Vehicle-mounted defrosting and demisting intelligent control system and method capable of achieving partition dynamic adjustment

Abstract

The invention discloses a vehicle-mounted defrosting and demisting intelligent control system and method for zoning dynamic adjustment, and relates to the technical field of vehicle-mounted control, and the method comprises the steps: initializing and activating a dual-camera visual sensor array, an infrared temperature field detection device, a pipeline pressure sensor, a vehicle cabin temperature and humidity sensor and a zoning air port execution mechanism; the method comprises the following steps of: performing multi-modal fusion processing on visible light images and infrared temperature data of the double cameras, obtaining characteristic indexes of moisture condensation and ice crystals on the surface of partitioned glass, generating continuous and adjustable total wind speed signals, dynamically distributing the total wind speed signals in a normalized manner according to a preset distribution index, obtaining partitioned wind speed signals, constructing a time sequence prediction model, and predicting the wind speed of the partitioned glass according to the time sequence prediction model. And the atomization risk score is output, a thermal field compensation coefficient is generated through mapping, the thermal field compensation coefficient is issued to a partition air opening executing mechanism, and partition defrosting and demisting are completed. The defrosting and demisting precision, the energy utilization rate and the response speed are remarkably improved by recognizing the frosting and dewing states of the partitions, optimizing the energy efficiency through wind speed distribution and conducting thermal field pre-compensation.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-mounted control technology, and in particular to a vehicle-mounted defrosting and defogging intelligent control system and method with dynamic zoning adjustment. Background Art

[0002] With the rapid development of new energy vehicles and intelligent driver assistance systems, in-vehicle comfort and driving safety are receiving increasing attention. Condensation, frost, or ice crystals on the windshield not only impair the driver's field of vision but can also cause the air conditioning and defogger systems to operate for extended periods, increasing vehicle energy consumption and exacerbating component wear. To improve defrosting and defogging efficiency, existing in-vehicle systems typically utilize a single heater or direct hot air flow, coupled with a fixed air volume to uniformly heat or blow air across the entire windshield surface.

[0003] However, fixed heating and uniform air supply methods struggle to balance energy consumption and defogging speed. On the one hand, some areas often overheat without condensation, while on the other hand, it's difficult to promptly remove fog and frost from remote areas. While some high-end models have introduced multi-zone air vents, their air volume distribution is often based on preset time periods or temperature thresholds, lacking precise perception and adaptive control of the microscopic condensation state and ice crystal distribution characteristics on the glass surface. Furthermore, existing systems often rely on simple temperature or humidity feedback, failing to integrate visible light images and infrared temperature field data, and failing to incorporate time-series prediction models to anticipate fogging risks. Consequently, system responses often lag behind actual fogging trends, preventing preventative dynamic compensation.

[0004] Therefore, there is an urgent need for an intelligent control method that can monitor the condensation and ice crystal status of different partitions of the front windshield in real time, dynamically adjust the partition air volume and heating power, and combine historical and real-time environmental data to predict the fogging risk, so as to achieve energy-saving and efficient partitioned adaptive adjustment while ensuring rapid defrosting and defogging. Summary of the Invention

[0005] Based on the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a vehicle-mounted defrosting and defogging intelligent control system and method with dynamic adjustment of partitions to solve the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of partitions, comprising:

[0007] S1: Initialize and activate the dual-camera visual sensor array, infrared temperature field detection device, duct pressure sensor, cabin temperature and humidity sensor, and partition air vent actuator;

[0008] S2: Perform multimodal fusion processing on the dual-camera visible light images and infrared temperature data to obtain the characteristic indicators of condensation and ice crystals on the partitioned glass surface;

[0009] S3: Based on the static pressure of the inlet airflow, a continuously adjustable total wind speed signal is generated through an energy-saving fluid amplifier. According to the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin, the total wind speed signal is dynamically normalized and distributed according to the preset distribution index to obtain the partitioned wind speed signal;

[0010] S4: Build a time series prediction model based on historical and real-time environmental data, evaluate the fogging risk within a preset lead time in the future, and output a fogging risk score;

[0011] S5: Generate thermal field compensation coefficient based on the predicted fogging risk score mapping and send it to the partitioned air outlet actuator to complete the partitioned defrosting and defogging.

[0012] The present invention is further configured such that step S1 comprises:

[0013] Initialize the dual-camera visual sensor array communication interface, load the camera calibration parameters and distortion correction matrix, and perform a self-check process to confirm that the image acquisition channel is complete;

[0014] Activate the infrared temperature field detection device, read the built-in temperature calibration point, calibrate the temperature drift and generate the initial temperature field baseline diagram;

[0015] Enable the pipeline pressure sensor, clear the output value and verify the pressure response characteristics, and confirm the range and linearity by injecting calibration gas source;

[0016] Start the cabin temperature and humidity sensor module, collect the initial values ​​of ambient temperature and humidity, calibrate the sensor bias, and store the reference humidity curve;

[0017] Establish a control bus connection for the partitioned air outlet actuator, send speed measurement commands in sequence, monitor the air outlet feedback position, and confirm the actuator's motion range and response delay;

[0018] Synchronize the trigger timestamp, synchronize the signal channel data stream, complete the initialization status confirmation and switch to the online monitoring mode.

[0019] The present invention is further configured such that step S2 includes:

[0020] Calibrate the internal and external parameters of the dual cameras and infrared module to obtain the camera distortion matrix and temperature mapping lookup table;

[0021] Perform dedistortion, color correction, and white balance processing on the visible light original image, and output a geometrically corrected high-fidelity image;

[0022] Perform temperature drift calibration and linear interpolation on the infrared raw data to generate the absolute temperature field map corresponding to each pixel;

[0023] Using feature point matching or mutual information maximization methods, the registration transformation matrix from the visible light image to the infrared temperature field is calculated, and the infrared image is resampled and aligned to the visible light coordinate system;

[0024] According to the preset partition outline, the registered visible light and infrared images are synchronously divided into blocks to extract the partitioned regions of interest;

[0025] Multi-scale Laplace enhancement and Gabor filter bank are applied to the partitioned region of interest to extract the edge response value of the fine water film and calculate the average temperature of the partitioned area.

