A method and device for overheat protection of an electric power steering gear for a heavy commercial vehicle

Abstract

This invention provides a method and device for overheat protection of electric power steering (EPS) in heavy-duty commercial vehicles. The method includes the following sub-steps: S1: Acquiring and filtering the input signal; S2: Estimating power consumption and updating the thermal model based on the input signal; S3: Predicting the winding temperature and power module temperature in parallel through two dual time windows (fast window and slow window); S4: Achieving thermal constraint closed-loop and solving; S5: Calibration and protection. The device includes a temperature sensor, an EPS control system, a constraint solver, and a power steering / speed control mapping module. This invention captures the heating pattern of the EPS in commercial vehicles and adopts a dual-time-window parallel prediction mechanism to achieve full-condition coverage, avoiding the adaptation limitations of a single prediction mode. Through smooth fusion constraints, multi-level buffer limits, thermal margin collaborative adaptation, and graded protection, the protection mechanism is gradually activated to avoid sudden changes in power steering assistance and ensure driving continuity and safety.

Description

Technical Field

[0001] This invention relates to the field of electric power steering control for commercial vehicles, and more particularly to a method and device for overheat protection of electric power steering systems in heavy-duty commercial vehicles. Background Technology

[0002] Electro-hydraulic power steering (EHPS) is currently the mainstream configuration for commercial vehicles. It reduces the driver's operating effort by introducing hydraulic devices or electronic control units, and provides assistance for steering operations during commercial vehicle operation to meet basic steering needs. However, as the automotive industry rapidly develops towards energy conservation and intelligence, electric power steering (EPS) is gradually becoming the future technology development trend of commercial vehicle steering systems due to its potential advantages in energy conservation and response speed, and is beginning to enter the market.

[0003] Compared to passenger car EPS, commercial vehicle EPS faces more stringent operating conditions: commercial vehicles have larger loads and more complex driving conditions, causing their motors to heat up faster and for longer periods. Especially in low-speed, frequent turning or long-term, high-load operating scenarios, the high current operation of the motor generates a large amount of heat, which can easily lead to damage to the motor and controller. This problem has become the core bottleneck restricting the large-scale application of commercial vehicle EPS. In addition to improvements through new materials and processes, optimizing control algorithms has become a key direction for alleviating the problem of motor overheating.

[0004] Existing overheat protection technologies have the following drawbacks:

[0005] 1. Poor adaptability: Existing overheat protection technology is only applicable to conventional steering systems. Under the heavy load conditions of commercial vehicles, it is difficult to balance the needs of steering performance and protection of electronic control devices. It does not fully consider the complex operating conditions of commercial vehicles and cannot effectively solve the overheat protection problem of EPS in commercial vehicles.

[0006] 2. Conventional overheat protection methods affect the driving experience and can easily cause safety hazards: They often use a fixed limit setting. When the system temperature reaches the limit, the protection mechanism will be activated suddenly, causing a sudden change in steering assist. Commercial vehicles are prone to triggering the limit condition in actual driving. This sudden change will cause a sudden change in steering feel, disrupt the continuity of driving operation, reduce comfort, and may also bring safety hazards. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and device for overheat protection of electric power steering systems in heavy-duty commercial vehicles.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] An overheat protection device for an electric power steering system in heavy-duty commercial vehicles includes a temperature sensor, an EPS control system, a constraint solver, and a power steering / speed regulation mapping module.

[0010] The temperature sensor is used to collect ambient temperature and power module temperature from the vehicle's signals;

[0011] The EPS control system is used to collect input signals, perform double buffering, preprocessing and filtering on the collected input signals, perform drive thermal model calculations, trigger dual-window prediction, and dynamically adjust the current / speed limits and assist torque commands based on the constraint solution results.

[0012] The constraint solver is used to obtain the total allowable power and convert it into the original value of the current / speed limit. It is then processed by smoothing and slope limiting to become the current / speed limit value that can be directly executed, and provides constraints such as thermal margin for the generation of the assist torque command.

[0013] The power assist / speed regulation mapping module is used to generate target angular velocity, torque conversion commands, etc.

[0014] A method for overheat protection of an electric power steering system in a heavy-duty commercial vehicle includes the following sub-steps:

[0015] S1: Acquire the input signal and perform filtering;

[0016] S11: Acquire input signal;

[0017] Input signals are periodically collected according to a preset control cycle. The input signals include q-axis current signal, bus voltage, motor speed, torque command value, ambient temperature, power module temperature, and steering wheel rotation pulse signal from the vehicle signals.

[0018] Specifically, the analog-to-digital converter (ADC) of the electric power steering system (EPS) performs hardware-triggered sampling at the midpoint of the pulse width modulation (PWM) signal, and periodically collects the q-axis current signal, bus voltage, and motor speed of the EPS according to a preset control cycle.

[0019] The embedded timer NHET of the electric power steering system EPS periodically acquires the duty cycle signal of the torque command according to a preset control cycle, and converts the duty cycle signal of the torque command into the torque command value through a preset mapping algorithm.

[0020] The power module temperature is collected by a temperature sensor installed in the driver board, and the ambient temperature in the vehicle signal is collected by an ambient temperature sensor.

[0021] The steering wheel angle sensor periodically collects pulse signals of steering wheel rotation according to a preset control cycle;

[0022] S12: Double-buffered transfer of the input signal;

[0023] The hardware transmission unit (HTU) of the electric power steering (EPS) system pre-initializes two independent buffers, the format and length of which are matched with the input signal; the two buffers are denoted as buffer A and buffer B, respectively.

