Automatic rearview mirror heating method and device based on windscreen wiper control

Abstract

The invention discloses a rearview mirror automatic heating method and device based on windscreen wiper control, and relates to the technical field of automobile rearview mirror heating, and the method comprises the steps that rearview mirror automatic heating function setting item signals sent by an IVI and windscreen wiper working state information sent by a BCM are received; acquiring current windscreen wiper working information according to the windscreen wiper working state information; generating a heating control strategy according to the windscreen wiper working information and the automatic heating function setting item signal; according to the heating control strategy, a heating control signal is generated through the LIN bus, and the heating control signal is sent to the rearview mirror heating device, so that the rearview mirror heating device conducts heating. According to the rearview mirror automatic heating method based on windscreen wiper control, heating control is conducted according to the situation of the windscreen wiper, and automatic control over the rearview mirror is achieved. Full coverage of scenes needing rearview mirror heating such as rainy days and basement is achieved, meanwhile, based on the vehicle basic windscreen wiper control function, additional cost does not need to be increased, and large-scale application can be achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle rearview mirror heating, in particular to a rearview mirror automatic heating method and device based on wiper control. BACKGROUND

[0002] The current mainstream rearview mirror automatic heating basically determines the starting condition of automatic heating through temperature, rainfall and image recognition. The temperature cannot cover the rainy day scene, and a temperature sensor is required. The method of determining whether to start heating according to the rainfall condition is high in cost, cannot be used in vehicles without light sensors, and cannot cover the scene of fogging of the rearview mirror caused by the temperature difference between the entrance and exit of the garage. The image recognition method is high in cost and greatly affected by light and shielding, and is very high in cost and difficult to arrange cameras, and does not have the conditions for mass production application. SUMMARY

[0003] The purpose of the present application is to provide a rearview mirror automatic heating method based on wiper control to at least solve one of the above technical problems.

[0004] The present application provides the following solutions:

[0005] According to one aspect of the present application, a rearview mirror automatic heating method based on wiper control is provided, and the rearview mirror automatic heating method based on wiper control comprises:

[0006] receiving a rearview mirror automatic heating function setting item signal sent by IVI and wiper working state information sent by BCM;

[0007] obtaining current wiper working information according to the wiper working state information;

[0008] generating a heating control strategy according to the wiper working information and the automatic heating function setting item signal;

[0009] generating a heating control signal according to the heating control strategy through a LIN bus, and sending the heating control signal to a rearview mirror heating device, so that the rearview mirror heating device heats.

[0010] Optionally, the wiper working state information includes a wiper gear, a continuous working time length and a load mutation feature of the wiper motor, and the load mutation feature includes an instantaneous pulse amplitude of the motor current when the wiper blade contacts the windshield and a pulse interval;

[0011] obtaining current wiper working information according to the wiper working state information comprises:

[0012] obtaining a dirt accumulation influence coefficient according to the instantaneous pulse amplitude and the pulse interval;

[0013] The wiper-fouling correlation matrix is obtained according to the fouling influence coefficient, the wiper gear position and the continuous working time length, and the current wiper working information includes the wiper-fouling correlation matrix.

[0014] Optionally, the heating control strategy is generated according to the wiper working information and the automatic heating function setting item signal, and the heating control strategy includes:

[0015] The core reference resonant frequency of the resonant heating system is obtained according to the wiper working information;

[0016] The heating strategy is generated according to the wiper working information and the core reference resonant frequency of the resonant heating system.

[0017] Optionally, the core reference resonant frequency of the resonant heating system is obtained by using the following formula:

[0018] ;

[0019] wherein, is the core reference resonant frequency of the resonant heating system; is the corrected fouling parameter; is the corrected fouling equivalent thickness; is the fouling and lens interface thermal resistance; is the fouling Stefan-Boltzmann coefficient correction term; is the lens density; is the lens specific heat capacity at constant pressure; is the lens thickness; is the lens effective heating area; is the ambient temperature; is the lens equivalent thermal resistance; is the lens equivalent heat capacity.

[0020] Optionally, the heating strategy includes:

[0021] The cleaning state strategy includes: when the fouling influence coefficient is less than a first preset threshold, the heating system is controlled to fine-tune the frequency within a narrow bandwidth of ±2Hz of the core reference resonant frequency of the resonant heating system;

[0022] The fouling state strategy includes: when the fouling influence coefficient is greater than or equal to the first preset threshold and less than or equal to a second preset threshold, the heating system is controlled to fine-tune the frequency within a narrow bandwidth of +5Hz of the core reference resonant frequency of the resonant heating system;

[0023] The aggravation state strategy includes: when the fouling influence coefficient is greater than the second preset threshold, the heating system is controlled to fine-tune the frequency within a narrow bandwidth of +10Hz of the core reference resonant frequency of the resonant heating system.

[0024] Optionally, the generating the heating control strategy according to the wiper working information and the automatic heating function setting item signal comprises:

[0025] obtaining a multi-physical field coupling model of the rearview mirror heating system;

[0026] inputting the obtained current wiper working information, environmental parameters and real-time working state of the heating device into the multi-physical field coupling model, thereby obtaining a prediction result of the rearview mirror lens surface in a future preset time period by solving the multi-physical field coupling model;

[0027] adjusting parameters of the heating control strategy dynamically according to the prediction result.

[0028] Optionally, the multi-physical field coupling model of the rearview mirror heating system comprises an electromagnetic induction heating field equation, a non-Fourier heat conduction-phase change field equation, a fluid dynamics-surface effect field equation and an infrared radiation-environment interaction field equation.

[0029] The solving the multi-physical field coupling model comprises:

[0030] initializing all coupled variables;

[0031] constructing a discretized simultaneous equation set according to the electromagnetic induction heating field equation, the non-Fourier heat conduction-phase change field equation, the fluid dynamics-surface effect field equation and the infrared radiation-environment interaction field equation;

[0032] solving the discretized simultaneous equation set by using the Newton-Raphson method.

[0033] Optionally, the adjusting parameters of the heating control strategy dynamically according to the prediction result comprises:

[0034] obtaining a trained Transformer prediction model;

[0035] inputting the prediction result of the rearview mirror lens surface in the future preset time period, the current wiper working information, the environmental parameters and the real-time working state of the heating device into the trained Transformer prediction model, thereby obtaining a key performance indicator prediction value in the future preset time;

[0036] generating parameters of the heating control strategy according to the key performance indicator prediction value in the future preset time.