[0026] After nonlinear mapping and normalization processing of the edge response value of the micro-water film and the average temperature of the partition, the condensation and ice crystal fusion characteristics of the partitioned glass surface are generated.

[0027] The present invention is further configured such that the logic for generating the condensation and ice crystal fusion characteristics on the partitioned glass surface is as follows: C i (t) is the condensation and ice crystal fusion characteristics of the glass surface in partition i at time t, S i (t) is the edge response value of the micro-water film in partition i at time t, T i (t) is the average temperature of partition i at time t, η fus is the visual fusion coefficient, κ ir is the infrared temperature sensitivity coefficient, ψ ice is the ice crystal response power exponent.

[0028] The present invention is further configured such that step S3 includes:

[0029] The inlet static pressure value is obtained according to the preset sampling frequency, and is processed by low-pass filtering, and the filtered static pressure value is range-verified and limited;

[0030] Load the calibration parameters of the energy-saving fluid amplifier and apply a gain mapping to the inlet static pressure value to generate a continuously adjustable total wind speed signal;

[0031] Convert the total wind speed signal into a drive command, send it through the control bus, and adjust the fan output;

[0032] The partition reference wind speed signal is generated based on the partition blade optimization coefficient and the partition allocation index, and the partition reference wind speed signal is corrected in combination with the regional temperature difference adjustment factor;

[0033] The corrected partition reference wind speed signal is normalized, and the total wind speed signal is distributed according to the partition ratio to obtain the partition wind speed signal.

[0034] The present invention is further configured such that the generation logic of the total wind speed signal is: F out(t) is the total wind speed signal at time t, Γ exp is the diffusion gain coefficient, λ amp is the exponential growth coefficient, P in (t) is the inlet static pressure value at time t, φ chan is the flow channel adjustment index;

[0035] The generation logic of the partition reference wind speed signal is: B i (t) is the reference wind speed signal of partition i at time t, Ψ leaf (i) is the partition blade optimization coefficient of partition i, δ sys Assign an index to the system;

[0036] The logic for correcting the partition reference wind speed signal is: B' i (t) = B i (t)×[1+θzone×ΔT i (t)],B' j (t) is the modified zone reference wind speed signal, θzone is the temperature difference adjustment factor, ΔT i (t) is the temperature difference between the glass surface and the cabin of partition i at time t;

[0037] The generation logic of the normalized partition wind speed signal is: U i (t) is the wind speed signal of partition i at time t, S adj (t) is the corrected partition reference wind speed signal and, N zone is the total number of partitions.

[0038] The present invention is further configured such that step S4 includes:

[0039] Aggregate historical environmental data and real-time sensor data, and divide historical series and real-time series into fixed time windows;

[0040] Normalization and difference preprocessing are performed on each time series to eliminate dimension differences and trend drift;

[0041] Construct multi-channel input tensors by data category and load the pre-trained multi-layer LSTM network model structure and parameters;

[0042] The processed historical sequence is forward calculated through a multi-layer LSTM network to extract the sequence time series feature vector;

[0043] Append the real-time sequence to the end of the historical sequence to form the prediction input, and obtain the risk mapping value within the future preset lead time at the LSTM output layer;

[0044] The mapped values ​​are adjusted by a nonlinear saturation function and a time weighting coefficient to generate a normalized atomization risk score.

[0045] The present invention is further configured such that step S5 includes:

[0046] Read the atomization risk score output by the time series prediction module, and convert the atomization risk score into a thermal field compensation coefficient according to the preset control gain and saturation mapping function;

[0047] The thermal field compensation coefficient is mapped to the target power output of the partition air outlet actuator, and the target power output is executed to compensate the output heat to achieve partitioned defrosting and demisting.

[0048] The present invention is further configured such that the calculation logic of the thermal field compensation instruction is: is the thermal field compensation coefficient of partition i at time t, ξ ctrl To control the gain, is the fogging risk score of partition i at time t, σ ctrl is the saturation mapping index.

[0049] The present invention also provides a vehicle-mounted defrosting and defogging intelligent control system with dynamic adjustment of partitions, which is used to implement the above-mentioned vehicle-mounted defrosting and defogging intelligent control method with dynamic adjustment of partitions, including:

[0050] Initial module: Initializes and activates the dual-camera visual sensor array, infrared temperature field detection device, duct pressure sensor, cabin temperature and humidity sensor, and partitioned air vent actuator;

[0051] Fusion module: Performs multimodal fusion processing on the dual-camera visible light images and infrared temperature data to obtain characteristic indicators of condensation and ice crystals on the partitioned glass surface;

[0052] Distribution module: Based on the static pressure of the inlet airflow, a continuously adjustable total wind speed signal is generated through an energy-saving fluid amplifier. Based on the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin, the total wind speed signal is dynamically normalized and distributed according to the preset distribution index to obtain the partitioned wind speed signal;

[0053] Prediction module: Builds a time series prediction model based on historical and real-time environmental data, evaluates the atomization risk within a preset lead time in the future, and outputs a atomization risk score;

[0054] Compensation module: Generates thermal field compensation coefficients based on the predicted fogging risk score mapping and sends them to the partitioned air outlet actuators to complete partitioned defrosting and defogging.