[0024] After the input signal is acquired, the hardware transmission unit (HTU) triggers signal transmission and writes the input signal into buffer A. At this time, buffer B is in a ready state, and the CPU of the electric power steering system (EPS) synchronously processes the historical data stored in buffer B. When buffer A is full, the hardware transmission unit (HTU) switches to buffer B to transmit the input signal, while the CPU switches to process the input signal in buffer A. The two sets of buffers alternately complete the transfer of the input signal.

[0025] S13: Perform preprocessing and filtering on the input signal after transportation;

[0026] The input signals after being moved between the two sets of buffers are preprocessed, including format unification, timing alignment, and removal of abnormal data.

[0027] Specifically, format standardization: the input signals in buffer A and buffer B are uniformly converted into a standardized format;

[0028] Timing alignment: Based on the sampling timing of the midpoint of the pulse width modulation (PWM) signal, the input signal is time-aligned;

[0029] Remove abnormal data: Remove input signals that exceed the corresponding preset threshold range;

[0030] The preprocessed input signal is then filtered.

[0031] Specifically, for input signals with high dynamic changes such as q-axis current signal and motor speed, a first-order low-pass filtering algorithm is used to filter out high-frequency sampling noise by setting a cutoff frequency; for input signals with low frequency and slow changes such as ambient temperature and bus voltage in the whole vehicle signal, a moving average filtering algorithm is used to smooth the instantaneous fluctuations of the input signal by averaging multiple consecutive sampling points.

[0032] S2: Power consumption estimation and thermal model update based on input signal;

[0033] S21: Based on the acquired input signal and preset calibration parameters, calculate the power consumption.

[0034] The preset calibration parameters include the motor stator winding at a universal reference temperature. The winding resistance below Temperature coefficient of resistance Iron loss coefficient and Switching loss coefficient Conduction loss coefficient Iron loss injection ratio in windings Winding heat capacity Power module thermal capacity Thermal resistance from winding to environment Thermal resistance from power module to environment Thermal resistance of coupling between winding and power module Switching frequency factor Temperature compensation coefficient , fast loop step length ;

[0035] in The coefficient for the first term of the speed is denoted as , representing the portion of iron loss that varies linearly with the motor speed. The coefficient for the quadratic term of the speed range represents the portion of iron loss that varies with the square of the motor speed; in the low-speed region... The effect is more direct, especially in the higher speed range. The item has a more significant impact on the growth of iron loss; the two annotations have the same meaning, both used to calculate the speed limit value based on the iron loss model.

[0036] Calculate the current winding resistance : ;

[0037] Here The initial value is the ambient temperature in the vehicle signal;

[0038] Calculate copper loss, iron loss, and drive loss based on the current winding resistance;

[0039] Copper loss is denoted as , ;

[0040] in, The effective value of the q-axis current obtained by converting the q-axis current signal acquired in step S1 using the root mean square (RMS) and other effective value algorithms. , as the target current;

[0041] Iron loss is recorded as , ;

[0042] in, The motor speed collected in step S1;

[0043] Let the driving loss be ;

[0044] ;

[0045] in, The ambient temperature is the signal from the vehicle collected in step S1.

[0046] The total injected power of the winding is calculated using the following formula. Power module injected power consumption :

[0047] Total power injection into windings ;

[0048] Power module injected power consumption ;

[0049] S22: Iteratively update winding temperature and power module temperature using a two-node thermal model;

[0050] Update the winding temperature using the following formula. :

[0051] ;

[0052] in, To control the cycle number, For the winding temperature of the previous control cycle, The winding temperature is updated after this control cycle; the initial value is set manually. Tamb represents the ambient temperature in the vehicle signals. The power module temperature and initial value for the previous control cycle are set manually. The total power injection into the windings in the previous control cycle is initialized manually; the power module temperature is updated using the following formula:

[0053] ;

[0054] in, This is the power module temperature after the update in this control cycle.

[0055] S3: Predicts winding temperature and power module temperature in parallel through two dual time windows, fast window and slow window, to capture transient temperature rise peak and long-term thermal saturation trend;

[0056] Preset steering wheel angular velocity threshold The update intervals corresponding to fast and slow windows ;

[0057] Based on the mapping relationship between the pulse signal of steering wheel rotation and the steering wheel deflection angle, the steering wheel deflection angle is obtained. The difference between the steering wheel deflection angles corresponding to two consecutive control cycles is divided by the fast loop step size Δt to obtain the steering wheel angular velocity.