[0037] Optionally, the generating parameters of the heating control strategy according to the key performance indicator prediction value in the future preset time comprises:

[0038] defining an optimization target and a constraint;

[0039] generating a candidate parameter combination by using a grid sampling and local refinement strategy;

[0040] For each candidate control parameter combination, constraint checking and objective function calculation are sequentially completed, so as to obtain an effective solution;

[0041] Pareto optimization and optimal solution selection are adopted to obtain an optimal solution from the effective solutions, and parameters of the heating control strategy are obtained based on the optimal solution.

[0042] The application also provides a rearview mirror automatic heating device based on wiper control, which comprises:

[0043] An information acquisition module is configured to receive a rearview mirror automatic heating function setting item signal sent by an IVI and wiper working state information sent by a BCM;

[0044] A current wiper working information acquisition module is configured to acquire current wiper working information according to the wiper working state information;

[0045] A heating control strategy acquisition module is configured to generate a heating control strategy according to the wiper working information and the automatic heating function setting item signal;

[0046] A heating control signal sending module is configured to generate a heating control signal according to the heating control strategy through a LIN bus and send the heating control signal to a rearview mirror heating device, so that the rearview mirror heating device is heated.

[0047] The rearview mirror automatic heating method based on wiper control provided by the application controls heating according to the condition of the wiper, and realizes automatic control of the rearview mirror. The method realizes full coverage of scenarios requiring heating of the rearview mirror in rainy days, garages and the like, and is based on the basic wiper control function of the vehicle, without the need for additional costs, and can realize large-scale application. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 FIG. 1 is a flowchart of the rearview mirror automatic heating method based on wiper control provided by an embodiment of the application.

[0049] Figure 2 FIG. 2 is a system schematic diagram of vehicle wiper and heating control in an embodiment of the application. DETAILED DESCRIPTION

[0050] The technical solutions of the application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are some of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.

[0051] See Figure 2 In the vehicle described in this application, the windshield wiper and heating control system is specifically as follows:

[0052] IVI: Cockpit Domain Controller. Provides settings for automatic rearview mirror heating, which can be turned on or off, and sends setting signals to the CAN bus.

[0053] SCS: Combination switch, manual control of wiper settings, sends operation commands such as electric wiper and automatic wiper to the CAN bus based on user operation.

[0054] BCM: Body Control Controller, responsible for driving the windshield wipers and sending the wiper status information to the CAN bus.

[0055] CGW: Central Domain Controller, receives the status of the automatic rearview mirror heating function, identifies the vehicle's power mode, detects the working status of the wipers and performs working timing, comprehensively judges the conditions for heating to be turned on, and controls the heating of the exterior rearview mirrors to be turned on and off via LIN.

[0056] Left / Right Rearview Mirrors: Activate exterior rearview mirror heating.

[0057] In this embodiment, the user sets the automatic heating function of the rearview mirror to be turned on and off via the IVI, and the IVI sends the setting item (automatic heating function setting item signal) to the bus.

[0058] The CGW receives the on / off status of the automatic rearview mirror heating function from the IVI on the CAN bus. Automatic rearview mirror heating control can only be performed when the function is on.

[0059] like Figure 1 The automatic rearview mirror heating method based on wiper control shown includes:

[0060] Receives the rearview mirror automatic heating function setting signal from the IVI and the wiper working status information from the BCM;

[0061] Obtain the current wiper operating information based on the wiper operating status information;

[0062] A heating control strategy is generated based on wiper operation information and automatic heating function settings.

[0063] According to the heating control strategy, a heating control signal is generated through the LIN bus and sent to the rearview mirror heating device, thereby enabling the rearview mirror heating device to heat up.

[0064] In an alternative embodiment, the user sets the automatic heating function of the rearview mirror to be turned on and off via the IVI, and the IVI sends the settings to the bus.

[0065] The CGW receives the rearview mirror automatic heating function opening and closing state sent by the IVI on the CAN bus, and only in the opening state, the automatic heating control of the outside rearview mirror can be performed.

[0066] When it is rainy or the basement glass is fogged, the user uses the manual wiper or the automatic wiper according to the actual use, and the BCM sends the wiper working state to the CAN bus according to the actual execution.

[0067] The CGW judges the power supply mode, the function opening state and the wiper working state, controls the opening and closing of the outside rearview mirror heating through the LIN bus, and the logic is as follows:

[0068] Manual wiper: triggered after 10s of continuous wiper operation, and exited after 15 minutes or the wiper stops

[0069] Manual wiper: triggered when the number of wiper operations in one minute is greater than or equal to 3 times, and exited after 15 minutes

[0070] Automatic wiper: triggered after 10s of automatic wiper operation, and exited after 15 minutes or the wiper stops

[0071] Automatic wiper: triggered when the number of wiper operations triggered by the sensor in one minute is greater than or equal to 3 times, and exited after 15 minutes.

[0072] In the embodiment, the wiper working state information includes the wiper gear, the continuous working time length and the load mutation characteristics of the wiper motor, and the load mutation characteristics include the instantaneous pulse amplitude and pulse interval of the motor current when the wiper blade contacts the windshield;

[0073] According to the wiper working state information, the current wiper working information is obtained, including:

[0074] According to the instantaneous pulse amplitude and the pulse interval, the dirt accumulation influence coefficient is obtained.

[0075] Specifically, the load mutation characteristics are the instantaneous pulse amplitude (ΔI) and the pulse interval (Δt) of the motor current when the wiper blade contacts the windshield, and the BCM records the pulse sequence through 1kHz high-frequency sampling.

[0076] In the embodiment, the dirt accumulation influence coefficient K is calculated by the following formula:

[0077] K=ΔI / Δt, and the dirt accumulation working condition is divided according to the K value: K>0.5A / ms is high dirt accumulation working condition, 0.2A / ms≤K≤0.5A / ms is medium dirt accumulation working condition, and K<0.2A / ms is low dirt accumulation working condition.

[0078] In the embodiment, the dirt equivalent permittivity and the dirt equivalent thickness are obtained by the dirt influence coefficient K, specifically, the dirt equivalent permittivity εp=1.8+2.2×K (clean glass ε=5.5), and the dirt equivalent thickness d=0.5+0.5×K (clean glass d=0.5mm), both of which constitute a dirt physical parameter set, and together with the wiper-dirt correlation matrix, form the wiper working information.

[0079] In the embodiment, the wiper gear can include four gears (closed, intermittent, low speed, and high speed) (indicated by the SCS combination switch, and synchronized to the bus by the BCM).

