Seat ventilation parameter prediction and control method and system based on dynamic environment perception

Abstract

The invention discloses a seat ventilation parameter prediction and control method and system based on dynamic environment perception, and relates to the technical field of data processing.The method comprises the steps that firstly, multi-source asynchronous sensor information of all seats of a target vehicle is obtained, and ventilation operation accumulation areas of all the seats are recognized through a space-time alignment and judgment method; meanwhile, the heat dissipation effective state of each accumulation area is determined and updated in real time by combining the heat dissipation rate and the air volume gear historical record; on this basis, ventilation parameters are preliminarily adjusted, then a future ventilation demand is generated through a ventilation demand model driven by short-time prediction, hardware constraint matching is carried out in combination with the actual executable range of a fan, temperature and humidity state changes of passengers and environmental disturbance can be responded in a short time, and the riding comfort is improved; meanwhile, through combination of feedforward control and hardware constraint matching, air volume overshoot or lag is avoided, safe and reliable operation of a ventilation system is ensured, and the seat temperature and humidity adjusting efficiency is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for predicting and controlling seat ventilation parameters based on dynamic environmental perception. Background Technology

[0002] In modern automotive cabin environments, the comfort and responsiveness of seat ventilation systems are receiving increasing attention from users. With the widespread adoption of smart cockpits and personalized riding experiences, seat ventilation systems need to be able to adjust airflow in real-time and precisely according to changes in the in-vehicle environment and vehicle operating conditions to ensure riding comfort, temperature and humidity balance, and localized physical experience. Meanwhile, the continuous changes in the internal and external environmental conditions of the vehicle cause the airflow and heat transfer inside the seat to exhibit complex nonlinear spatiotemporal dynamic characteristics. Furthermore, the complex porous structure of the seat means that airflow is affected by the resistance of the porous medium, local channel morphology, thermal coupling effects, and heat exchange between the seat surface and the occupant's skin. This results in a high degree of non-uniformity and strong spatial dependence in the internal airflow distribution and local pressure. Traditional ventilation strategies, based on single sensors, fixed rules, or linear control, struggle to accurately capture these local flow characteristics and cannot effectively adapt to dynamically changing environmental conditions. Dynamic environmental perception refers to the system's ability to acquire, analyze, and understand the state and changes of the surrounding environment in real time. In automotive cabin seat ventilation, it involves not only measuring static environmental parameters, such as interior temperature, humidity, or seat surface temperature, but also sensing the characteristics of environmental changes over time. This allows control strategies to be flexibly adjusted according to environmental changes, rather than relying solely on pre-set static parameters or fixed models, thereby achieving more precise, comfortable, and efficient seat ventilation control.

[0003] For example, CN120197559B discloses a method, device, and computer for determining automotive seat ventilation parameters. The method includes: acquiring three-dimensional contour information of a target object on an automotive seat, the object's physiological information, and in-vehicle environmental information; performing comfort analysis based on the object's physiological information and the in-vehicle environmental information to determine comfortable physiological information; using the comfortable physiological information as a target condition, constructing ventilation duct information for the automotive seat based on the three-dimensional contour information; inputting the comfortable physiological information and ventilation duct information into a fan control model to obtain initial fan operating parameters for each fan; performing ventilation simulation based on the initial fan operating parameters to obtain fan operating information for each fan; for any given fan, optimizing the initial fan operating parameters for energy saving based on the fan operating information to obtain target fan operating parameters; and fusing the target fan operating parameters to obtain the automotive seat ventilation parameters.

[0004] Current control strategies typically combine occupant physiological states with environmental conditions to conduct comfort analysis, determine target comfort physiological information, generate target operating parameters for each fan, and then fuse these parameters to form overall ventilation parameters for the car seat, achieving preliminary control and comfort management of seat airflow. However, in practical applications, this method still faces challenges. The airflow distribution inside the seat is highly non-uniform due to the resistance of porous media, local channel shape, and thermal coupling effects. Existing technologies rely on fixed duct models and fan simulation parameters, making it difficult to accurately reflect local flow field characteristics. Secondly, occupant states and in-vehicle environmental conditions are constantly changing. Therefore, existing control strategies lack the ability to respond quickly to dynamic environmental changes and lack efficient data fusion and dynamic adjustment mechanisms, limiting the intelligence and adaptability of the seat ventilation system under complex operating conditions. Summary of the Invention