[0055] The present invention provides an intelligent control system and method for dynamically adjusting vehicle defrosting and defogging by partition. The method initializes and activates a dual-camera visual sensor array, an infrared temperature field detection device, a duct pressure sensor, a cabin temperature and humidity sensor, and a partitioned air outlet actuator; performs multimodal fusion processing on the dual-camera visible light image and infrared temperature data to obtain characteristic indicators of condensation and ice crystals on the partitioned glass surface; generates a continuously adjustable total wind speed signal based on the inlet airflow static pressure through an energy-saving fluid amplifier; dynamically normalizes and distributes the total wind speed signal according to a preset distribution index based on the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin to obtain a partitioned wind speed signal; constructs a time series prediction model based on historical and real-time environmental data, evaluates the fogging risk within a preset lead time in the future, and outputs a fogging risk score; generates a thermal field compensation coefficient based on the predicted fogging risk score mapping, and sends it to the partitioned air outlet actuator to complete the partitioned defrosting and defogging, resulting in the following beneficial effects:

[0056] 1. Improve defrosting and defogging accuracy: Combined with the internal and external parameter calibration and image registration of visible light and infrared temperature fields, the edge response value and average temperature of the fine water film are extracted to accurately reflect the condensation and ice crystal status of each partitioned glass surface, realizing differentiated identification and processing of different areas;

[0057] 2. Energy efficiency optimization and dynamic wind speed distribution: An energy-saving fluid amplifier generates a continuously adjustable total wind speed signal and dynamically normalizes wind speed distribution based on the partitioned blade optimization coefficient and the temperature difference within the glass cabin. This not only meets the needs of local rapid demisting, but also effectively reduces overall fan power consumption and improves energy utilization.

[0058] 3. Active prediction and advance compensation: Based on a multi-layer LSTM time series prediction model, the fogging risk within the preset advance time is accurately assessed, a fogging risk score is generated, and the score is mapped to a thermal field compensation coefficient. This enables preheating compensation for high-risk zones, avoiding the delays and blind spots of traditional passive waiting-based defoggers.

[0059] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without inventive efforts. In the drawings:

[0061] Figure 1 This is a flow chart of an intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to an exemplary embodiment of the present invention;

[0062] Figure 2 The figure is a schematic structural diagram of a vehicle-mounted defrosting and defogging intelligent control system with dynamic adjustment of partitions according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0063] The following describes the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art will readily appreciate the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the various details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are intended only to illustrate the present invention and are not intended to limit the scope of protection of the present invention.

[0064] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0065] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0066] Example 1

[0067] Intelligent control method for vehicle defrosting and defogging with dynamic adjustment of partitions, such as Figure 1 Shown, including:

[0068] S1: Initialize and activate the dual-camera visual sensor array, infrared temperature field detection device, duct pressure sensor, cabin temperature and humidity sensor, and partition air vent actuator;

[0069] S2: Perform multimodal fusion processing on the dual-camera visible light images and infrared temperature data to obtain the characteristic indicators of condensation and ice crystals on the partitioned glass surface;

[0070] S3: Based on the static pressure of the inlet airflow, a continuously adjustable total wind speed signal is generated through an energy-saving fluid amplifier. According to the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin, the total wind speed signal is dynamically normalized and distributed according to the preset distribution index to obtain the partitioned wind speed signal;

[0071] S4: Build a time series prediction model based on historical and real-time environmental data, evaluate the fogging risk within a preset lead time in the future, and output a fogging risk score;

[0072] S5: Generate thermal field compensation coefficient based on the predicted fogging risk score mapping and send it to the partitioned air outlet actuator to complete the partitioned defrosting and defogging.

[0073] The present invention is further configured such that step S1 comprises:

[0074] Initialize the dual-camera visual sensor array communication interface, load camera calibration parameters and distortion correction matrix, and perform a self-test process to confirm the integrity of the image acquisition channel. Specifically, open the serial or Ethernet interface of the left and right cameras in turn, load the internal and external parameters (including focal length, principal point, and distortion coefficient) obtained by offline calibration, and calculate the distortion correction matrix on the board. Confirm the synchronization and image integrity of the left and right cameras by sending frame grab commands and checking the frame rate and resolution.

[0075] Activate the infrared temperature field detection device, read the built-in temperature calibration points, calibrate the temperature drift, and generate an initial temperature field baseline diagram. Specifically, enable power supply and sampling of the infrared detector, read the sensor's response values ​​at multiple blackbody temperatures, correct the temperature drift through an interpolation algorithm, generate a standardized temperature-voltage mapping table, and collect a frame of temperature field diagram under no-load conditions as the environmental baseline.

[0076] Enable the pipeline pressure sensor, clear the output value and verify the pressure response characteristics. Confirm the range and linearity by injecting the calibration gas source. Specifically, reset the pressure sensor output to the zero reference, apply a known standard pressure (calibration gas source) and record the output value. Compare the output with the standard value to verify whether the upper and lower limits of the range and the linear deviation meet the design specifications.

[0077] The cabin temperature and humidity sensor module is activated to collect initial ambient temperature and humidity values, calibrate the sensor bias, and store a baseline humidity curve. Specifically, the initial values ​​of the cabin temperature and humidity sensor are read and compared with those of an environmental calibrator to calculate the zero bias. Multi-point data is collected in different humidity environments, and a calibration curve is generated by fitting and written into the sensor module as a baseline.

[0078] Establish a control bus connection for the partitioned air outlet actuators, send speed measurement commands sequentially, and monitor the air outlet feedback position to confirm the actuator's range of motion and response delay. Specifically, register the address of each air outlet actuator on the CAN or LIN bus and cyclically send commands for minute opening changes. Read feedback through the built-in position or speed sensor to measure the minimum controllable opening, maximum opening, and the time delay from command to execution.

[0079] The system triggers the timestamp synchronously, aligns the signal channel data streams, completes the initialization status confirmation, and switches to the online monitoring mode. Specifically, the main control unit outputs a unified clock synchronization pulse (or PTP protocol) to align the camera, infrared, pressure, and temperature and humidity data streams on the time axis; after verifying that the delay deviation of each channel is within an acceptable range, the system status is set to "online" to start the formal data acquisition and control cycle; step S1 lays the foundation for the high-precision, multi-modal, and low-latency collaborative work of the entire system through self-inspection, calibration, and synchronization of each perception and execution module, ensuring the reliability and consistency of subsequent environmental perception, prediction, and execution control.