[0080] The continuous working time length is counted from the moment when the wiper starts to work, and the unit is second, with an accumulated accuracy of ±0.1s, and the wiper is reset after stopping.

[0081] In the embodiment, the wiper-dirt correlation matrix has the following dimensions: 3 (dirt level) x 4 (wiper gear) x N (continuous working time length interval: 0-5s / 5-10s / above 10s), and each element is the comprehensive dirt influence coefficient K corresponding to the combination of the dirt level-wiper gear-time length interval. com The matrix elements are the pre-set comprehensive influence weights (range 0-1), for example, the weight of high dirt-high speed-above 10s is 0.95, and the weight of low dirt-intermittent-0-5s is 0.3 (as shown in Table 1).

[0082] Table 1: Wiper-dirt correlation matrix table

[0083]

[0084] According to the dirt level corresponding to the current wiper working condition (gear + time length) and the original K, the unique K is extracted from the matrix. com

[0085] In the embodiment, the generation of the heating control strategy according to the wiper working information and the automatic heating function setting item signal includes:

[0086] According to the wiper working information, the core reference resonant frequency of the resonant heating system is obtained.

[0087] According to the wiper working information and the core reference resonant frequency of the resonant heating system, the heating strategy is generated.

[0088] In the embodiment, the core reference resonant frequency of the resonant heating system is obtained by the following formula:

[0089] ;

[0090] wherein, is the core reference resonant frequency of the resonant heating system; ​​for correcting the dirt parameter; for correcting the equivalent thickness of the dirt after correction; for the thermal resistance between the dirt and the lens; for the dirt Stefan-Boltzmann coefficient correction term; for the lens density; for the lens constant-pressure specific heat capacity; for the lens thickness; for the effective heating area of the lens; for the ambient temperature; for the equivalent thermal resistance of the lens; for the equivalent heat capacity of the lens.

[0091] In the embodiment, the dirt parameter is corrected The formula is as follows:

[0092] , wherein is the reference emissivity of the dirt in the clean state (a pre-calibration constant determined by the dirt material, such as dust dirt = 0.8, water stain dirt = 0.9); is the dirt emissivity correction coefficient (calibrated by real vehicle testing, which can be set by the user as needed to reflect the influence degree of the dirt emissivity);

[0093] In the embodiment, the equivalent thickness of the dirt after correction is calculated by the following formula:

[0094] ;

[0095] , wherein is the equivalent thickness of the dirt after correction (unit: mm, which is equivalent to the thickness of a uniform thin layer of the non-uniformly distributed dirt, and is used to quantify the hindering effect of the dirt on heat conduction); is the dirt equivalent thickness correction coefficient (calibrated by real vehicle testing, which reflects the com influence degree of the dirt thickness);

[0096] In the embodiment, the heating strategy includes:

[0097] The clean state strategy includes: when the dirt influence coefficient is less than a first preset threshold, the heating system is controlled to fine-tune the frequency within a narrow bandwidth of ± 2 Hz of the core reference resonant frequency of the resonant heating system;

[0098] The dirt state strategy includes: when the dirt influence coefficient is greater than or equal to the first preset threshold and less than or equal to a second preset threshold, the heating system is controlled to fine-tune the frequency within a narrow bandwidth of + 5 Hz of the core reference resonant frequency of the resonant heating system;

[0099] An aggravated state strategy is provided, which includes: when the contamination impact coefficient is greater than a second preset threshold, controlling the heating system to fine-tune the frequency within a narrow bandwidth of +10Hz from the core reference resonant frequency of the resonant heating system.

[0100] Specifically, the first preset threshold is 0.2A / ms (the boundary between low and medium dirt accumulation), and the second preset threshold is 0.5A / ms (the boundary between medium and high dirt accumulation).

[0101] Level 3 strategy execution logic:

[0102] Clean-state strategy: When K < first preset threshold, the heating system in Dynamic fine-tuning within a narrow bandwidth of ±2Hz, with a step size of 0.5Hz and an adjustment period of 200ms.

[0103] Accumulated contamination strategy: When 0.2 A / ms ≤ K ≤ 0.5 A / ms, the heating system in Fine-tuning within a narrow bandwidth of +5Hz, with a step size of 1Hz and an adjustment period of 100ms.

[0104] Increased-state strategy: When K > 0.5 A / ms, the heating system in Fine-tuning within a narrow bandwidth of +10Hz, with a step size of 1.5Hz and an adjustment period of 50ms.

[0105] Strategy switching mechanism: Real-time monitoring of K value changes across thresholds and immediate strategy switching; after the wipers stop, the current strategy is maintained for 5 seconds, and then gradually adjusted to the clean state according to K decay.

[0106] This solution calculates K by using the instantaneous pulse amplitude ΔI and pulse interval Δt of the wiper motor current, and directly reuses the original current monitoring function of the BCM for the wiper motor (the BCM needs to monitor the motor status in real time to prevent overload). It can achieve dirt accumulation detection without adding new hardware, and is not affected by light or obstruction, making it suitable for all scenarios such as "rainy days, temperature difference fogging in underground parking lots, and dust accumulation".

[0107] This solution achieves quantitative perception of the state of dirt accumulation, replacing traditional qualitative judgment. Traditional solutions often trigger heating based on qualitative thresholds such as rainfall and temperature, which cannot distinguish differences in dirt thickness and adhesion. This solution uses K to convert the impact of dirt accumulation on wiper resistance into a quantitative value, providing quantitative input for subsequent precise heating and avoiding misjudgments such as strong heating for light dirt accumulation or weak heating for heavy dirt accumulation.

[0108] This application uses a matrix to correlate the dirt accumulation level (K classification) + wiper speed + continuous working time, and outputs K. com It dynamically corrects the actual state of dirt accumulation, so that the heating control is matched with the "dirt generation-removal balance" in real time.

[0109] This application implements adaptive correction based on operating conditions, improving control robustness in complex scenarios. Faced with complex scenarios such as wiper speed switching (e.g., from intermittent to high speed) and changes in dirt type (e.g., from rainwater to mud), the matrix can quickly adjust K through pre-calibrated weights. com (For example, the Kcom corresponding to the same K in the mud scenario is 20% higher than that in the rainwater scenario), so there is no need to recalculate the model parameters, ensuring the real-time performance and accuracy of the heating strategy, and solving the problems of fixed parameters and poor scenario adaptability of traditional solutions.