[0005] This invention provides a method and system for predicting and controlling seat ventilation parameters based on dynamic environmental perception. The technical solution provided by this application is as follows: According to a first aspect of this application, a method for predicting and controlling seat ventilation parameters based on dynamic environment perception is provided, the method comprising: The system acquires multi-source asynchronous information from the target vehicle's sensors and performs spatiotemporal alignment determination to obtain the ventilation operation accumulation area of ​​each seat in the target vehicle. Simultaneously, based on the correspondence between temperature change and airflow level records, it obtains the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle and updates the ventilation operation accumulation area of ​​each seat in the target vehicle.

[0006] Based on the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, the active collection and compensation inference of the sensor triggers the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle to be initially adjusted.

[0007] Based on the ventilation demand inference model driven by short-term prediction, the ventilation demand prediction results of the target vehicle are generated, and the ventilation demand prediction results of the target vehicle are dynamically monitored and compared to implement feedforward control of the future ventilation intensity of each accumulation area of ​​the target vehicle.

[0008] Generate an executable airflow control signal and obtain the fan's executable range for the target vehicle, then perform hardware constraint matching on the executable airflow control signal.

[0009] According to another aspect of this application, a seat ventilation parameter prediction and control system based on dynamic environment perception is provided, including a multi-source sensing and accumulation area determination module, which is used to acquire multi-source asynchronous information from the target vehicle's sensors and perform spatiotemporal alignment determination to obtain the ventilation operation accumulation area of ​​each seat in the target vehicle. At the same time, based on the correspondence between temperature change and airflow level records, the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle is obtained, and the ventilation operation accumulation area of ​​each seat in the target vehicle is updated.

[0010] The active data acquisition and ventilation parameter compensation module is used to trigger the active data acquisition and compensation inference of the sensors based on the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, and to make preliminary adjustments to the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle.

[0011] The short-term prediction-driven feedforward control module is used to generate ventilation demand prediction results for the target vehicle based on the short-term prediction-driven ventilation demand inference model, and to dynamically monitor and compare the ventilation demand prediction results of the target vehicle to perform feedforward control on the future ventilation intensity of each accumulated area of ​​the target vehicle.

[0012] The air volume control signal generation and hardware constraint matching module is used to generate executable air volume control signals, obtain the fan's executable range for the target vehicle, and perform hardware constraint matching on the executable air volume control signals.

[0013] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: (1) Obtain multi-source asynchronous sensor information of each seat in the target vehicle, and identify the ventilation operation accumulation area of ​​each seat through spatiotemporal alignment and judgment method. At the same time, combine the heat dissipation rate and the historical record of air volume level to determine the effective heat dissipation status of each accumulation area and update it in real time. On this basis, the ventilation parameters are initially adjusted. Then, the future ventilation demand is generated through the ventilation demand model driven by short-term prediction to generate an executable air volume control signal. Combine the actual executable range of the fan to perform hardware constraint matching. It can respond to changes in the temperature and humidity of the occupants and environmental disturbances in a short time and improve the riding comfort. At the same time, by combining feedforward control and hardware constraint matching, air volume overshoot or lag is avoided, ensuring the safe and reliable operation of the ventilation system and effectively improving the seat temperature and humidity regulation efficiency.

[0014] (2) By establishing a time-series confidence window, the multi-source asynchronous sensor information of each seat in the target vehicle is obtained. Combined with the current ambient temperature and solar radiation intensity, the seat area is dynamically marked as an accumulation area with high heat load or humidity retention. At the same time, based on the historical air volume level and temperature change records, the effective heat dissipation status of each accumulation area is judged. For areas with insufficient heat dissipation, the heat load or humidity retention status is further identified, thereby obtaining the ventilation operation accumulation area of ​​each seat area. Based on dynamic environmental perception and changes in the temperature and humidity status of the occupants, the ventilation operation accumulation area can be accurately identified, providing a precise basis for subsequent temperature and humidity gradient acquisition, compensation inference and ventilation parameter prediction. This ensures that the seat ventilation system can respond to environmental disturbances and occupant needs in real time and accurately, improve riding comfort and optimize ventilation efficiency.