[0058] Relative steering wheel angular velocity to steering wheel angular velocity threshold Compare; if it is greater than or equal to the steering wheel angular velocity threshold... Then, execute step S31 according to the update interval corresponding to the fast window; if it is less than the steering wheel angular velocity threshold. Then, step S32 is executed according to the update interval corresponding to the slow window;

[0059] S31: Reset the fast window prediction sequence;

[0060] S311: Perform a one-step prediction;

[0061] The Electric Power Steering (EPS) system integrates the pulse signal from the steering wheel rotation and the torque command value obtained in step S11 to obtain the actual assist demand. Based on the preset mapping relationship between assist demand and q-axis current, it obtains the q-axis current actually requested by the driver, denoted as... , as the actual current;

[0062] Furthermore, the winding injection power based on the driver's request is calculated using the following formula. Power injected by the power module based on driver request ;

[0063] ;

[0064] The Pcu= ; ;

[0065] ;

[0066] The ;

[0067] And calculate the predicted temperature for the next cycle;

[0068] ;

[0069] in, For the predicted winding temperature, the The updated winding temperature after step S2;

[0070]

[0071] in, The predicted temperature value of the power module, the For the power module temperature updated in step S2, Inject power into the windings based on the driver's request;

[0072] S312: Perform multi-step prediction;

[0073] The following formula is used for multi-step prediction to predict the temperature trajectory within a fast window, thus obtaining the future... The predicted temperature value for the step;

[0074]

[0075] The The number of iterations within the set window; For the first After one control cycle, in the future The predicted temperature value, including future... Predicted winding temperature of step ,future The predicted temperature value of the power module in step; A is the preset thermal model state matrix. This is the iteration index variable, and its value range is... B is the preset thermal model input matrix; Let the current temperature vector be... , constitute; The predicted power sequence for the next h steps is derived from... and The column vector formed;

[0076] The peak temperature of the winding is extracted from the temperature trajectory within the window using the following formula. :

[0077] ;

[0078] in, This is a function that iterates through the range and takes the maximum value among the given values; in this formula... To traverse the future Take the maximum value among all predicted winding temperatures for each step;

[0079] The peak temperature of the power module is extracted from the fast-window prediction using the following formula. :

[0080] ;

[0081] In this formula, To traverse the future Take the maximum value among all predicted power module temperatures for each step;

[0082] S32: Update the slow window prediction sequence;

[0083] Based on long-term statistical power consumption and equivalent thermal models, the temperature rise trend and peak temperature of the windings and power modules are predicted, including the following sub-steps:

[0084] S321: Calculate the equivalent thermal resistance and thermal time constant;

[0085] The equivalent thermal resistance of the winding is calculated using the following formulas. Equivalent thermal resistance of power module :

[0086] ;

[0087] ;

[0088] The winding thermal time constant is calculated using the following formula. Thermal time constant of power module :

[0089] ;

[0090] ;

[0091] S322: Calculate statistical power;

[0092] The statistical average of copper consumption in the current cycle is calculated using the following formula:

[0093] ;

[0094] ;

[0095] ;

[0096] in, For the set smoothing coefficient, This represents the statistical average of copper consumption in the current cycle. This is the statistical average of copper consumption from the previous cycle, with the initial value being the preset baseline copper consumption. The current cycle copper consumption is calculated in step S21;

[0097] This is the statistical average of iron consumption in the current cycle; This is the statistical average of iron consumption in the previous cycle, with the initial value being the preset baseline iron consumption. The current cycle iron loss is calculated in step S21;

[0098] This represents the statistical average of the driving losses in the current cycle. This is the statistical average of the drive loss in the previous cycle, with the initial value being the preset baseline drive loss. The driving loss calculated in step S21;

[0099] S323: Predicts the temperature saturation trend of windings and power modules;

[0100] The long-term heat injection power is calculated using the following formula:

[0101] ,

[0102] ;

[0103] in, Power injection for long-term heat transfer into windings The statistical average of copper consumption Inject the preset iron loss into the winding ratio, The statistical average value of iron loss Power injection for long-term heat dissipation in power modules The statistical average of the drive losses; , , The average value of copper loss, iron loss and drive loss in long-term statistics in step S21 is calculated by using the average value formula.

[0104] The long-term winding temperature Tw_slow(t) can be predicted using the following formula:

[0105] ;

[0106] The For the time variable within the slow window, , Given the slow window length, It is a natural constant;

[0107] The following formula is used to predict the temperature of the power module. :

[0108] ;

[0109] S324: Calculate the peak temperature within the slow window;

[0110] Iterate through the predicted temperatures at all times within the slow window length and extract the peak temperature;

[0111] ;

[0112] in, The peak temperature of the winding. For the winding temperature at any time during the slow window, To iterate through all times within the slow window length (from 0 to ... That is, to traverse the entire long-term temperature trajectory and take the maximum value of the winding temperature;

[0113] ;

[0114] in, The peak temperature of the power module. For the power module temperature at any time during the slow window, To iterate through all times within the slow window length (from 0 to ... That is, to traverse the entire long-term temperature trajectory and take the maximum value of the power module temperature.

[0115] This step introduces two independent prediction channels: a fast window and a slow window. The fast window is used to capture the transient temperature rise peak during aggressive driving (to prevent instantaneous burnout); the slow window is used to calculate the thermal saturation trend under long-term load (to prevent long-term overheating).

[0116] S4: Achieve thermal constraint closed-loop and solve the problem;

[0117] S41: Obtain the final executable allowable power;

[0118] S411: Calculate the upper limit of the derived power of the fast loop constraint;

[0119] The constraint solver calculates the upper limit of the winding temperature constraint derived power using the following formulas. Upper limit of derived power due to winding temperature rise rate constraint ;

[0120] ;

[0121]

[0122] Similarly, the upper limit of the power module's temperature-constrained derived power is calculated. Power module temperature rise rate constraint derives power upper limit value ;

[0123]

[0124] ;

[0125] ;

[0126] in, For the known winding hard limit temperature, For the set safety redundancy temperature, For the set maximum temperature rise rate of the winding, For power module hard-limit temperature, The maximum temperature rise rate of the power module is set.