[0110] The reference resonant frequency formula mentioned above can solve the technical problem that traditional fixed power heating cannot adapt to the thermophysical characteristics of dirt accumulation and lens. By integrating parameters such as the corrected emissivity of dirt accumulation, the corrected equivalent thickness of dirt accumulation, the thermal resistance of the lens, and the thermal capacity of the lens, a reference resonant frequency that is precisely matched with the thermophysical characteristics of "dirt accumulation-lens-environment" is calculated. This allows the heating energy to efficiently penetrate the dirt layer and be absorbed by the lens, improving the dirt removal efficiency by more than 20% while avoiding lens overheating.

[0111] In this embodiment, the heating system used in the rearview mirror of this application is a resonant heating system, which is a heating element that utilizes the principle of electromagnetic resonance to achieve efficient energy conversion and directional heating. The working principle is as follows:

[0112] The essence of a resonant heating rod is an electromagnetic resonant coupling system, and its working process can be divided into three stages:

[0113] High-frequency electromagnetic field excitation: The heating rod has a built-in high-frequency inverter circuit (an LC resonant inverter circuit built around the transmitting coil (L) + resonant capacitor (C)) to convert the vehicle's 12V / 24V DC voltage into MHz-level high-frequency AC power (typical frequency range: 1-10MHz), and generates an alternating strong electromagnetic field through the transmitting coil.

[0114] Electromagnetic resonant coupling: A receiving coil (usually a flexible PCB coil or a miniature wound coil) is integrated inside or on the back of the rearview mirror lens. When the natural resonant frequencies of the transmitting coil and the receiving coil are matched, the two coils generate strong electromagnetic coupling, and the receiving coil induces a high-frequency induced current (eddy current).

[0115] Joule heating is generated in a directional manner: induced current flows in the conductive layer of the lens (such as ITO transparent conductive film, metal nano-coating) or the dirt layer (water, frost, oil, etc.), and Joule heating is generated due to resistance loss. The heat is directly applied to the dirt area to achieve rapid defrosting / defogging.

[0116] The core reference resonant frequency mentioned above is the inherent resonant frequency of the LC circuit at the transmitter.

[0117] In another alternative embodiment, the generating the heating control strategy according to the wiper working information and the automatic heating function setting item signal further comprises:

[0118] obtaining a multi-physical field coupling model of the rearview mirror heating system;

[0119] inputting the obtained current wiper working information, environmental parameters and real-time working state of the heating device into the multi-physical field coupling model, thereby obtaining a prediction result of a surface of a rearview mirror lens in a future preset time period by solving the multi-physical field coupling model;

[0120] dynamically adjusting parameters of the heating control strategy according to the prediction result.

[0121] In the embodiment, the multi-physical field coupling model of the rearview mirror heating system comprises an electromagnetic induction heating field equation, a non-Fourier heat conduction-phase change field equation, a fluid dynamics-surface effect field equation and an infrared radiation-environment interaction field equation.

[0122] In the embodiment, the electromagnetic induction heating field equation is as follows:

[0123] ;

[0124] wherein, is a vector differential operator (nabla operator), r is a spatial coordinate, T is temperature, μ(r, T) is a medium magnetic permeability varying with space and temperature, A(r, t) is a vector magnetic potential varying with space and time, σ(r, T) is a medium electrical conductivity varying with space and temperature, t is time, Js(r, f) is an excitation current density source varying with space and heating frequency, and f is heating frequency. is a volume heat generation rate varying with space and time.

[0125] In the embodiment, the non-Fourier heat conduction-phase change field equation is as follows:

[0126] ;

[0127] wherein, is a medium density varying with space, r is a spatial coordinate, c(r, T) is a specific heat capacity at constant pressure varying with space and temperature, T is temperature, is a first-order partial derivative with respect to time, t is time, is a thermal relaxation time varying with space, is a second-order partial derivative with respect to time, k(r, T) is a thermal conductivity varying with space and temperature, is a volume heat generation rate varying with space and time, and L(r) is a phase change latent heat varying with space. is the spatially and temporally varying liquid volume fraction; n is the outward normal vector at the boundary; is the spatially and temporally varying convective heat transfer heat flux density, is the spatially and temporally varying radiative heat transfer heat flux density. denotes the value on the lens surface boundary.

[0128] In the present embodiment, the fluid dynamics-surface effects field equations are as follows:

[0129]

[0130] is the first order partial derivative with respect to time, h(r,t) is the spatially and temporally varying liquid film thickness, r is the spatial coordinate, t is the time, u(r,t) is the spatially and temporally varying liquid film flow velocity, pl is the liquid density, k(r,T) is the spatially and temperature varying thermal conductivity, T is the temperature, T(r,t) is the spatially and temporally varying temperature, n is the outward normal vector at the boundary, L(r) is the spatially varying latent heat of phase change, E(r,t) is the spatially and temporally varying evaporation rate. u(r,t) is the spatially and temporally varying liquid film flow velocity, r is the spatial coordinate, t is the time, pl is the liquid density, g is the gravitational acceleration, h(r,t) is the spatially and temporally varying liquid film thickness, mI(T) is the temperature varying liquid dynamic viscosity, T is the temperature, g(T) is the temperature varying surface tension, cos is the cosine function, 0(r,T) is the spatially and temperature varying contact angle. is the spatially and temporally varying convective heat transfer heat flux density, r is the spatial coordinate, t is the time, is the spatially and temporally varying convective heat transfer coefficient, is the spatially and temporally varying lens surface temperature, is the time varying ambient temperature. is the spatially and temporally varying convective heat transfer coefficient, r is the spatial coordinate, t is the time, C is an empirical constant, Re(u,h) is the Reynolds number (dimensionless) characterized by the liquid film flow velocity and thickness, u is the liquid film flow velocity, h is the liquid film thickness, 0.6 is the Reynolds number exponent, Pr(mI,T) is the Prandtl number (dimensionless) characterized by the liquid dynamic viscosity and temperature, mI is the liquid dynamic viscosity, T is the temperature, 0.4 is the Prandtl number exponent, is the temperature varying fluid thermal conductivity, h is the liquid film thickness.

[0131] In the present embodiment, the infrared radiation-environment interaction field equations are as follows:

[0132] ​;

[0133] is the heat flux density of radiative heat transfer varying with space and time, r is the spatial coordinate, t is the time, ϵ(r, T) is the surface emissivity varying with space and temperature, T is the temperature, σ is the Stefan-Boltzmann constant, Ts(r, t) is the surface temperature of the mirror varying with space and time, is the ambient equivalent radiation temperature varying with time, is the surface radiation reflectivity varying with space and temperature, is the solar radiation heat flux density varying with space and time.