[0015] (3) The ventilation operation accumulation area of ​​each seat in the target vehicle is dynamically updated within the time-series confidence window. At the same time, the disturbance increment is calculated by combining the temperature and humidity interaction of the neighborhood to obtain the temperature and humidity disturbance increment of each area, and the length of the time-series confidence window is dynamically adjusted accordingly. On this basis, temperature and humidity compensation inference values ​​are further generated, and the ventilation parameters of each seat area are initially adjusted according to these compensation inference values, thereby forming an updated ventilation operation accumulation area. This area can reflect the temperature and humidity status of the occupants and the changes in local environmental disturbances in real time, improve the identification and response capabilities of high heat load and humid retention areas, provide accurate input for subsequent ventilation demand prediction and feedforward control, and ensure that the seat ventilation system can quickly and accurately adjust the air volume and wind speed in a dynamic environment, so as to achieve a synergistic improvement in occupant comfort and energy efficiency optimization.

[0016] (4) After preliminary parameter adjustment of the ventilation operation accumulation areas of each seat in the target vehicle, the temperature gradient sequence of the high heat load area, the humidity gradient sequence of the humid retention area, and the local environmental disturbance characteristics within the time confidence window are input into the lightweight random forest model. The ventilation demand prediction results of each accumulation area are output, and the air volume, air speed, and ventilation duration of each accumulation area are controlled in a feedforward manner. The temperature and humidity status is collected in real time during each monitoring cycle to dynamically correct the future ventilation intensity. Finally, the system generates an executable air volume control signal by combining the key performance parameters of the fan, and performs hardware constraint matching on the instantaneous change of the predicted air volume to ensure that the issued air volume command is within the executable range of the fan. This achieves accurate prediction and feedforward adjustment of the ventilation demand of each area of ​​the seat, which can respond in advance to changes in the temperature and humidity status of the occupants, improve the local comfort of the seat, prevent excessive or insufficient temperature and humidity, and ensure the safety and controllability of the fan operation. Thus, the seat ventilation performance and occupant experience are optimized under dynamic environmental perception.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0018] The accompanying drawings are provided for a better understanding of this solution and do not constitute a limitation of this application. Wherein: Figure 1 This is a flowchart of the seat ventilation parameter prediction and control method based on dynamic environment perception provided in the embodiments of the present invention; Figure 2 This is a flowchart of the multi-source sensor information acquisition and ventilation operation accumulation area update provided in the embodiments of the present invention; Figure 3 This is a flowchart of seat ventilation demand prediction and feedforward control based on dynamic environment perception provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the lightweight random forest model provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of a seat ventilation parameter prediction and control system module based on dynamic environment perception provided in an embodiment of the present invention. Detailed Implementation

[0019] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0020] This invention provides a method for predicting and controlling seat ventilation parameters based on dynamic environmental perception, such as... Figure 1 The flowchart shown is for a method of predicting and controlling seat ventilation parameters based on dynamic environment perception. The processing flow of this method may include the following steps: The system acquires multi-source asynchronous information from the target vehicle's sensors and performs spatiotemporal alignment determination to obtain the ventilation operation accumulation area of ​​each seat in the target vehicle. Simultaneously, based on the correspondence between temperature change and airflow level records, it obtains the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle and updates the ventilation operation accumulation area of ​​each seat in the target vehicle.

[0021] Based on the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, the active collection and compensation inference of the sensor triggers the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle to be initially adjusted.

[0022] Based on the ventilation demand inference model driven by short-term prediction, the ventilation demand prediction results of the target vehicle are generated, and the ventilation demand prediction results of the target vehicle are dynamically monitored and compared to implement feedforward control of the future ventilation intensity of each accumulation area of ​​the target vehicle.

[0023] Generate an executable airflow control signal and obtain the fan's executable range for the target vehicle, then perform hardware constraint matching on the executable airflow control signal.