[0080] The present invention is further configured such that step S2 includes:

[0081] The dual cameras and infrared module are calibrated for internal and external parameters to obtain the camera distortion matrix and temperature mapping lookup table. Specifically, the binocular visible light cameras are calibrated using a checkerboard grid to obtain internal parameters (including focal length, principal point, and distortion coefficient) and external parameters (including rotation matrix and translation vector). The infrared detection module is sampled at multiple blackbody temperatures to construct a temperature-to-digital value lookup table.

[0082] Dedistortion, color correction, and white balance are performed on the original visible light image to output a geometrically corrected high-fidelity image. Specifically, the distortion matrix obtained by calibration is applied to the original image to dedistort it and eliminate radial and tangential lens distortion. Color correction and white balance are performed to restore the true color distribution, and a geometrically and color-corrected high-fidelity image is output.

[0083] Perform temperature drift calibration and linear interpolation on the infrared raw data to generate an absolute temperature field map corresponding to each pixel. Specifically, perform temperature drift calibration on the infrared raw data to compensate for the offset caused by the temperature change of the detector itself. Use a linear interpolation algorithm to expand the discrete lookup table value into an absolute temperature field map corresponding to each pixel.

[0084] Using feature point matching or mutual information maximization methods, the registration transformation matrix from the visible light image to the infrared temperature field is calculated, and the infrared image is resampled and aligned to the visible light coordinate system. Specifically, based on feature point matching or mutual information maximization strategies for the visible light and infrared images, the homography or rigid transformation matrix between the two is calculated. The infrared temperature field map is resampled so that its pixel grid is completely aligned with the visible light image.

[0085] The registered visible light and infrared images are synchronously divided into blocks according to the preset partition outline, and the partitioned regions of interest are extracted. Specifically, according to the pre-defined windshield partition outline, the registered visible light image and infrared temperature field are synchronously divided into blocks; the region of interest (ROI) of each area is extracted so that subsequent feature processing only acts on the valid area;

[0086] Multi-scale Laplacian enhancement and Gabor filter banks are applied to the partitioned regions of interest to extract the edge response values ​​of the fine water film and calculate the average temperature of the partitioned regions. Specifically, multi-scale Laplacian enhancement and Gabor filter banks are applied to the visible light image of each region ROI to extract the response of the water film edge and fine texture. The average temperature of all pixels in the corresponding infrared ROI is calculated to reflect the characteristics of ice crystal aggregation or temperature difference.

[0087] After nonlinear mapping and normalization of the edge response values ​​of the micro-water film and the average temperature of each partition, the condensation and ice crystal fusion characteristics of the partitioned glass surface are generated. Specifically, Sigmoid, logarithmic, or power mapping is applied to the edge response values ​​and the average temperature to achieve nonlinear amplification and noise suppression. After normalizing the mapping results to a unified dimension, the condensation and ice crystal fusion characteristics of each partition are generated.

[0088] The present invention is further configured such that the logic for generating the condensation and ice crystal fusion characteristics on the partitioned glass surface is as follows: C i (t) is the condensation and ice crystal fusion characteristics of the glass surface in partition i at time t, S i (t) is the edge response value of the micro-water film in partition i at time t, T i (t) is the average temperature of partition i at time t, η fus is the visual fusion coefficient, κ ir is the infrared temperature sensitivity coefficient, ψ ice is the ice crystal response power index. Specifically, The Sigmoid function is used to convert the edge response value S of the micro water film into i (t) is mapped to (0,1), where the mapping slope is determined by the visual fusion coefficient η fus Decision. When S i When (t) is low, Close to 0, suppressing false texture areas; when S i When (t) is high, Close to 1, amplifying the contribution of the real texture area to the fusion feature; S i The calculation logic of (t) is: Ω i is the set of all pixels in partition i, |Ω i| is the total number of pixels in partition i, S vis (p, t) is the visible light texture response value at pixel p, which is extracted by multi-scale Laplacian operator and Gabor filter and then normalized to [0, 1]; The average temperature of the region T i First, the power operation is performed to characterize the nonlinear sensitivity of different crystal forms to temperature (exponent ψ ice ), then multiplied by the temperature sensitivity coefficient κ ir And add one to take the logarithm to suppress the excessive amplification caused by extreme temperature values. T(p,t) is the absolute temperature value at pixel point p; the gated texture weight Mapping values ​​with temperature Multiplication ensures that the temperature effect is activated only when there is an obvious water film or ice crystal texture, and can also adaptively adjust the fusion feature amplitude according to the temperature field intensity. The visual fusion coefficient η fus Controls the slope of the Sigmoid mapping, with a value range of [0.1, 10], and the infrared temperature sensitivity coefficient κ ir The temperature power value is mapped to the fusion feature scale with a value range of [0.01, 1], and the ice crystal response power index ψ ice Characterizes the sensitivity of different crystal forms to temperature, with a value range of [1,3].

[0089] The present invention is further configured such that step S3 includes:

[0090] The inlet static pressure value is obtained according to the preset sampling frequency, and is low-pass filtered. The filtered static pressure value is then range-verified and amplitude-limited. Specifically, the raw inlet static pressure value is read from the pipeline pressure sensor at the preset sampling frequency. The raw static pressure value is low-pass filtered to eliminate high-frequency noise interference. The filtered data is compared with the upper and lower limits of the sensor range. If the range is exceeded, amplitude limiting is performed to ensure the safety and reliability of subsequent mapping input.