[0134] In the present embodiment, for the convenience of reading, the data required to be used by the multi-physics coupling model of the rearview mirror heating system are described as follows:

[0135] Current wiper working information:

[0136] Wiper gear: converted into the initial value h0 of the initial contaminant thickness h(r, t), the higher the wiper gear (fast > slow > intermittent), the greater the determined precipitation intensity, the greater the initial liquid film thickness h0, which is used as the initial condition of the fluid field to substitute into the liquid film thickness equation, for specific acquisition method, please refer to the following;

[0137] Wiper continuous working time: converted into the correction coefficient of the initial contaminant thickness h(r, t), the longer the working time, the more sufficient the determination of the contamination being wiped off, the decay correction is made to h0 (h0 is multiplied by 0.3 when the time is longer than 5 seconds, and multiplied by 1.0 when the time is less than 1 second), and the final corrected h0 is substituted into the fluid field equation.

[0138] Wiper motor load mutation characteristics (contamination influence coefficient Kcont + initial liquid phase volume fraction );

[0139] Environmental parameters:

[0140] Ambient temperature : directly as the ambient boundary condition of the heat conduction field, the calculation input of the liquid dynamic viscosity in the fluid field , and the basic value of the ambient equivalent radiation temperature in the radiation field, which are substituted into the heat conduction equation, the liquid film flow rate equation, and the radiative heat transfer equation, respectively;

[0141] Ambient relative humidity: converted into the correction term of the ambient equivalent radiation temperature , the higher the humidity, the closer to (correction formula:

[0142] , the corrected Substitute into the radiation heat transfer equation.

[0143] Wind speed: converted into the calculation input of the convective heat transfer coefficient The greater the wind speed, the greater the Reynolds number Re, and the greater the calculated by the convective heat transfer coefficient formula, the final Substitute into the convective heat transfer heat flux density equation;

[0144] Solar radiation heat flux density ; directly as an input item of the radiation heat transfer equation, participate in the calculation of the radiation heat transfer heat flux density .

[0145] Real-time working state of the heating device:

[0146] Current heating power : directly as the core parameter of the excitation current density source , the frequency change directly changes the amplitude of , and then affects the vector magnetic potential A(r, t) and the volumetric heat generation rate q′′′(r, t), which are substituted into the electromagnetic induction field equation.

[0147] Current heating power P: converted into the calibration item of the excitation current density source , the power and satisfy , the amplitude of is calibrated in reverse through the power value, and the volumetric heat generation rate is ensured to be calculated accurately.

[0148] Real-time current of the heating device: converted into the verification item of the excitation current density source , the real-time current is proportional to , if the current measurement value deviates from the derived value by more than 10%, the value of is corrected, and the electromagnetic induction field equation is ensured to be solved accurately Real-time temperature of the lens surface

[0149] : directly as the initial temperature condition of the heat conduction field , and is used to correct the temperature-dependent parameters such as magnetic permeability μ(r, T), electrical conductivity σ(r, T), surface tension γ(T), surface emissivity ϵ(r, T), and substituted into the related equations of the electromagnetic induction field, the heat conduction field, the fluid field, and the radiation field. In this embodiment, the initial liquid film thickness basic value

[0150] is obtained by the following formula:

[0151] ; ​

[0152] wherein, is the initial liquid film thickness base value (unit: m), k g is the gear coefficient (experimental calibration value: k g = 5 x 10-6m / gear), G is the rain wiper gear quantization value (0 = stop -> 0, 1 = intermittent -> 1, 2 = slow -> 2, 3 = fast -> 3).

[0153] In this embodiment, the thickness correction coefficient can also be obtained by the following formula:

[0154] ;

[0155] wherein, is the time length correction coefficient (dimensionless, 0 < k_t < 1), is the rain wiper continuous working time length (unit: s), is the time constant (k = 3 s).

[0156] The initial liquid phase volume fraction is classified as follows:

[0157] ;

[0158] In this embodiment, the final initial liquid film thickness is as follows:

[0159] .

[0160] In this embodiment, the convective heat transfer coefficient is obtained by the following formula:

[0161] ;

[0162] wherein, the Reynolds number ( is the air kinematic viscosity, which varies with , refer to the air property table);

[0163] the Prandtl number ( is the air dynamic viscosity, is the air constant-pressure specific heat capacity, both of which vary with );

[0164] C = 0.664 (flat plate forced convection empirical coefficient), = 0.15 m (rearview mirror characteristic length, fixed), is the wind speed (unit: m / s).

[0165] In the present embodiment, the excitation current density source The calibration is performed by the following equation:

[0166] ;

[0167] wherein, is the calibrated excitation current density source, is the initial , P is the current measured heating power, is the corresponding theoretical power.

[0168] In the present embodiment, the current measured heating power is as follows:

[0169] .

[0170] The excitation current density source The verification correction method is as follows:

[0171] ;

[0172] wherein, is the final corrected , is the measured heating current, is the corresponding theoretical current ( , is the cross-sectional area of the coil).

[0173] In the present embodiment, the solving of the multi-physical field coupling model comprises:

[0174] initializing all coupling variables;

[0175] Specifically, the initial conditions are set as follows: (no initial magnetic field), (initial temperature = ambient temperature), (initially no liquid contaminants), (initial liquid film thickness estimated by the wiper contamination coefficient);

[0176] The boundary conditions are set as follows: , , (the heating frequency f is known).

[0177] According to the electromagnetic induction heating field equation, the non-Fourier heat conduction-phase change field equation, the fluid dynamics-surface effect field equation, and the infrared radiation-environment interaction field equation, a discrete set of simultaneous equations is constructed;

[0178] Specifically, the spatial domain Ω is discretized into a grid using the finite volume method, and the time domain is discretized into steps Δt;

[0179] The partial differential equations (PDEs) of each field are discretized into algebraic equations, and the discrete algebraic equations of all fields are combined into a large sparse matrix equation set: ;

[0180] Wherein is the discrete value vector of all coupled variables; M is the coefficient matrix (containing physical parameters and coupling correlation terms of all fields); B is the right-hand side (containing initial conditions, boundary conditions, and excitation sources).

[0181] The Newton-Raphson method is used to solve the discretized simultaneous equations.

[0182] Specifically, assume that the variable value of the current iteration step is (k is the iteration number);

[0183] Calculate the residual ;

[0184] If the residual (ε is the convergence threshold, such as 10 −6 ), the current X k is the solution of this time step;

[0185] If it does not converge, calculate the Jacobian matrix , update the variable , and repeat the construction of the discretized simultaneous equations and the full-coupling iterative solution until convergence.