[0024] It's important to note that acquiring multi-source asynchronous information from the target vehicle's sensors and performing spatiotemporal alignment involves integrating environmental and passenger status data from different sensors, sampling frequencies, and trigger times into a continuous time sequence that can be used to determine the accumulated areas of each seat ventilation system. First, asynchronously uploaded data streams are received from the pressure-sensitive matrix, local humidity patch, seat temperature sensor, in-vehicle air duct temperature sensor, solar radiation sensor, and in-vehicle ambient temperature and humidity sensor. These data streams typically exhibit differences in sampling rates, upload latency, and trigger modes. Therefore, a unified time reference, such as the vehicle's central clock or controller timestamp, is needed to map these asynchronous data onto the same timeline. Interpolation, synchronization window matching, and event alignment are then used to compensate for the time gaps caused by different frequencies, ensuring that each sensor has corresponding valid data in the same time slice.

[0025] It should be noted that the target vehicle's seating area is divided into four zones for each seat: the seat cushion is divided into front and rear zones to reflect the differences in heat and ventilation between the thighs and ischial tuberosities, respectively; the backrest is divided into upper and lower zones to reflect the differences in heat and contact between the shoulders and lower back, respectively. This four-zone division allows for the capture of the temperature and humidity differences most critical for ventilation control, without introducing a high-resolution pressure-sensitive matrix, utilizing existing or low-cost scalable zone pressure sensors, several surface temperature points, and local humidity sensors.

[0026] like Figure 2 As shown, Figure 2 The multi-source sensor information acquisition and ventilation operation accumulation area update flowchart provided in this embodiment of the invention describes the process of acquiring asynchronous temperature, humidity, and environmental data from sensors at each seat of the vehicle, generating a preliminary ventilation operation accumulation area through spatiotemporal alignment determination, and judging the effective heat dissipation status of each area based on heat dissipation rate and airflow level records. Simultaneously, within a time-series confidence window, the incremental temperature and humidity disturbances in high heat load and humid retention areas are continuously monitored and updated, thereby providing dynamic and accurate regional information for subsequent preliminary adjustment and prediction of ventilation parameters. The generation of the preliminary ventilation operation accumulation area through spatiotemporal alignment determination includes obtaining the ventilation operation accumulation area for each seat of the target vehicle, the specific process of which is as follows: A time-series confidence window is established based on the sensor sampling delay. Multi-source asynchronous information from the target vehicle's sensors is extracted from the time-series confidence window, including the seat surface temperature, heat dissipation rate, absolute humidity change rate, and cabin relative humidity change rate of each seat area of ​​the target vehicle.

[0027] Extract the surface temperature of the seats in each seat area of ​​the target vehicle and collect the current ambient temperature and current solar radiation intensity.

[0028] The current solar radiation intensity is compared with the solar radiation threshold stored in the database, and the surface temperature of the seats in each seat area of ​​the target vehicle is compared with the current ambient temperature.

[0029] If the surface temperature of the seat in each seat area of ​​the target vehicle is higher than or equal to the current ambient temperature and the solar radiation intensity is higher than or equal to the solar radiation threshold, then the seat area of ​​the target vehicle is marked as the first heat accumulation area, thus obtaining each first heat accumulation area; otherwise, no marking is required. The absolute humidity change rate of each seat area in the target vehicle is compared with the relative humidity change rate of the vehicle cabin. If the absolute humidity change rate of each seat area in the target vehicle is higher than or equal to the relative humidity change rate of the vehicle cabin, then that seat area in the target vehicle is marked as the second moisture accumulation area, thus obtaining each second moisture accumulation area. Otherwise, no marking is required.

[0030] Specifically, based on the correspondence between temperature changes and recorded airflow levels, the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle is obtained. The specific process is as follows: Obtain the ventilation setting of the target vehicle and calculate the temperature change per unit time for each of the first heat accumulation areas and each of the second moisture accumulation areas of the target vehicle at that ventilation setting.

[0031] If the temperature change of any accumulated area within a unit of time at the ventilation setting is lower than or equal to the temperature change threshold stored in the database, then the accumulated area is further marked as an area with insufficient heat dissipation; otherwise, no marking is required.

[0032] If a certain accumulation area is simultaneously marked as the first heat accumulation area and the insufficient heat dissipation area, then the effective heat dissipation state of that area is determined to be a high heat load state.