[0091] Loading the energy-saving fluid amplifier calibration parameters and applying a gain mapping to the inlet static pressure value generates a continuously adjustable total wind speed signal. Specifically, the energy-saving fluid amplifier calibration parameters, including diffusion gain, exponential gain, and flow channel adjustment index, are loaded. Applying the calibration gain mapping to the pre-processed static pressure value generates a continuously adjustable total wind speed signal. The total wind speed signal is converted into a fan drive instruction through a mapping table or algorithm, and is issued through the vehicle control bus, so that the fan accurately outputs the corresponding wind speed.

[0092] The total wind speed signal is converted into a drive instruction, which is sent down through the control bus to adjust the fan output. Specifically, the partition base wind speed share is calculated according to the fan blade optimization coefficient and system allocation index of each partition, reflecting the air supply priority of different areas.

[0093] A partitioned reference wind speed signal is generated based on the partitioned blade optimization coefficient and the partitioned allocation index, and then corrected using the regional temperature difference adjustment factor. Specifically, the temperature difference between the glass surface and the cabin of each partition is read, and the temperature difference adjustment factor is applied to incrementally correct the reference share to enhance the air supply response in high-fogging areas.

[0094] Normalize the corrected partition reference wind speed signal and distribute the total wind speed signal according to the partition ratio to obtain the partition wind speed signal. Specifically, sum and normalize all the corrected partition reference shares to ensure that the total distribution ratio is 1; distribute the total wind speed signal to each partition according to the normalized ratio to obtain the final partition wind speed instruction; convert the partition wind speed instruction into the opening degree of each air outlet actuator or fan speed control command to complete the air supply zoning;

[0095] The present invention is further configured such that the generation logic of the total wind speed signal is: F out (t) is the total wind speed signal at time t, Γ exp is the diffusion gain coefficient, λ amp is the exponential growth coefficient, P in (t) is the inlet static pressure value at time t, φ chan is the flow channel adjustment index; specifically, by exponentially increasing the inlet static pressure and combining it with the diffusion gain, and then adjusting the flow channel index, the continuous, smooth amplification and controllable output of the total wind speed signal are achieved, providing a stable and reliable airflow benchmark for subsequent zoned air volume distribution. In the calculation logic, the real-time static pressure value P output by the pipeline pressure sensor is converted into in (t) is used as the basic pneumatic drive signal for P in (t) applies an exponential increase in the form of The exponential growth factor λ amp Controlling the steepness of the static pressure to flow rate amplification curve can adapt to the amplification requirements under different working conditions; the exponential term and the diffusion gain coefficient Γ exp Multiply them together to simulate the secondary pressurization of the airflow by the vortex and Venturi effect in the energy-saving fluid amplifier; and then apply the flow channel adjustment index φ to the above product as a whole. chan The power operation is used to fine-tune the sensitivity and saturation characteristics of different wind speed ranges, so that both low and high wind speed sections can obtain smooth and controllable output changes, and the total wind speed signal F is obtained. out (t), as the reference input for partition air volume distribution; diffusion gain coefficient Γ exp According to the amplifier geometry and experimental wind tunnel data calibration, the value range is [1.0,5.0], the exponential amplification coefficient λ amp According to the static pressure change range and the required wind speed output range, the value range is [0.01, 1.0], the flow channel adjustment index φchan Used to balance the response of low wind and high wind sections, the value range is [0.5, 2.0];

[0096] The generation logic of the partition reference wind speed signal is: B i (t) is the reference wind speed signal of partition i at time t, Ψ leaf (i) is the partition blade optimization coefficient of partition i, δ sys Assign an index to the system; specifically, by exponentially scaling the total wind speed signal and combining it with the partition blade optimization coefficient, the basic wind speed share of each partition is generated to reflect the difference in air supply efficiency and priority in different areas, laying the foundation for subsequent temperature difference correction and normalization allocation; the total wind speed signal F out (t) Apply system distribution index δ sys The power operation of By adjusting δ sys , the concentration of the controllable wind speed contribution to the partition air volume benchmark share: δ sys When <1, the high-speed segment share difference is weakened and tends to be evenly distributed. sys When Ψ > 1, the regional priority of the high-speed section is strengthened to highlight the key partitions when there is high traffic demand; the total wind speed term after exponential scaling is combined with the blade optimization coefficient Ψ of partition i leaf (i) Multiplication, Ψ leaf (i) Reflects the differences in blade geometry and aerodynamic efficiency of the partitioned air outlet, allowing the high-efficiency area to obtain a larger share of basic air volume; ultimately, the partitioned reference wind speed signal is obtained, which serves as the initial input for subsequent temperature difference increment correction and normalization allocation; the partitioned blade optimization coefficient Ψ of partition i leaf (i) Calibrated according to the experimental data of blade efficiency, the value range is [0.5, 2.0], the system allocation index δ sys A nonlinear index used to control the impact of total wind speed on the partition benchmark share, adjust the concentration of air volume distribution between partitions, and optimize based on the system's requirements for uniformity and concentration. The value range is [0.8, 1.5];

[0097] The logic for correcting the partition reference wind speed signal is: B' i (t) = B i (t)×[1+θzone×ΔT i (t)],B' j (t) is the modified zone reference wind speed signal, θzone is the temperature difference adjustment factor, ΔT i (t) is the temperature difference between the glass surface and the cabin of partition i at time t; specifically, at the uncorrected partition reference wind speed B i (t), increase the temperature difference ΔT between the partitions i(t) is proportional to the incremental term, completing the dynamic response of air volume to temperature difference demand; multiplying the partition temperature difference by the temperature difference adjustment factor θ zone , get the incremental ratio of the wind speed benchmark for each partition, add one to the incremental ratio and multiply it with the benchmark signal to generate the corrected partition benchmark wind speed signal, ensuring that no additional air volume is introduced when the temperature difference is zero, and the wind speed benchmark is linearly amplified when the temperature difference increases; temperature difference adjustment factor θ zone Defines the proportional coefficient of the effect of temperature difference on air volume correction, which is used to control the sensitivity of air volume to temperature difference changes. It is set based on the glass area and air outlet layout of different vehicle models, and the trade-off between energy efficiency and defogger speed. The value range is [0.05, 0.3];