[0186] After completing the solution of the current time step t, X k is used as the initial condition for the next time step t+Δt;

[0187] Repeat the construction of the discretized simultaneous equations and the full-coupling iterative solution until the solution of the preset prediction time window (such as 10 seconds in the future) is completed, and the evolution curves of all variables over time (T(r,t), h(r,t), clearance efficiency η(t), etc.) are obtained.

[0188] In this embodiment, the current wiper working information includes the wiper fouling coefficient K, the environmental temperature , the relative humidity RH, the wind speed , the current heating frequency , the current heating power , the initial temperature of the lens surface ;

[0189] The above wiper working information constitutes a fusion feature vector .

[0190] In the embodiment, the prediction result of the mirror lens surface in the future preset time period includes the following information:

[0191] High-precision physical field data for Np=10 time steps (each step Δt=0.5s, a total of 5 seconds), including:

[0192] Lens surface temperature field distribution (r is a spatial coordinate, Δt=1,2,...,10);

[0193] Contaminant (ice / water) thickness distribution ;

[0194] Contaminant liquid volume fraction distribution ;

[0195] Lens internal thermal stress distribution (Avoid lens damage due to overheating / stress overload).

[0196] In the embodiment, the parameters of the heating control strategy dynamically adjusted according to the prediction result include:

[0197] Obtain a trained Transformer prediction model;

[0198] Input the prediction result of the mirror lens surface in the future preset time period, the current wiper working information, the environmental parameters, and the real-time working state of the heating device into the trained Transformer prediction model, so as to obtain the key performance indicator prediction value in the future preset time;

[0199] In the embodiment, the prediction result of the mirror lens surface in the future preset time period is converted into low-dimensional and interpretable statistical features to form a physical feature vector (dimension=24), as shown in the following Table 2:

[0200] Table 2:

[0201]

[0202] (dimension=7) and (dimension=24) are spliced and standardized to obtain the final input feature vector Z(t) (dimension=31) of the deep learning model;

[0203] In the embodiment, the final input feature vector Z(t) is input into the Transformer prediction model, so as to obtain the prediction result (i.e., the key performance indicator prediction value in the future preset time, including the future 10-step removal rate trajectory (0≤η≤1), the future 10-step surface maximum temperature trajectory​ Maximum thermal stress trajectory in 10 steps (MPa), total clearing time , total energy consumption ).

[0204] Generating parameters of the heating control strategy according to the predicted values of the key performance indicators within the future preset time.

[0205] In this embodiment, the generating parameters of the heating control strategy according to the predicted values of the key performance indicators within the future preset time comprises:

[0206] defining optimization objectives and constraints;

[0207] System constraint parameters (calibrated by hardware specifications and material characteristics, fixed and unchanged):

[0208] Physical safety constraints: lens surface temperature safety threshold, lens maximum thermal stress safety threshold;

[0209] Hardware operation constraints: heating frequency range f ∈ [20 kHz, 100 kHz], heating power range P ∈ [5 W, 30 W];

[0210] Control time domain parameters: optimization time domain Nc=5 (covering the future 5 control periods, each period 0.5 seconds), control parameter step size (frequency step size 5 kHz, power step size 2 W);

[0211] Multi-physical field coupling model related parameters: calibration coefficients of convective heat transfer coefficient , volumetric heat generation rate , and excitation current density source .

[0212] In this embodiment, the multi-objective function is as follows:

[0213] Objective 1 (J1): minimizing total clearing time (priority: safety > fast, weight λ1=0.4);

[0214] Objective 2 (J2): minimizing total energy consumption (priority: energy saving second, weight λ2=0.3);

[0215] Objective 3 (J3): maximizing physical constraint margin (priority: safety highest, weight λ3=0.3), defined as:

[0216] ;

[0217] Comprehensive objective function: wherein , are the objective functions of all candidate solutions , The maximum value of the objective function is normalized to eliminate the dimensional difference, and the final goal is to minimize .

[0218] Hard constraints must be met as follows, otherwise the candidate solution is directly invalid:

[0219] Temperature constraint: maximum surface temperature at all prediction time steps ;

[0220] Stress constraint: maximum thermal stress at all prediction time steps ;

[0221] Hardware constraint: heating frequency , heating power ;

[0222] Soft constraints (to be met as much as possible, and penalties are imposed if violated):

[0223] Clearance rate constraint: at the end of the control time domain (2.5 seconds in the future), the clearance rate η ≥ 95%, and if violated, the penalty term is Penalty = 10 × (0.95 - η), which is added to .

[0224] The strategy of grid sampling and local refinement is adopted to generate candidate parameter combinations;

[0225] Specifically, the strategy of global grid sampling + local fine sampling is adopted to ensure coverage of the optimal parameter region, while controlling the number of candidate solutions to improve real-time performance:

[0226] Global grid sampling: within the constraint range of heating frequency (20 kHz, 25 kHz,..., 100 kHz) and heating power (5 W, 7 W,..., 30 W), uniform sampling is performed at a pre-set step size to generate 36 initial candidate control parameter combinations (frequency-power pairs);

[0227] Local fine sampling: extract the optimal control parameter of the last round of optimization (if it is the first run, take the middle value of frequency 50 kHz and power 15 W), construct a local sampling window (frequency ± 15 kHz, power ± 8 W) around it, and supplement 14 candidate combinations at the same step size to avoid missing the optimal solution;

[0228] Candidate solution deduplication and screening: merge the results of the two samplings, remove duplicate combinations, and finally obtain 50 effective candidate control parameter combinations .

[0229] For each candidate control parameter combination, constraint checking and objective function calculation are performed in turn to obtain valid solutions;

[0230] Specifically, for each candidate control parameter combination, physical constraint checking, multi-physical field coupling calculation and objective function evaluation are completed in turn to ensure the feasibility and accuracy of the candidate solution:

[0231] Physical constraint pre-checking: the candidate parameters are substituted into the simplified physical model (focusing on temperature and thermal stress calculation, time-consuming <1 ms, (lightweight neural network trained offline, input f, P, output temperature peak, stress peak), quickly judge whether it violates the hard constraint:

[0232] If the predicted surface maximum temperature exceeds 60℃ or the maximum thermal stress exceeds 80MPa, it is directly marked as "invalid solution" and the subsequent evaluation is skipped;

[0233] If it meets the hard constraint, go to the next step of evaluation.