[0033] If a certain accumulation area is simultaneously marked as a second moisture accumulation area and a heat dissipation inadequate area, then the effective heat dissipation state of that area is determined to be a moisture retention state.

[0034] Specifically, the accumulated ventilation operation areas of each seat in the target vehicle are updated. The specific process is as follows: Within a time-series confidence window, several monitoring periods are preset to acquire local environmental disturbance characteristic data of each first heat accumulation area and each second moisture accumulation area of ​​the target vehicle. The local environmental disturbance characteristic data of each first heat accumulation area of ​​the target vehicle includes the short-time change rate of the surface temperature of each first heat accumulation area and the temperature gradient sequence of each adjacent first heat accumulation area. The local environmental disturbance characteristic data of each second moisture accumulation area includes the absolute humidity change rate of each second moisture accumulation area and the humidity gradient change rate between each adjacent second moisture accumulation area. At the same time, the temperature disturbance increment of each first heat accumulation area and the humidity disturbance increment of each second moisture accumulation area of ​​the target vehicle are obtained by incrementally calculating the local disturbances in conjunction with the interaction of temperature and humidity in the neighborhood.

[0035] The timing confidence window length is dynamically adjusted based on the temperature disturbance increment of each first heat accumulation zone of the target vehicle and the humidity disturbance increment of each second moisture accumulation zone of the target vehicle, and the ventilation operation accumulation zone of each seat of the target vehicle is updated.

[0036] It should be noted that the specific process for obtaining the temperature disturbance increment of each first heat accumulation area and the humidity disturbance increment of each second moisture accumulation area of ​​the target vehicle is as follows: For high heat load areas, the short-term rate of change of its surface temperature is first used as its own disturbance benchmark. At the same time, the temperature change rates of adjacent first heat accumulation areas within the same time period are compared, and the gradient difference or change amplitude between the two is calculated. Then, this gradient difference is accumulated with its own rate change to obtain the temperature disturbance increment of the area affected by heat transfer from the neighboring region, reflecting the additional changes in the area caused by its own heat accumulation and neighboring heat interaction. For moisture retention areas, the absolute humidity change rate is also used as the benchmark. Combined with the humidity gradient change of adjacent second moisture accumulation areas, the humidity change difference is calculated and superimposed on its own change rate to obtain the humidity disturbance increment, which is used to describe the additional effects caused by local moisture diffusion lag and neighboring moisture transfer.

[0037] It should be noted that the short-term rate of change of surface temperature in each region can be obtained by dividing the difference in surface temperature of that region at consecutive time points by the time interval, representing the speed of temperature change per unit time. The temperature gradient sequence between adjacent first heat accumulation regions is obtained by recording the difference in surface temperature of adjacent regions at each sampling time point, and then combining it with the time interval between these two regions to map this temperature difference into gradient values. These gradient values ​​form a sequence at consecutive time points, i.e., the temperature gradient sequence, reflecting the direction and rate of heat transfer along the seat surface. The absolute humidity change rate of each second moisture accumulation region is obtained by the ratio of the humidity difference of each region at consecutive time points to the corresponding time interval, representing the rate of humidity accumulation or dissipation per unit time. The humidity gradient change between adjacent second moisture accumulation regions is obtained by calculating the humidity difference of each pair of neighboring regions and considering the spatial relationship between the regions, reflecting the directionality and intensity of moisture diffusion.

[0038] It should be noted that dynamically adjusting the time-series confidence window length specifically involves inputting the temperature and humidity disturbance increments as dynamic quantitative indicators into the database to determine the adjustment strategy for the window length and sampling frequency. When the temperature or humidity disturbance increment exceeds the corresponding threshold stored in the database, it is input into the time-series confidence window adjustment table. This table is indexed by different levels of local disturbance intensity (based on temperature and humidity disturbance increments). Each disturbance intensity corresponds to a recommended window length and a corresponding sensor sampling frequency. The time-series confidence window length corresponding to the temperature or humidity disturbance increment is directly obtained from the table, and the current window length is adjusted to the value suggested in the table. Simultaneously, the sampling frequency of the sensor in that area is adjusted to the suggested value in the table. If both the temperature and humidity disturbance increments exceed the threshold, they are input into the mapping table respectively. If two window lengths are matched, the shorter window length among the two matching results is selected to ensure the strongest sensitivity to disturbance response.