[0098] The generation logic of the normalized partition wind speed signal is: U i (t) is the wind speed signal of partition i at time t, S adj (t) is the corrected partition reference wind speed signal and, N zone Specifically, the wind speed distribution ratio of each partition is generated by normalizing the sum of the corrected partition reference wind speed signals, and then the ratio is multiplied by the continuously adjustable total wind speed signal to obtain the actual partition wind speed instruction, ensuring that the sum of the wind speeds of each partition is equal to the total wind speed while taking into account the regional priority; the corrected reference wind speed B' of all partitions is i (t) Accumulate and get the normalized denominator S adj (t), which represents the total amount of the corrected reference wind speed of each partition under the current wind volume and environmental requirements; for each partition i, its corrected reference wind speed B' i (t) and the sum S adj (t) ratio, calculate the wind speed distribution ratio of the partition Ensure that the sum of the distribution ratios of all partitions is 1, so that the wind speed distribution is reasonable and standardized in terms of proportion; multiply the normalized ratio by the total wind speed signal F out (t), the actual partition wind speed signal is obtained, which is used as the opening or wind speed control command sent to the partition air outlet actuator to ensure that the total partition wind speed is equal to the system output wind speed.

[0099] The present invention is further configured such that step S4 includes:

[0100] Aggregate historical environmental data and real-time sensor data, and divide the historical series and real-time series into fixed time windows. Specifically, synchronously obtain historical environmental data such as outdoor temperature, humidity, wind speed, wind direction, and vehicle speed from the vehicle database and sensor bus, and collect the latest readings of each sensor in real time. According to the preset time window length, divide the historical data into fixed-length historical series, and use the sliding window composed of the latest readings as the real-time series, ensuring that both historical and real-time information can be used in the prediction.

[0101] Normalization and differencing preprocessing are performed on each time series to eliminate dimensional differences and trend drift. Specifically, normalization is applied to the time series data of each channel to map the data to a unified numerical range to eliminate the influence of different dimensions. Difference operations are performed on the normalized series to remove long-term trends and seasonal drift and highlight short-term change characteristics.

[0102] Construct a multi-channel input tensor by data category and load the pre-trained multi-layer LSTM network model structure and parameters. Specifically, stack each channel sequence according to the data category (temperature, humidity, wind speed, wind direction, vehicle speed, etc.) to construct a network input tensor of the form [number of channels × time step]. Load the offline pre-trained multi-layer LSTM network architecture and its weight parameters to ensure consistency between the online prediction and offline training environments.

[0103] The processed historical sequence is forward-calculated through a multi-layer LSTM network to extract the sequence time series feature vector. Specifically, the pre-processed history tensor is input into the multi-layer LSTM network, and the long-term and short-term dependencies are captured through the gating units of each layer, and the hidden state vector at the last moment is output as the time series feature.

[0104] The real-time sequence is appended to the end of the historical sequence to form the prediction input, and the risk mapping value within the preset lead time in the future is obtained at the LSTM output layer. Specifically, the real-time sequence generated by sliding is appended to the end of the historical sequence to form a new input sequence, which is fed into the same LSTM network for forward calculation. At the output layer, the hidden state at the last moment is converted into the original risk mapping value within the preset lead time in the future through a fully connected mapping.

[0105] The mapped values ​​are adjusted using a nonlinear saturation function and a time-weighted coefficient to generate a normalized fog risk score. Specifically, a nonlinear saturation function is applied to the risk mapped values ​​to suppress extreme values ​​and enhance responses in medium- and high-risk segments. Time-weighted coefficients are assigned to the mapped values ​​based on different lead times to balance the importance of short- and medium-term forecasts. Finally, a normalized fog risk score is generated, limited to the range [0, 1], for use in downstream control logic.

[0106] The implementation logic of the above-mentioned fogging risk scoring is existing technology and will not be elaborated here. Through the unified management and preprocessing of historical and real-time environmental data, combined with a multi-channel LSTM time series network, key time series features are extracted and risk mapping and weighting are performed. A quantitative assessment of the fogging risk within a preset lead time is formed, providing a decision-making basis for preventive thermal field compensation.

[0107] The present invention is further configured such that step S5 includes:

[0108] The atomization risk score output by the time series prediction module is read and converted into a thermal field compensation coefficient according to the preset control gain and saturation mapping function. Specifically, the normalized atomization risk score of each partition is obtained from the time series prediction module. The risk score is amplified or compressed according to the preset control gain to adjust the heating response intensity. The gain-adjusted score is converted into a thermal field compensation coefficient through the saturation mapping function to ensure that the compensation coefficient changes smoothly within a reasonable range and does not produce a sudden response.

[0109] The thermal field compensation coefficient is mapped to the target power output of the partition air outlet actuator, and the target power output is executed to compensate the output heat to achieve partitioned defrosting and defogger. Specifically, the thermal field compensation coefficient is mapped to the target power output of the partition air outlet actuator, and the partition electric heating power target is generated, and the incremental heating of high-risk areas is achieved in combination with the partition priority; this is achieved here through historical statistical data, which is a prior art and will not be described in detail; according to the characteristic curve of the electric heating element (PtC semiconductor), the power target is mapped to the corresponding current or voltage drive signal; the drive signal is packaged and sent to the electric heating element control unit of each partition through the on-board control bus; the control unit adjusts the output power according to the instruction, accurately controls the temperature of the heating wire, and realizes real-time compensation of the heating amount of each partition of the windshield;

[0110] Through real-time mapping and power allocation of fogging risk scores, the prediction results are converted into heating instructions, and the electric heating elements are driven in a zoned manner to ensure early release of heat in high-risk areas, achieving fast and energy-saving zoned defrosting and demisting.