[0234] Multi-physical field coupling calculation:

[0235] According to the candidate frequency Update the excitation current density source , combined with the candidate power Calibrate the volume heat generation rate ;

[0236] Based on the temperature prediction value corresponding to the candidate parameters, correct the temperature-dependent parameters such as convective heat transfer coefficient and liquid dynamic viscosity , to ensure consistency with the multi-physical field coupling model.

[0237] Fusion prediction evaluation: substitute the adapted candidate parameters into the Transformer prediction model in the above to get the fusion prediction result corresponding to the candidate solution, including the corrected clearance rate trajectory, temperature trajectory, stress trajectory, total clearance time and total energy consumption ;

[0238] Extract the maximum value (maximum value in all ), and maximum value (maximum value in all );

[0239] Second step: normalize (dimensionless effect);

[0240] Calculate the of the candidate solution (based on the predicted temperature peak, stress peak, J3 definition formula calculation);

[0241] Calculate the comprehensive target value (the smaller the targets 1, 2 are, the larger the target 3 is, adjust the sign to unify the optimization direction to minimization, if the soft constraint is violated, add a penalty term): ;

[0242] Use Pareto optimization and optimal solution selection to obtain the optimal solution from each effective solution, and obtain the parameters of the heating control strategy based on the optimal solution.

[0243] Select non-dominated solutions (i.e. there is no other solution that is better in all objectives) from all effective candidate solutions; the final reserved non-dominated solutions form the Pareto front (usually containing 5-8 solutions);

[0244] Optimal solution strategy: use the weighted Tchebyshev method to select the final solution from the Pareto front, which can find the balanced solution closest to the ideal point when there is a multi-objective conflict:

[0245] Define the ideal point: (shortest cleaning time), (lowest energy consumption), (highest safety margin);

[0246] Calculate the weighted distance of each front solution to the ideal point:

[0247] ;

[0248] Where is the minimum value of in the Pareto front;

[0249] Select the front solution with the smallest as the optimal control parameter .

[0250] Optimal control parameter structure, including:

[0251] Core control parameters: optimal heating frequency (precise to 1 kHz), optimal heating power (precise to 1 W);

[0252] Safety redundancy information: temperature margin , stress margin ;

[0253] Predictive performance indicators: predicted cleaning time corresponding to the optimal parameters, predicted total energy consumption .

[0254] The output will be sent directly to the heating device driving circuit through the LIN bus, completing the dynamic update of the heating strategy, and at the same time serving as the historical information input for the next round of optimization, forming a closed-loop control.

[0255] With the above technical solution, the present application has the following advantages:

[0256] The multi-physical field coupling model covers the whole chain of "energy input-heat conduction-phase change-fluid motion-environment interaction". The electromagnetic induction heating field equation accurately calculates the heating energy generation through vector magnetic potential, electrical conductivity and other parameters. The non-Fourier heat conduction-phase change field equation considers the thermal relaxation effect and phase change latent heat. The fluid dynamics equation describes the liquid film flow and scraping process. The infrared radiation field equation integrates the effects of environmental radiation and solar radiation. The four equations are coupled with each other to avoid the error caused by ignoring the cross-influence of single physical field modeling.

[0257] The input parameters are quantitatively mapped to the physical model. The wiper gear is converted into the initial liquid film thickness through the gear coefficient calibrated by experiment. The working time is adjusted by an exponential correction formula to optimize the liquid film thickness. The environmental humidity is optimized by a correction formula to optimize the radiation temperature. The heating power and excitation current density source are calibrated by a quantitative formula. All inputs are not qualitative judgments but accurate physical quantities that can be directly used by the model to ensure the accuracy of the modeling input.

[0258] First, high-precision physical field data (temperature, thickness, stress, etc.) are obtained by solving the multi-physical field coupling model, and then converted into a 24-dimensional physical feature vector, which is spliced with a 7-dimensional real-time state vector to form a 31-dimensional input. The Transformer model can capture the evolution law of physical quantities over time, accurately predict future cleaning rate, temperature and stress trajectory, and provide "forward-looking" basis for control parameter adjustment to avoid the hysteresis of traditional feedback control.

[0259] The comprehensive objective function balances the core requirements through weight allocation (safety 0.3, speed 0.4, energy saving 0.3), directly relates to the lens material characteristics (temperature ≤60℃, stress ≤80MPa) and hardware specifications (frequency 20-100kHz), and ensures the cleaning effect through the penalty term. This design avoids the trade-off caused by single objective (such as only pursuing fast heating while ignoring lens damage), and eliminates the dimensional difference through normalization to ensure the scientificity of the optimization logic.

[0260] First, a lightweight neural network (input frequency, power, output temperature / stress peak value) is used to quickly pre-check the candidate parameters, and the parameters that violate the hard constraints are directly eliminated. Then, the temperature-dependent parameters (such as thermal conductivity, viscosity) are corrected through multi-physical field coupling calculation to ensure the consistency of the parameters and the model. Finally, the performance indicators at all time steps are predicted by the Transformer model, and the threefold verification avoids the safety risks such as over-temperature and over-stress during the heating process.

[0261] The real-time current of the heating device is associated with the excitation current density source through a formula, and dynamic correction is performed when the deviation exceeds 10%, so as to ensure the accuracy of energy generation calculation; the optimal parameters of the last round of optimization are used as the core of local sampling in the next round, and the control strategy is continuously adjusted in combination with the dynamic changes of environmental parameters (wind speed and humidity), so as to avoid the reliability decline of fixed parameters under complex working conditions.

[0262] The global grid sampling covers the whole parameter interval (36 candidates) according to the step length, so as to ensure that the global optimal solution is not missed; the local fine sampling supplements 14 candidates around the historical optimal solution, so as to improve the local search accuracy; finally, 50 candidate solutions can not only ensure the optimization effect, but also avoid the calculation delay caused by full-space exhaustion, so as to meet the real-time control demand (0.5 seconds per control cycle).

[0263] The physical constraint pre-check uses a lightweight neural network (time consumption <1 ms) trained offline to replace the complex multi-physical field full solution, so as to quickly filter out invalid solutions; the solution results (such as temperature field data) of the multi-physical field coupling model can be reused as input features of the Transformer model, so as to avoid repeated calculation and ensure that the whole optimization process is completed in a short time, thereby adapting to the real-time requirement of the vehicle-mounted scene.

[0264] The application also provides a rearview mirror automatic heating device based on wiper control.