[0039] It should be noted that after adjusting the time series confidence window length, the update of the ventilation operation accumulation area of ​​each seat in the target vehicle is a process of re-evaluating and correcting the area determination results previously obtained based on longer or shorter windows, based on changes in time resolution and data coverage.

[0040] Specifically, the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle are initially adjusted. The specific process is as follows: In each monitoring cycle, the sampling rate of the temperature sensor of each first heat accumulation area of ​​the target vehicle is increased within the time confidence window. At the same time, the temperature change rate of each adjacent first heat accumulation area of ​​the target vehicle is collected in each monitoring cycle. The temperature change rate of each adjacent first heat accumulation area of ​​the target vehicle is subtracted to obtain the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle. The temperature compensation inference value of each adjacent first heat accumulation area of ​​the target vehicle is obtained based on the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle.

[0041] Similarly, the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle is obtained, and the humidity compensation inference value of each adjacent second moisture accumulation area of ​​the target vehicle is obtained based on the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle.

[0042] It should be noted that, at the same time point, the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle can be obtained by taking the difference between the instantaneous temperature values ​​of the two areas of each adjacent first heat accumulation area of ​​the target vehicle. This reflects the direction and intensity of heat diffusion between adjacent areas and is also used to determine whether there is local thermal blockage. Similarly, the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle is also obtained by taking the difference between the instantaneous temperature values ​​of the two areas of each adjacent second moisture accumulation area of ​​the target vehicle.

[0043] Based on the temperature compensation inference value of each adjacent first heat accumulation zone of the target vehicle and the humidity compensation inference value of each adjacent second moisture accumulation zone of the target vehicle, the ventilation parameters of the ventilation operation accumulation zone of each seat of the target vehicle are initially adjusted.

[0044] It should be noted that the specific process for obtaining the inferred temperature compensation values ​​for each adjacent first heat accumulation zone of the target vehicle is as follows: The temperature gradient sequence of this region is matched with a temperature compensation mapping table stored in the database. This mapping table is established based on historical mass-produced vehicle experimental and simulation data, recording the gradient-corresponding compensation increment sequence between the temperature change response of the seating zones and the required ventilation compensation amount under different gradient amplitudes, durations, and combinations of environmental conditions. Each gradient value in the compensation increment sequence corresponds to a set of reference compensation increments. These increments reflect the expected reduction in temperature deviation under the current heat load conditions by adjusting the ventilation volume and duration. First, the corresponding compensation increment sequence for the continuous periodic gradient values ​​is looked up in the mapping table. Then, based on historical trends and the magnitude of changes in the neighboring gradients, combined with the current ambient temperature, humidity, and occupant pressure sensitivity coverage, the increments obtained from the table are linearly or non-linearly integrated. The system maps historical temperature trends, neighborhood gradients, current ambient temperature, cabin relative humidity changes, and solar radiation intensity to corresponding environmental correction values ​​in the database. Finally, it sums the compensation increments for each time point in the gradient sequence obtained from the lookup table with the environmental correction values, and then takes the average to form the temperature compensation inference value for the region in the current sampling period. This inference value represents the potential temperature rise in the region under continuous heat input and environmental conditions, and can be directly used as an input parameter for ventilation control to adjust the airflow and ventilation duration. Through continuous sampling and real-time updates, it can dynamically respond to changes in high heat load, keeping the seat surface temperature stable within the occupant's comfort range. It also considers neighborhood synergy effects and environmental correction factors to achieve closed-loop prediction and compensation control. The method for obtaining the humidity compensation inference value of each adjacent second moisture accumulation area of ​​the target vehicle is similar to the logic of the temperature compensation inference value. First, the humidity gradient value of the continuous period is looked up in the mapping table for the corresponding compensation increment sequence. Then, combined with the current ambient humidity, cabin temperature, solar radiation intensity, and occupant pressure sensitivity coverage, each humidity compensation increment value obtained from the table is environmentally corrected. Finally, the average value of the corrected sequence is taken to obtain the humidity compensation inference value of each adjacent second moisture accumulation area of ​​the target vehicle.