[0127] S412: The allowable power is obtained by smoothly integrating temperature and temperature rise rate constraints through a soft minimum function;

[0128] ;

[0129] ;

[0130] ;

[0131] in, For the allowable power of the winding, Allow power for the power module For the total allowable power of the fast loop, The smoothing parameters are those used when the temperature constraints and temperature rise rate constraints of the same component (winding / power module) are combined.

[0132] S413: Slow Window Budgeting and Smoothing;

[0133] The constraint solver calculates the winding temperature margin using the following formula. and power module temperature margin ;

[0134] ;

[0135] ;

[0136] The temperature margin is converted into the corresponding slow window allowable power using the following formula;

[0137] ;

[0138] ;

[0139] in, For the long-term permissible power of the winding, This refers to the long-term permitted power of the power module.

[0140] The allocable budget is calculated using the following formula:

[0141] ;

[0142] ;

[0143] ;

[0144] in, For windings, an allocable budget Budget can be allocated for power modules, The original value of the slow window budget;

[0145] The total budget for the current period after smoothing is calculated using the following formula. :

[0146] ;

[0147] in, For the slow window budget after smoothing from the previous cycle, obtained through a delay, The time constant for budget smoothing;

[0148] The final executable total allowed power is obtained by fusing fast and slow windows using a soft minimum function. ;

[0149] ;

[0150] ;

[0151] in, For fast and slow window fusion power, For pre-set fast and slow window blending smoothing parameters, The known physical maximum power limit;

[0152] S42: Converts the final executable total allowable power into directly executable current / speed limits;

[0153] S421: Current-side closed-loop limit;

[0154] The final executable total allowable power is converted into the original value of the current limit using the following formula. :

[0155] ;

[0156] , ;

[0157] ;

[0158] in, Power budget for the current side;

[0159] To account for the iron loss corresponding to the angular velocity requested by the driver, Request the angular velocity for the driver. For the set switching loss coefficient, The set switching loss reference coefficient, For the bus voltage collected in step S1, The set current-side loss comprehensive coefficient, For the reference coefficient of conduction loss, It is the square root function;

[0160] S422: Speed-side closed limit;

[0161] The final executable total permissible power is converted to the original speed limit value using the following formula. ;

[0162] ;

[0163] ;

[0164] in, For speed-side power budget, To comply with the driver's request Copper loss due to shaft current variation The drive loss varies with the q-axis current requested by the driver;

[0165] S423: Convert the raw values ​​of the current / speed limit into continuously operable values;

[0166] The gradual change of the original value of the current / velocity limit is achieved by the following first-order filtering formula;

[0167] ;

[0168] ;

[0169] in, This is the current limit after smoothing for the current cycle; the initial value is set manually. This is the current limit after smoothing from the previous cycle; The smoothing time constant for the set current limit; This is the speed limit after smoothing for the current cycle; the initial value is set manually. This is the speed limit after smoothing from the previous cycle; The time constant is used to smooth out the set speed limit.

[0170] The upper limit of the rate of change of the current / speed limit is set using the following slope limiting formula:

[0171] ;

[0172] ;

[0173] in, The upper limit of the current slope is set. This is the upper limit of the set velocity slope.

[0174] S43: Generate assist torque command;

[0175] S431: Combines current / velocity limits with thermal margin to dynamically generate the final target angular velocity adapted to the thermal state;

[0176] S4311: Obtain the basic homing target angular velocity;

[0177] The base homing target angular velocity is calculated using the following formula. ;

[0178] ;

[0179] in, The set gain coefficient for the homing speed; The steering wheel deflection angle is set as positive when the steering wheel deflects to the right and negative when it deflects to the left.

[0180] S4312: Obtain the heat margin normalization coefficient;

[0181] The minimum thermal margin of the winding and power module is calculated using the following formula, and the thermal margin normalization coefficient is obtained based on the minimum thermal margin.

[0182] ;

[0183] ;

[0184] ;

[0185] in, For the minimum remaining heat margin of the winding, Minimum residual thermal margin for power modules For the heat margin normalization coefficient, For the amplitude limiting function, The lower limit of heat margin set by humans The upper limit of thermal margin is set by the user. ;

[0186] S4313: Adjust the return force and response bandwidth of the electric power steering system (EPS) based on the thermal margin normalization coefficient;

[0187] The target value for the positive return weight is calculated based on the following formula;

[0188] ;

[0189] in, To correct the target value of the weight; This is a limiting function; the output value is between 0 and 1. The set positive weight base value, The set positive heat synergy increment;

[0190] The calculated target value of the positive return weight is processed by a first-order low-pass filter using the following formula to obtain the positive return weight for the current period.

[0191] ;

[0192] in, For the current cycle to return to positive weight, To correct the weight of the previous period stored in history, The time constant for the positive weighting smoothing;

[0193] The target bandwidth value is calculated using the following formula;

[0194] ;

[0195] in, For bandwidth target value, For bandwidth lower limit, For bandwidth limit, For hyperbolic tangent function, The margin mapping smoothing coefficient;

[0196] The bandwidth coefficient for the current period is obtained by performing a first-order low-pass filter on the target bandwidth value using the following formula.