[0265] An information acquisition module is configured to receive a rearview mirror automatic heating function setting item signal sent by an IVI and wiper working state information sent by a BCM.

[0266] A current wiper working information acquisition module is configured to acquire current wiper working information according to the wiper working state information.

[0267] A heating control strategy acquisition module is configured to generate a heating control strategy according to the wiper working information and the automatic heating function setting item signal.

[0268] A heating control signal sending module is configured to generate a heating control signal according to the heating control strategy through a LIN bus and send the heating control signal to a rearview mirror heating device, so that the rearview mirror heating device is heated.

[0269] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions recorded in the above embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for automatic heating of a rearview mirror based on windshield wiper control, characterized in that, The automatic rearview mirror heating method based on wiper control includes: Receives the rearview mirror automatic heating function setting signal from the IVI and the wiper working status information from the BCM; Obtain the current wiper operating information based on the wiper operating status information; A heating control strategy is generated based on wiper operation information and automatic heating function settings. According to the heating control strategy, a heating control signal is generated through the LIN bus and sent to the rearview mirror heating device, thereby enabling the rearview mirror heating device to heat up.

2. The automatic rearview mirror heating method based on wiper control as described in claim 1, characterized in that, The wiper operating status information includes wiper setting, continuous operating time, and load change characteristics of the wiper motor. The load change characteristics include the instantaneous pulse amplitude and pulse interval of the motor current when the wiper blade contacts the windshield. The current wiper operating information obtained from the wiper operating status information includes: The contamination impact coefficient is obtained based on the instantaneous pulse amplitude and pulse interval; The wiper-dirt accumulation correlation matrix is ​​obtained based on the dirt accumulation impact coefficient, wiper speed setting, and continuous working duration. The current wiper working information includes the wiper-dirt accumulation correlation matrix.

3. The automatic rearview mirror heating method based on wiper control according to claim 2, characterized in that, The heating control strategy generated based on wiper operation information and automatic heating function setting signals includes: The core reference resonant frequency of the resonant heating system is obtained based on the wiper operation information. A heating strategy is generated based on the wiper operation information and the core reference resonant frequency of the resonant heating system.

4. The automatic rearview mirror heating method based on wiper control as described in claim 3, characterized in that, The core reference resonant frequency of the resonant heating system is obtained using the following formula: ； in, This is the core reference resonant frequency of the resonant heating system; To correct the dirt accumulation parameters; The corrected equivalent thickness of the accumulated dirt; Thermal resistance at the interface between dirt accumulation and lens; This is a correction term for the Stefan-Boltzmann coefficient of accumulated pollution; Lens density; The specific heat capacity at constant pressure of the lens; Lens thickness; The effective heating area of ​​the lens; Ambient temperature; The equivalent thermal resistance of the lens; This is the equivalent heat capacity of the lens.

5. The automatic rearview mirror heating method based on wiper control as described in claim 4, characterized in that, The heating strategy includes: The cleaning state strategy includes: when the contamination impact coefficient is less than a first preset threshold, controlling the heating system to fine-tune the frequency within a narrow bandwidth of ±2Hz from the core reference resonant frequency of the resonant heating system; The contamination state strategy includes: when the contamination influence coefficient is greater than or equal to a first preset threshold and less than or equal to a second preset threshold, controlling the heating system to fine-tune the frequency within a narrow bandwidth of +5Hz from the core reference resonant frequency of the resonant heating system. An aggravated state strategy is provided, which includes: when the contamination impact coefficient is greater than a second preset threshold, controlling the heating system to fine-tune the frequency within a narrow bandwidth of +10Hz from the core reference resonant frequency of the resonant heating system.

6. The automatic rearview mirror heating method based on wiper control as described in claim 1, characterized in that, The heating control strategy generated based on wiper operation information and automatic heating function setting signals includes: Obtain the multiphysics coupling model of the rearview mirror heating system; The current wiper operation information, environmental parameters, and real-time operating status of the heating device are input into the multiphysics coupling model. By solving the multiphysics coupling model, the predicted results of the rearview mirror lens surface within a preset time period are obtained. Based on the prediction results, the parameters of the heating control strategy are dynamically adjusted.

7. The automatic rearview mirror heating method based on wiper control as described in claim 6, characterized in that, The multi-physics coupling model of the rearview mirror heating system includes the electromagnetic induction heating field equation, the non-Fourier heat conduction-phase transition field equation, the fluid dynamics-surface effect field equation, and the infrared radiation-environment interaction field equation. The solution to the multiphysics coupling model includes: Initialize all coupling variables; A set of discretized simultaneous equations is constructed based on the electromagnetic induction heating field equation, the non-Fourier heat conduction-phase transition field equation, the fluid dynamics-surface effect field equation, and the infrared radiation-environment interaction field equation. The Newton-Raphson method is used to solve the discretized system of equations.

8. The automatic rearview mirror heating method based on wiper control as described in claim 7, characterized in that, The parameters for dynamically adjusting the heating control strategy based on the prediction results include: Obtain the trained Transformer prediction model; The predicted results of the rearview mirror lens surface, the current wiper operation information, environmental parameters, and the real-time operating status of the heating device within a preset time period are input into the trained Transformer prediction model to obtain the predicted values ​​of key performance indicators within the preset time period. The parameters for the heating control strategy are generated based on the predicted values ​​of key performance indicators within a preset timeframe.

9. The automatic rearview mirror heating method based on wiper control as described in claim 8, characterized in that, The parameters for generating the heating control strategy based on predicted values ​​of key performance indicators within a preset future timeframe include: Define the optimization objective and constraints; Candidate parameter combinations are generated using a strategy of grid sampling and local refinement. For each candidate combination of control parameters, constraint verification and objective function calculation are performed sequentially to obtain an effective solution; Pareto optimization and optimal solution selection are employed to obtain the optimal solution from all valid solutions, and the parameters of the heating control strategy are obtained based on the optimal solution.

10. An automatic rearview mirror heating device based on wiper control, characterized in that, The automatic rearview mirror heating device based on wiper control includes: The information acquisition module is used to receive the rearview mirror automatic heating function setting signal from the IVI and the wiper working status information from the BCM. The current wiper working information acquisition module is used to acquire current wiper working information based on wiper working status information. A heating control strategy acquisition module is used to generate a heating control strategy based on wiper operating information and automatic heating function setting signals. A heating control signal sending module is used to generate a heating control signal via a LIN bus according to a heating control strategy, and send the heating control signal to the rearview mirror heating device, thereby enabling the rearview mirror heating device to heat up.

Images