[0045] It should be noted that the initial adjustment of ventilation parameters for the accumulated ventilation operation areas of each seat in the target vehicle is performed as follows: A control mapping table corresponding to the current vehicle model and seat configuration is retrieved from the database. This table is organized by index item: seat material type, zone ID, and current ventilation level. Each entry in the mapping table maps the temperature compensation level and humidity compensation level to a set of candidate control actions. Each candidate action consists of three elements: a suggested increase or decrease in airflow, a suggested adjustment of the fan speed, and the minimum duration of action. The real-time temperature compensation and humidity compensation inference values ​​are first quantified into discrete compensation levels (e.g., mild / moderate / severe), and the quantification rules are also stored in the database. After obtaining the levels, the corresponding candidate control actions are retrieved from the mapping table, and the candidate actions are compared with the current actual airflow, fan speed, and remaining duration of action to form the target control intent. When only a temperature compensation inference value exists in a certain zone, prioritize using the temperature-wind control action obtained from the lookup table: if the compensation level is slight, the table suggests slightly increasing the airflow or fan speed by one level and maintaining it for a short time; if it is moderate, it is recommended to increase the airflow or fan speed by two levels and extend the application time; if it is severe, it is recommended to directly use a higher airflow level and extend the duration until the temperature drops or reaches the comfort threshold. When both temperature and humidity compensation inference values ​​exist in a certain zone, first apply a short-duration medium-high airflow pulse to quickly remove significant heat, then switch to a lower fan speed but extend the duration to promote water vapor evaporation and removal, thereby simultaneously achieving cooling and dehumidification.

[0046] like Figure 3 and Figure 4 As shown, Figure 3 This is a flowchart of seat ventilation demand prediction and feedforward control based on dynamic environment perception provided in an embodiment of the present invention. Figure 4This is a schematic diagram of the lightweight random forest model provided in this embodiment of the invention. It describes how a prediction model generates ventilation demand prediction results, while simultaneously collecting real-time temperature and humidity data of each accumulation area in each monitoring cycle and performing feedforward ventilation control. The airflow is then constrained to the maximum output value of the fan before generating control commands, which are finally sent to the fan to achieve precise, real-time ventilation adjustment and hardware constraint matching. The ventilation demand prediction results generated by the prediction model use the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle, the humidity gradient sequence of each adjacent second humidity accumulation area, and the local environmental disturbance features collected within the time-series confidence window as the core input data for model prediction. The full set of features is deconstructed into multiple sets of cross-correlated feature subsets, and feature dimensionality reduction is used to achieve the lightweight nature of the model, reducing computational overhead in in-vehicle scenarios. Each feature subset is input into the corresponding lightweight decision trees 1 to n for parallel inference operations. Each decision tree independently outputs intermediate results containing the expected airflow level and expected ventilation mode predicted by a single tree. Finally, the ventilation demand prediction results for each accumulation area of ​​the target vehicle are output, providing data support for the precise control of the vehicle seat ventilation system. The process of generating ventilation demand prediction results through a prediction model includes generating ventilation demand prediction results for the target vehicle based on a short-term prediction-driven ventilation demand inference model. The specific process is as follows: After initially adjusting the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle, the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle, the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle, and the local environmental disturbance features within the time series confidence window are input into the lightweight random forest model to output the ventilation demand prediction results of each accumulation area of ​​the target vehicle. The ventilation demand prediction results of each accumulation area of ​​the target vehicle include the expected air volume level and the expected ventilation mode.

[0047] It's important to note that after identifying and classifying areas of heat or humidity accumulation, if initial adjustments to ventilation parameters aren't made immediately, the temperature gradients, humidity gradients, and various disturbance characteristics collected in subsequent monitoring windows may be biased. This is because these characteristics reflect the bias caused by inappropriate ventilation settings. In other words, if a seat area is marked as having a high heat load or humidity retention, and the original airflow level is maintained, that area will continue to accumulate heat or moisture, leading to abnormally amplified gradients and distorted disturbance responses. This results in a strongly skewed distribution in the random forest model's predicted input, reducing the reliability of the model's output. Initial adjustments are made to ensure that the temperature and humidity gradients when the system enters the time-series confidence window are more realistic, stable, and identifiable. This allows the subsequent lightweight random forest to receive sample inputs closer to real-world needs, resulting in more accurate outputs of desired airflow levels and ventilation modes, bringing temperature and humidity changes within the monitoring window back to an interpretable range. After initial adjustments, these features are then input into a random forest model for high-level demand forecasting, which helps to ensure that the final ventilation control can respond quickly and be finely tuned and optimized.