[0197] ;

[0198] in, This is the bandwidth coefficient for the current period, with an initial value set manually. This represents the bandwidth coefficient of the previous historical storage period. The bandwidth smoothing time constant;

[0199] S4314: Obtain the target angular velocity;

[0200] Let the target angular velocity be ;

[0201] ;

[0202] in, The driver requests the angular velocity by dividing the difference in steering wheel deflection angle between adjacent control cycles by the fast-loop step size. get; The target angular velocity is used as the basis for correction.

[0203] S432: Converts the final target angular velocity / current request for thermal margin adaptation into motor angular velocity commands and assist torque commands;

[0204] S4321: Obtain the target angular velocity after the speed limit is applied;

[0205] The boost / speed control mapping module calculates the target angular velocity after speed limiting using the following formula;

[0206] ;

[0207] in, The target angular velocity after speed limiting; This is a sign function; it returns 1 when the input is positive, -1 when the input is negative, and 0 when the input is 0.

[0208] S4322: Receive the angular velocity command and apply constraints;

[0209] The target angular velocity after speed limiting is smoothed by first-order filtering using the following formula to obtain the angular velocity command for the current cycle.

[0210] ;

[0211] in, The angular velocity command for the current cycle has an initial value that is manually set.

[0212] The angular velocity command stored in the previous cycle;

[0213] The slope limit is set for the rate of change of the angular velocity command in the current cycle using the following formula;

[0214] ;

[0215] in, The upper limit of the slope of the set angular velocity;

[0216] Set the angular acceleration limit using the following formula;

[0217] ;

[0218] in, The angular velocity command stored in the previous cycle, This is the upper limit of the set angular acceleration;

[0219] S4323: Obtain the bandwidth coefficient;

[0220] The current scaling factor is obtained using the following formula;

[0221] ;

[0222] in, For the set current scaling factor, For current limits, The set current protection threshold is zero.

[0223] The angular velocity scaling factor is obtained using the following formula;

[0224] ;

[0225] Where s_ω is the set angular velocity scaling factor, and ω_max is the angular velocity limit. This is the set threshold for preventing zero angular velocity;

[0226] The bandwidth factor is calculated using the following formula;

[0227] ;

[0228] in, The set bandwidth coefficient;

[0229] S4324: Converts bandwidth coefficient into current command;

[0230] The boost / speed control mapping module generates current pre-commands using the following formula. ;

[0231] ;

[0232] The current command for the current cycle is obtained by performing a first-order filter smoothing process on the current pre-command using the following formula.

[0233] ;

[0234] in, For the current cycle current command, The current command from the previous cycle has an initial value that is manually set.

[0235] The current slope amplitude of the current command is set using the following formula;

[0236] ;

[0237] in, The upper limit of the current slope is set;

[0238] S4325: Converts the current command into the actual output steering assist torque;

[0239] The current cycle assist torque is calculated using the following formula;

[0240] ;

[0241] in, To provide torque for the current cycle, The set motor torque coefficient;

[0242] The upper limit of the torque slope is obtained using the following formula;

[0243] ;

[0244] in, This represents the upper limit of the torque slope.

[0245] The torque change rate constraint for steering assist torque is applied using the following formula;

[0246] .

[0247] This step is based on the smooth constraint fusion of soft minimum functions, which is different from the traditional hard lookup table or piecewise linear constraints. It adopts a soft minimum function mathematical model to smoothly integrate temperature constraints, temperature rise rate constraints, and physical limits using mathematical methods, thus solving the problem of sudden changes in the steering feel when triggering protection in commercial vehicles. It dynamically adjusts the return speed and system bandwidth according to the thermal margin, and while protecting the motor, it actively adjusts the steering feel so that the driver can perceive that the thermal fade is gradual.

[0248] S5: Perform calibration and protection;

[0249] Based on the temperature prediction, power constraint, and torque command generation results obtained in steps S1-S4, relevant thresholds are calibrated and control parameters are dynamically adjusted.

[0250] Specifically, adjustable thermal protection warning thresholds are preset, including winding warning temperature and power module warning temperature; and safety buffer margins are configured.

[0251] Real-time monitoring of winding temperature, power module temperature and temperature rise rate. When the heat load increases (such as when the temperature approaches the warning threshold), the torque change rate is gradually reduced according to the preset grading rules: the torque change rate is maintained at a high value at room temperature, and decreases linearly or piecewise as the temperature rises, and drops to the lowest value at high temperature; and the boost gain and return slope are reduced simultaneously.

[0252] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0253] This method employs a layered RC two-node thermal model of the stator winding and power module, combined with periodic power consumption decomposition estimation and real-time thermal model updates, to capture the heating pattern of EPS in commercial vehicles. It adopts a dual-time-window parallel prediction mechanism, with a fast window adapting to short-term heavy-load scenarios with high-speed steering wheel rotation and a slow window adapting to long-term steady-state load scenarios, achieving full coverage of operating conditions and avoiding the adaptation limitations of a single prediction mode. It uses a thermal constraint closed-loop and speed regulation coordination strategy to dynamically adjust the power assist torque and return slope based on the temperature of the steering system and the current power assist demand, no longer relying on fixed parameters, and can adapt to the complex operating conditions of commercial vehicles.

[0254] By employing smooth fusion constraints, multi-level buffer limits, thermal margin collaborative adaptation, and graded protection, the protection mechanism is activated gradually to avoid sudden changes in steering assist and ensure driving continuity and safety. Attached Figure Description

[0255] Figure 1 This is a flowchart illustrating the steps of an overheat protection method for an electric power steering system in a heavy-duty commercial vehicle according to the present invention. Detailed Implementation

[0256] To provide a further understanding of the purpose, structure, features, and functions of the present invention, detailed descriptions are provided below with reference to specific embodiments.

