning step size is set to be less than the minimum detection accuracy of the vehicle controller. The normalized vehicle controller output signal is substituted into the state difference function for real-time calculation. The change amplitude of the output state is quantified by calculating the difference between adjacent steady-state sampling points. Once the calculated result of the state difference function is detected to exceed the preset state switching judgment threshold, a logic flip is determined, and the current excitation signal value is immediately locked and extracted, which is defined as the static DC operating point for subsequent closed-loop testing. The current excitation signal value is the critical boundary value.

3. The test method for the electric vehicle controller according to claim 1, characterized in that, The process of generating a dynamic disturbance signal containing orthogonal components and superimposing it onto a static reference value specifically includes: Two AC components that are coprime and have frequencies higher than the closed-loop bandwidth of the control algorithm are selected, and a digital waveform of the dynamic disturbance signal is constructed using a waveform synthesis algorithm under the condition of phase orthogonality. A random initial phase angle is generated by a pseudo-random number generator and added to the digital waveform of the dynamic perturbation signal to construct a dynamic perturbation sequence with differentiated initial states. The dynamic perturbation sequence is superimposed with the static reference value to obtain a digital sequence of composite excitation signal; The digital sequence of the composite excitation signal is converted into a physical electrical signal through a digital-to-analog converter and power amplification stage. A stress field that continuously fluctuates around the state switching judgment threshold is constructed at the input of the vehicle controller to stimulate the high-frequency switching behavior of the control logic.

4. The test method for the electric vehicle controller according to claim 1, characterized in that, The calculation of the effective thermal stress power within the sliding time window based on the voltage change rate specifically includes: The voltage data of the measured load is subjected to first-order differential operation to obtain the voltage change rate, and the voltage change rate is compared with the voltage change rate threshold, which is determined based on the boundary between the linear amplification region and the saturation region of the power device. Based on the comparison results, an oscillation weighting factor is generated in real time: when the voltage change rate exceeds the voltage change rate threshold, the circuit is determined to be in a high-frequency switching state, and the preset high-frequency loss coefficient is retrieved as the oscillation weighting factor; otherwise, the oscillation weighting factor is set to a unit value. Within a set sliding time window, the product of voltage and current data is weighted and integrated using the oscillation weighting factor, thereby converting high-frequency switching losses and skin effect losses into an increase in effective thermal stress power.

5. The test method for the electric vehicle controller according to claim 1, characterized in that, The process of running the electrothermal impedance hysteresis model, calculating the virtual internal temperature based on the effective thermal stress power, and calculating the target equivalent impedance value based on the virtual internal temperature, specifically includes: The thermal equilibrium differential equation is invoked and solved in real time using the discrete difference method. The virtual internal temperature of the previous calculation step is used as the state variable, and the currently calculated effective thermal stress power is used as the input variable. The virtual internal temperature at the current moment is calculated through numerical integration iteration. Substituting the virtual internal temperature into the impedance mapping equation, if the load under test is linear, the impedance drift is calculated through the linear equation, and the target equivalent impedance value is obtained based on the impedance drift and the nominal impedance value. If the load under test is nonlinear, the corresponding target equivalent impedance value is obtained by calling the impedance-temperature characteristic curve data table through the lookup table method. The calculated target equivalent impedance value is subjected to safety limiting processing to obtain an impedance command. The final output impedance command is limited to the safe range defined by the short-circuit protection threshold and the open-circuit protection threshold of the vehicle controller.

6. The test method for the electric vehicle controller according to claim 1, characterized in that, The determination of the stability of the vehicle controller's control logic further includes: The virtual internal temperature sequence within the observation time window is extracted. The local peaks and valleys of the waveform of the virtual internal temperature sequence are identified by extreme value search to define the continuous oscillation period in the time domain. The amplitude ratio of adjacent oscillation periods is calculated to generate the oscillation decay rate. The calculated oscillation decay rate is compared with the pass / fail threshold: if the oscillation decay rate is less than the pass / fail threshold, it is determined to be thermal convergence; if the oscillation decay rate is greater than or equal to the pass / fail threshold, it is determined that the vehicle controller has a risk of thermal runaway and the test is terminated.

7. The test method for the electric vehicle controller according to claim 6, characterized in that, The determination of the stability of the vehicle controller's control logic also includes dual verification based on state transition density: The total number of logic level switches in the output signal of the vehicle controller within the observation time window is counted and divided by the duration of the observation time window to generate the state flip density. The state flip density is compared with the logic stability threshold determined based on the physical lifetime of the execution device; The vehicle controller is confirmed to have reached stability at both the thermodynamic macroscopic level and the logical microscopic level only when the determined oscillation decay rate is qualified and the state reversal density is simultaneously lower than the logic stability threshold. The vehicle controller is then deemed to have passed the test.

8. The test method for the electric vehicle controller according to claim 3, characterized in that, The setting of the AC component amplitude in generating the dynamic disturbance signal containing orthogonal components follows the following constraints: An upper limit constraint is added to the frequency setting of the AC component, requiring that the frequency be lower than the cutoff frequency of the low-pass filter at the input of the vehicle controller. In the digital-to-analog conversion stage, the update rate of the digital-to-analog converter is set to be greater than ten times the highest frequency of the AC components; In the signal driving stage, the bandwidth of the linear power amplifier circuit is configured to cover the range from DC to 1 kHz, and it has low output impedance characteristics.

9. A test system for an electric vehicle controller, applied to the test method for the electric vehicle controller according to any one of claims 1-8, characterized in that, include: The main control module is used to initialize the test environment and establish an electrothermal impedance hysteresis model, receive voltage data fed back by the data acquisition module, calculate the voltage change rate based on the voltage data, calculate the effective thermal stress power, run the electrothermal impedance hysteresis model to solve the virtual internal temperature and target equivalent impedance value in real time, generate and send the corresponding impedance command to the load simulation module, and determine the control logic stability of the electric vehicle controller based on the convergence trend of the virtual internal temperature and the switching frequency of the control state. The simulation module is used to run the vehicle dynamics model and interact with the vehicle controller through the vehicle communication bus. It sends wake-up messages and simulated vehicle speed signals to the vehicle controller, so that the functional loop of the vehicle controller under test is in an active standby state. The excitation injection module is used to respond to the instructions of the main control module. During the stage of acquiring the critical boundary value, it outputs a discrete excitation sequence that changes in a step-like manner. During the closed-loop test stage, it generates a composite excitation signal containing orthogonal components and superimposed on the static reference value to drive the vehicle controller to operate in the critical state. The data acquisition module is used to monitor the output signal status of the vehicle controller in real time to identify logic flipping actions, and synchronously acquire the voltage and current data applied by the vehicle controller to the two ends of the load simulation module and transmit them to the main control module. The load simulation module receives impedance commands sent by the main control module and dynamically adjusts its own physical input impedance to simulate the target equivalent impedance value calculated from the simulation.

Images