[0111] The present invention is further configured such that the calculation logic of the thermal field compensation instruction is: is the thermal field compensation coefficient of partition i at time t, ξ ctrl To control the gain, is the fogging risk score of partition i at time t, σ ctrl is the saturation mapping index; specifically, by applying nonlinear saturation mapping and control gain amplification to the partition atomization risk score, the risk degree is converted into a thermal field compensation coefficient to control the intensity of the heating power, ensuring that high-risk areas obtain more heat compensation and low-risk areas reduce energy consumption; read the atomization risk score R of partition i i (t), ranging from [0,1], reflects the intensity of the fogging trend in the area; the risk score is exponentially operated and then added to itself for ratio mapping: where the saturation mapping index σ ctrl Adjust the steepness of the mapping curve to make the response in the medium and high risk areas more obvious; multiply the saturation mapping result by the control gain ξ ctrl , and obtain the thermal field compensation coefficient Determines the overall heating intensity and can be adjusted according to energy consumption and response speed requirements to control gain ξ ctrl The coefficient for linearly amplifying the mapping results is used to adjust the overall sensitivity of the heating system to risk changes. The value range is [0.5, 2.0] according to the power limit and energy consumption target of the vehicle electric heating system. The saturation mapping index σ ctrl The shape of the risk score-to-compensation coefficient mapping curve is controlled. The larger the exponent, the more pronounced the mapping amplifies high risks. The curve is optimized based on the desired steepness and response speed, with a value range of [2,6]. Through nonlinear saturation mapping and exponential operations, a refined response to different risk levels is achieved, avoiding overheating in low-risk areas. By adjusting the heating intensity through gain control, flexible and controllable heat distribution is achieved, balancing rapid defogger speed with optimal energy consumption. Through zoned compensation coefficient output, high-risk areas are prioritized for heat compensation, improving overall defrosting and defogging efficiency and comfort.

[0112] Example 2

[0113] See also Figure 2 The exemplary vehicle-mounted defrosting and defogging intelligent control system with dynamic adjustment of partitions is used to implement the above-mentioned vehicle-mounted defrosting and defogging intelligent control method with dynamic adjustment of partitions, including:

[0114] Initial module: Initializes and activates the dual-camera visual sensor array, infrared temperature field detection device, duct pressure sensor, cabin temperature and humidity sensor, and partitioned air vent actuator;

[0115] Fusion module: Performs multimodal fusion processing on the dual-camera visible light images and infrared temperature data to obtain characteristic indicators of condensation and ice crystals on the partitioned glass surface;

[0116] Distribution module: Based on the static pressure of the inlet airflow, a continuously adjustable total wind speed signal is generated through an energy-saving fluid amplifier. Based on the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin, the total wind speed signal is dynamically normalized and distributed according to the preset distribution index to obtain the partitioned wind speed signal;

[0117] Prediction module: Builds a time series prediction model based on historical and real-time environmental data, evaluates the atomization risk within a preset lead time in the future, and outputs a atomization risk score;

[0118] Compensation module: Generates thermal field compensation coefficients based on the predicted fogging risk score mapping and sends them to the partitioned air outlet actuators to complete partitioned defrosting and defogging.

[0119] It should be noted that the intelligent control system for vehicle-mounted defrost and defogger with dynamic adjustment of partitions provided in the above-mentioned embodiment and the intelligent control method for vehicle-mounted defrost and defogger with dynamic adjustment of partitions provided in the above-mentioned embodiment belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiment and will not be repeated here. In actual applications, the intelligent control system for vehicle-mounted defrost and defogger with dynamic adjustment of partitions provided in the above-mentioned embodiment can, as needed, allocate the above-mentioned functions to different functional modules, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0120] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0121] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0122] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0123] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0124] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0125] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0126] In the several embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0127] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0128] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0129] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0130] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. Intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones, characterized in that: include: S1: Initialize and activate the dual-camera visual sensor array, infrared temperature field detection device, duct pressure sensor, cabin temperature and humidity sensor, and partition air vent actuator; S2: Perform multimodal fusion processing on the dual-camera visible light images and infrared temperature data to obtain the characteristic indicators of condensation and ice crystals on the partitioned glass surface; S3: Based on the static pressure of the inlet airflow, a continuously adjustable total wind speed signal is generated through an energy-saving fluid amplifier. According to the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin, the total wind speed signal is dynamically normalized and distributed according to the preset distribution index to obtain the partitioned wind speed signal; S4: Build a time series prediction model based on historical and real-time environmental data, evaluate the fogging risk within a preset lead time in the future, and output a fogging risk score; S5: Generate thermal field compensation coefficient based on the predicted fogging risk score mapping and send it to the partitioned air outlet actuator to complete the partitioned defrosting and defogging.

2. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 1 is characterized in that: Step S1 includes: Initialize the dual-camera visual sensor array communication interface, load the camera calibration parameters and distortion correction matrix, and perform a self-check process to confirm that the image acquisition channel is complete; Activate the infrared temperature field detection device, read the built-in temperature calibration point, calibrate the temperature drift and generate the initial temperature field baseline diagram; Enable the pipeline pressure sensor, clear the output value and verify the pressure response characteristics, and confirm the range and linearity by injecting calibration gas source; Start the cabin temperature and humidity sensor module, collect the initial values ​​of ambient temperature and humidity, calibrate the sensor bias, and store the reference humidity curve; Establish a control bus connection for the partitioned air outlet actuator, send speed measurement commands in sequence, monitor the air outlet feedback position, and confirm the actuator's motion range and response delay; Synchronize the trigger timestamp, synchronize the signal channel data stream, complete the initialization status confirmation and switch to the online monitoring mode.

3. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 1 is characterized in that: Step S2 includes: Calibrate the internal and external parameters of the dual cameras and infrared module to obtain the camera distortion matrix and temperature mapping lookup table; Perform dedistortion, color correction, and white balance processing on the visible light original image, and output a geometrically corrected high-fidelity image; Perform temperature drift calibration and linear interpolation on the infrared raw data to generate the absolute temperature field map corresponding to each pixel; Using feature point matching or mutual information maximization methods, the registration transformation matrix from the visible light image to the infrared temperature field is calculated, and the infrared image is resampled and aligned to the visible light coordinate system; According to the preset partition outline, the registered visible light and infrared images are synchronously divided into blocks to extract the partitioned regions of interest; Multi-scale Laplace enhancement and Gabor filter bank are applied to the partitioned region of interest to extract the edge response value of the fine water film and calculate the average temperature of the partitioned area. After nonlinear mapping and normalization processing of the edge response value of the micro-water film and the average temperature of the partition, the condensation and ice crystal fusion characteristics of the partitioned glass surface are generated.

4. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 3 is characterized in that: The generation logic of the condensation and ice crystal fusion characteristics on the partitioned glass surface is as follows: C i (t) is the condensation and ice crystal fusion characteristics of the glass surface in partition i at time t, S i (t) is the edge response value of the micro-water film in partition i at time t, T i (t) is the average temperature of partition i at time t, η fus is the visual fusion coefficient, κ ir is the infrared temperature sensitivity coefficient, ψ ice is the ice crystal response power exponent.

5. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 1 is characterized in that: Step S3 includes: The inlet static pressure value is obtained according to the preset sampling frequency, and is processed by low-pass filtering, and the filtered static pressure value is range-verified and limited; Load the calibration parameters of the energy-saving fluid amplifier and apply a gain mapping to the inlet static pressure value to generate a continuously adjustable total wind speed signal; Convert the total wind speed signal into a drive command, send it through the control bus, and adjust the fan output; The partition reference wind speed signal is generated based on the partition blade optimization coefficient and the partition allocation index, and the partition reference wind speed signal is corrected in combination with the regional temperature difference adjustment factor; The corrected partition reference wind speed signal is normalized, and the total wind speed signal is distributed according to the partition ratio to obtain the partition wind speed signal.

6. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 5 is characterized in that: The generation logic of the total wind speed signal is: F out (t) is the total wind speed signal at time t, Γ exp is the diffusion gain coefficient, λ amp is the exponential growth coefficient, P in (t) is the inlet static pressure value at time t, φ chan is the flow channel adjustment index; The generation logic of the partition reference wind speed signal is: B i (t) is the reference wind speed signal of partition i at time t, Ψ leaf (i) is the partition blade optimization coefficient of partition i, δ sys Assign an index to the system; The logic for correcting the partition reference wind speed signal is: B' i (t) = B i (t)×[1+θ zone ×ΔT i (t)],B′ j (t) is the modified partition reference wind speed signal, θ zone is the temperature difference adjustment factor, ΔT i (t) is the temperature difference between the glass surface and the cabin of partition i at time t; The generation logic of the normalized partition wind speed signal is: U i (t) is the wind speed signal of partition i at time t, S adj (t) is the corrected partition reference wind speed signal and, N zone is the total number of partitions.

7. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 1 is characterized in that: Step S4 includes: Aggregate historical environmental data and real-time sensor data, and divide historical series and real-time series into fixed time windows; Normalization and difference preprocessing are performed on each time series to eliminate dimension differences and trend drift; Construct multi-channel input tensors by data category and load the pre-trained multi-layer LSTM network model structure and parameters; The processed historical sequence is forward calculated through a multi-layer LSTM network to extract the sequence time series feature vector; Append the real-time sequence to the end of the historical sequence to form the prediction input, and obtain the risk mapping value within the future preset lead time at the LSTM output layer; The mapped values ​​are adjusted by a nonlinear saturation function and a time weighting coefficient to generate a normalized atomization risk score.

8. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 1, characterized in that: Step S5 includes: Read the atomization risk score output by the time series prediction module, and convert the atomization risk score into a thermal field compensation coefficient according to the preset control gain and saturation mapping function; The thermal field compensation coefficient is mapped to the target power output of the partition air outlet actuator, and the target power output is executed to compensate the output heat to achieve partitioned defrosting and demisting.

9. The intelligent control method for vehicle-mounted defrosting and defogging with dynamic adjustment of zones according to claim 8, characterized in that: The calculation logic of the thermal field compensation instruction is: is the thermal field compensation coefficient of partition i at time t, ξ ctrl is the control gain, R i (t) is the fogging risk score of partition i at time t, σ ctrl is the saturation mapping index.

10. An intelligent vehicle defrosting and defogging control system with dynamic adjustment of partitions, used to implement the intelligent vehicle defrosting and defogging control method with dynamic adjustment of partitions according to any one of claims 1 to 9, characterized in that: include: Initial module: Initializes and activates the dual-camera visual sensor array, infrared temperature field detection device, duct pressure sensor, cabin temperature and humidity sensor, and partitioned air vent actuator; Fusion module: Performs multimodal fusion processing on the dual-camera visible light images and infrared temperature data to obtain characteristic indicators of condensation and ice crystals on the partitioned glass surface; Distribution module: Based on the static pressure of the inlet airflow, a continuously adjustable total wind speed signal is generated through an energy-saving fluid amplifier. Based on the partitioned fan blade optimization coefficient and the temperature difference between the glass surface and the cabin, the total wind speed signal is dynamically normalized and distributed according to the preset distribution index to obtain the partitioned wind speed signal; Prediction module: Builds a time series prediction model based on historical and real-time environmental data, evaluates the atomization risk within a preset lead time in the future, and outputs a atomization risk score; Compensation module: Generates thermal field compensation coefficients based on the predicted fogging risk score mapping and sends them to the partitioned air outlet actuators to complete partitioned defrosting and defogging.

Images