[0257] Example 1

[0258] For low-speed, heavy-load operating conditions, the specific implementation method is as follows:

[0259] (1) Set initial operating conditions;

[0260] Ambient temperature Initial temperature of winding Initial temperature of power module ;

[0261] , , ;

[0262] , ;

[0263] , , ,

[0264] , ;

[0265] , , ; ;

[0266] , , ;

[0267] , , ; , ;

[0268] , , ; , , ; , , ;

[0269] (2) Perform input signal acquisition and filtering;

[0270] according to Fast loop periodic acquisition signal, including shaft current Bus voltage Motor speed And so on, through HTU double buffering and filtering suppression;

[0271] (3) Power consumption estimation and thermal model update;

[0272] Winding resistance calculation: ;

[0273] by , Perform power consumption decomposition:

[0274] Copper loss Iron consumption Drive loss Total power consumption ;

[0275] Fast loop constraint allowable power:

[0276] ;

[0277] ;

[0278] ;

[0279] Slow-release window budget:

[0280] ;

[0281] ;

[0282] ;

[0283] ;

[0284] Statistical injection approximation: , , ;

[0285] , ;

[0286] After smoothing (initial);

[0287] Total permissible power and limits:

[0288] ;

[0289] Current-side budget ,Pick , ;

[0290] , ;

[0291] Limit smoothing and slope Continuously available;

[0292] Speed-side budget:

[0293] ; after smoothing and Restricting the formation of executables ;

[0294] Coordination and speed / amplitude limiting:

[0295] Margin: Calculated based on the peak values ​​of the fast and slow windows. , Smooth to ;

[0296] Synthetic return angular velocity:

[0297] ;

[0298] Target angular velocity limit:

[0299] ;

[0300] Speed ​​command smoothing and constraints: First-order filtering + limit;

[0301] Current command limit: , ;

[0302] (4) Results and indicators;

[0303] After 60 seconds of continuous pullback, , Do not exceed the upper limit;

[0304] The steering wheel response time decreases by approximately 10–15%, with a delay of ≤60–80ms; there are no sudden changes or vibrations; the return to center becomes less flexible when thermal pressure increases, but the handling consistency remains acceptable; and the warning zone is clearly indicated.

[0305] Example 2

[0306] For long-term load conditions, the specific implementation method is as follows:

[0307] Signals are acquired in a timing sequence of 200µs fast loop, 2ms medium speed loop, and 20ms low speed loop, including q-axis current, bus voltage, motor speed, ambient temperature, etc. The data is processed by HTU double buffering and filtering to ensure the stability of long-term data transmission.

[0308] (1) Set initial operating conditions;

[0309] Ambient temperature Initial temperature , ;

[0310] The motor / drive and thermal parameters are the same as in Example 1; the target vehicle speed is high, and the duty cycle is high due to small-angle correction.

[0311] Constraints and windows are in the same order; Slow window is sensitive to heat debt

[0312] (2) Sampling and power consumption estimation;

[0313] , ;

[0314] (Slightly increased after the temperature rises);

[0315] ;

[0316] , ;

[0317] (3) Dual-window temperature prediction;

[0318] , , ;

[0319] Slow window response: , The temperature continues to rise, and it is predicted to approach the warning level in 10–12 minutes.

[0320] Integration of slow window budgeting and fast loop:

[0321] ;

[0322] ;

[0323] , , ;

[0324] The fast-loop Pfast is typically higher in this operating condition. It is dominated by a slow window and descends smoothly;

[0325] Limits and bandwidth coordination:

[0326] and After smoothing and slope, Limit the formation of continuous limits;

[0327] Heat margin normalization Gradually reduce, It smoothly decreased from 1.0 to the 0.8–0.85 range;

[0328] Positive weight Increase the ratio (e.g., from 0.3 to 0.45) to improve the mechanical homing ratio and reduce high-frequency energy consumption;

[0329] (4) Results and indicators;

[0330] 15 minutes later, , The hard limit was not triggered.

[0331] A decrease of approximately 12–18%, Decrease of approximately 10–20% (as the price decreases) Smoothing and speed limiting);

[0332] Good directional stability with a response delay of ≤ 60 ms; slight changes in driving perception, maintaining comfort.

[0333] The present invention has been described in the above-described embodiments; however, these embodiments are merely examples for implementing the present invention. It must be noted that the disclosed embodiments do not limit the scope of the present invention. Conversely, any modifications and refinements made without departing from the spirit and scope of the present invention are within the scope of patent protection of the present invention.

Claims

1. A method for overheat protection of an electric power steering system in a heavy-duty commercial vehicle, characterized in that: Includes the following steps: S1: Acquire the input signal and perform filtering; S11: Acquire input signal; S12: Double-buffered transfer of the input signal; S13: Perform preprocessing and filtering on the input signal after transportation; S2: Power consumption estimation and thermal model update based on input signal; S21: Based on the acquired input signal and preset calibration parameters, calculate the power consumption. S22: Iteratively update winding temperature and power module temperature using a two-node thermal model; S3: Predicts winding temperature and power module temperature in parallel through two dual time windows, fast window and slow window, to capture transient temperature rise peak and long-term thermal saturation trend; S31: Reset the fast window prediction sequence; S32: Update the slow window prediction sequence; S4: Achieve thermal constraint closed-loop and solve the problem; S41: Obtain the final executable allowable power; S42: Converts the final executable total allowable power into directly executable current / speed limits; S43: Generate assist torque command; S5: Perform calibration and protection.

2. The overheat protection method for electric power steering systems in heavy-duty commercial vehicles as described in claim 1, characterized in that: The specific details of step S1 are as follows: S11: Acquire input signal; Input signals are periodically collected according to a preset control cycle. The input signals include q-axis current signal, bus voltage, motor speed, torque command value, ambient temperature, power module temperature, and steering wheel rotation pulse signal from the vehicle signals. S12: Double-buffered transfer of the input signal; Two independent buffers are pre-initialized in the hardware transmission unit (HTU). The format and length of the buffers are matched with the input signal. The two buffers alternately complete the input signal transfer. S13: Perform preprocessing and filtering on the input signal after transportation.

3. The overheat protection method for electric power steering systems in heavy-duty commercial vehicles as described in claim 1, characterized in that: The specific details of step S2 are as follows: S21: Based on the acquired input signal and preset calibration parameters, calculate the power consumption. The preset calibration parameters include the motor stator winding at a universal reference temperature. The winding resistance below Temperature coefficient of resistance Iron loss coefficient and Switching loss coefficient Conduction loss coefficient Iron loss injection ratio in windings Winding heat capacity Power module thermal capacity Thermal resistance from winding to environment Thermal resistance from power module to environment Thermal resistance of coupling between winding and power module Switching frequency factor Temperature compensation coefficient E, fast loop step size ; in, The coefficient for the first term of the speed is denoted as , representing the portion of iron loss that varies linearly with the motor speed. The coefficient for the quadratic term of the speed is denoted as , representing the portion of iron loss that varies with the square of the motor speed. Calculate the current winding resistance using the following formula. : ； Winding temperature here The initial value is the ambient temperature in the vehicle signal; Calculate copper loss, iron loss, and drive loss based on the current winding resistance; Copper loss is denoted as , ; in, The q-axis current RMS value is obtained by converting the q-axis current signal acquired in step S1 into the q-axis current RMS value using an RMS algorithm. ; Iron loss is recorded as , ; in, The motor speed collected in step S1; Let the driving loss be ; ； in, The ambient temperature is the signal from the vehicle collected in step S1. Total power injection into windings ; Power module injected power consumption ; S22: Iteratively update winding temperature and power module temperature using a two-node thermal model; Update winding temperature : ； Where k is the control cycle number, For the winding temperature of the previous control cycle, The winding temperature is updated after this control cycle; the initial value is set manually. Tamb represents the ambient temperature in the vehicle signals. The power module temperature and initial value for the previous control cycle are set manually. The total power consumption injected into the winding in the previous control cycle is initially set manually. Update power module temperature: ； Where Tp[k+1] is the power module temperature after the current control cycle update.

4. The overheat protection method for the electric power steering system of heavy-duty commercial vehicles as described in claim 1, characterized in that: The specific details of step S3 are as follows: S31: Reset the fast window prediction sequence; S311: Perform a one-step prediction; The electric power steering (EPS) system receives the q-axis current actually requested by the driver, denoted as... Based on q-axis current Obtain the winding injection power based on the driver's request. Power injected into the power module based on driver request And calculate the predicted temperature for the next cycle; The temperature prediction values ​​include winding temperature prediction values ​​and power module temperature prediction values. ; S312: Perform multi-step prediction; Perform multi-step prediction to predict the temperature trajectory within the fast window, and obtain the predicted temperature value for the next h steps after the k-th control cycle. ; Extract the peak temperature of the winding from the temperature trajectory within the snapshot window. : ； in, This is a function that iterates through the range and takes the maximum value among the given values; in this formula... To iterate through all predicted winding temperatures for the next h steps, take the maximum value among them; Peak temperature of power module extracted from fast window prediction : ； In this formula, To iterate through all predicted power module temperatures for the next h steps, take the maximum value among them; S32: Update the slow window prediction sequence; Based on long-term statistical power consumption and equivalent thermal models, the temperature rise trend and peak temperature of the windings and power modules are predicted, including the following sub-steps: S321: Calculate the equivalent thermal resistance and thermal time constant; Let the equivalent thermal resistance of the winding be... The equivalent thermal resistance of the power module is : ； ； Denote the thermal time constant of the winding. Thermal time constant of power module : ； ； S322: Calculate statistical power; Statistical average of copper consumption in the current cycle The calculation method is as follows: ； in, For the set smoothing coefficient, This represents the statistical average of copper consumption in the current cycle. This is the statistical average of copper consumption from the previous cycle, with the initial value being the preset baseline copper consumption. The current cycle copper consumption is calculated in step S21; Similarly, the statistical average iron loss for the current cycle can be calculated. Statistical average of current cycle driving losses ; S323: Predicts the temperature saturation trend of windings and power modules; Based on the statistical averages of copper loss, iron loss, and drive energy consumption across different cycles, the statistical averages of copper loss, iron loss, and drive energy consumption are obtained using a mean formula. Based on these statistical averages, the long-term heat injection power is then calculated. ; Predicting long-term winding temperature : ； t is the time variable within the slow window, 0 <t≤Wslow、 The known slow window length is given, and e is the natural constant. Predicting power module temperature : ； S324: Calculate the peak temperature within the slow window; Iterate through the predicted temperatures at all times within the slow window length and extract the peak temperature; ； in, The peak temperature of the winding. The winding temperature at any given moment during the slow window; ； in, The peak temperature of the power module. This represents the power module temperature at any given moment during the slow window.

5. The overheat protection method for electric power steering systems in heavy-duty commercial vehicles as described in claim 1, characterized in that: The specific details of step S4 are as follows: S41: Obtain the final executable allowable power; S411: Calculate the upper limit of the derived power of the fast loop constraint; The upper limit of the derived power due to winding temperature constraint is calculated using the following formulas. Upper limit of derived power due to winding temperature rise rate constraint ; ； ； Similarly, the upper limit of the power module's temperature-constrained derived power is calculated. Power module temperature rise rate constraint derives power upper limit value ; S412: The allowable power is obtained by smoothly integrating temperature and temperature rise rate constraints through a soft minimum function; Including winding allowable power Power module allows power Total allowable power of fast loop ; S413: Slow Window Budgeting and Smoothing; The winding temperature margin is calculated using the following formula. and power module temperature margin ; ； ； The temperature margin is converted into the corresponding slow-window allowable power, including the long-term allowable power of the winding. and power module long-term permitted power ; Based on the long-term allowable power of the winding and power module long-term permitted power Obtain the winding allocable budget Power module allocable budget Original value of slow window budget ; The total budget for the current period after smoothing is calculated using the following formula. : ； in, The initial value is set manually for the slow window budget after smoothing from the previous cycle. The preset budget smoothing time constant; The final executable total allowed power is obtained by fusing fast and slow windows using a soft minimum function. ; S42: Converts the final executable total allowable power into directly executable current / speed limits; S421: Current-side closed-loop limit; The final executable total allowable power is converted into the original value of the current limit. : S422: Speed-side closed limit; Convert the final executable total allowable power to the original value of the speed limit. ; S423: Convert the raw values ​​of the current / speed limit into continuously operable values; The smoothed current limit for the current period is obtained using a first-order filtering formula. The initial value is set manually; the current cycle smoothed speed limit. The initial value is set manually; The upper limit of the rate of change of the current / velocity limit is set by the slope limiting formula: ； ； in, The upper limit of the current slope is set. The upper limit of the set velocity slope; S43: Generate assist torque command; S431: Combines current / velocity limits with thermal margin to dynamically generate the final target angular velocity adapted to the thermal state; S4311: Obtain the basic homing target angular velocity; Let the target angular velocity for base homing be... , ; in, The set gain coefficient for the homing speed; This refers to the steering wheel deflection angle; S4312: Obtain the heat margin normalization coefficient; The minimum residual heat margin of the winding is obtained by using the min function. Minimum residual heat margin of power module The heat margin normalization coefficient is obtained based on the minimum heat margin using the limiting function. ; S4313: Adjust the return force and response bandwidth of the electric power steering system (EPS) based on the thermal margin normalization coefficient; Set the target value for the positive weight. , ; in, The set positive weight base value, The set positive heat synergy increment; The target value of the positive return weight is subjected to a first-order low-pass filter to obtain the positive return weight for the current period. ; Let the target bandwidth be _____. , ； in, , For the set lower / upper bandwidth, The margin mapping smoothing coefficient; The bandwidth target value is subjected to a first-order low-pass filter to obtain the bandwidth coefficient for the current period. ; S4314: Obtain the target angular velocity; Let the target angular velocity be ; ； in, The driver requests the angular velocity by dividing the difference in steering wheel deflection angle between adjacent control cycles by the fast-loop step size. get; The target angular velocity is used as the basis for correction. S432: Converts the final target angular velocity / current request for thermal margin adaptation into motor angular velocity commands and assist torque commands; S4321: Obtain the target angular velocity after the speed limit is applied; Let the target angular velocity after the speed limit be... ; ； S4322: Receive the angular velocity command and apply constraints; The target angular velocity after speed limiting is smoothed by a first-order filter to obtain the angular velocity command for the current cycle. ; Set an upper limit for the angular rate slope of the rate of change of the angular velocity command in the current cycle. Upper limit of angular acceleration ; S4323: Obtain the bandwidth coefficient; The current scaling factor is obtained through the min function. Angular velocity linkage scaling factor ; Let the bandwidth coefficient be , ; S4324: Converts bandwidth coefficient into current command; Assume the current pre-command is ; ; The current command is smoothed by first-order filtering to obtain the current command for the current cycle. And set the upper limit of the current slope for the current command. ; S4325: Converts the current command into the actual output steering assist torque; Let the current cycle assist torque be ; ; Kt is the set motor torque coefficient; Let the upper limit of the torque slope be ; ; And the torque change rate is constrained by steering assist torque.

6. The overheat protection method for the electric power steering system of heavy-duty commercial vehicles as described in claim 1, characterized in that: In step S5, adjustable thermal protection warning thresholds are preset, including winding warning temperature and power module warning temperature; and a safety buffer margin is configured. Real-time monitoring of winding temperature, power module temperature and temperature rise rate. When the heat load increases, the torque change rate is gradually reduced according to the preset graded rules: the torque change rate is maintained at a high value at room temperature, and decreases linearly or piecewise with the temperature rise, and drops to the lowest value at high temperature. Simultaneously reduce the boost gain and the homing slope.

Images