[0048] It should be noted that, in order to accurately predict the ventilation requirements of the target vehicle in the front, rear, upper, and lower areas of the seat cushion within a short timescale, the controller uses temperature and humidity gradient sequences acquired within a time-series confidence window, combined with local environmental disturbance characteristics, as joint inputs to a lightweight random forest model to obtain the predicted ventilation requirements of the target vehicle. The lightweight random forest model is an ensemble learning model composed of multiple shallow decision trees. The depth and number of trees in each decision tree are controlled, enabling the model to achieve fast inference and low-latency response on computationally limited vehicle controllers. Through training on historical operating data, the model can capture the characteristics of local heat accumulation in the seat, humidity retention trends, changes in occupant contact behavior, and the impact of vehicle environmental disturbances on the heat dissipation process. Regarding input features, the temperature gradient sequence reflects the change pattern of the temperature difference between adjacent first heat accumulation areas over time within a continuous monitoring period, used to describe the direction of local heat transfer and the rate of heat load accumulation; the humidity gradient sequence represents the evolution trend of the humidity difference between adjacent second humidity accumulation areas over multiple sampling periods, used to reflect the accumulation and diffusion rate of local humidity between the textile layer and the occupant's back. Local environmental disturbance features are used to describe the impact of environmental condition changes on local temperature and humidity within the window time length, specifically including but not limited to the instantaneous change rate of solar radiation, the rate of change of cabin ambient temperature, the rate of change of cabin relative humidity, the intensity of occupant posture changes, the displacement velocity of the occupant pressure-sensing matrix center of gravity, and the short-term impact of the current ventilation setting switch on the local heat dissipation rate. The controller collects the temperature gradient sequence and humidity gradient sequence of each seat area of ​​the target vehicle within the time-series confidence window, and simultaneously collects local environmental disturbance features within the time-series window, including instantaneous changes in solar radiation, the rate of change of cabin ambient temperature, the rate of change of cabin relative humidity, occupant posture changes, and pressure-sensing matrix changes. During the fusion processing, the controller pairs the temperature gradient sequence, humidity gradient sequence, and environmental disturbance features within each monitoring cycle according to their temporal correspondence. The gradient information and environmental disturbance information of the same monitoring cycle are merged into a single feature vector, with each feature vector corresponding to the overall temperature and humidity state within that monitoring cycle. Subsequently, the feature vectors of all monitoring cycles within the time-series confidence window are arranged in chronological order to form a sequence data set. Each element in the sequence simultaneously contains temperature gradient, humidity gradient, and environmental disturbance features, used to describe the joint state of local temperature and humidity accumulation and environmental disturbance within that monitoring cycle. Through this processing, the controller can organize multi-source information into an input sequence in chronological order, ensuring that temperature and humidity changes and environmental disturbances are comprehensively considered within the same cycle. A random forest model then performs internal discrimination and prediction for different cycles and different types of features, obtaining the expected airflow level, expected ventilation mode, and expected ventilation demand for each high heat load area and humid stagnation area.

[0049] It should be noted that the expected ventilation demand includes the expected surface temperature, expected rate of temperature change, expected rate of absolute humidity change, and expected humidity gradient change of each accumulation area in the next monitoring cycle.

[0050] Specifically, the ventilation demand forecast results of the target vehicle are dynamically monitored and compared, and the future ventilation intensity of each accumulation area of ​​the target vehicle is controlled in a feedforward manner. The specific process is as follows: The desired airflow level of each accumulation area of ​​the target vehicle is mapped to the fan output of the future control cycle. The airflow, air speed and ventilation duration of each accumulation area of ​​the target vehicle are adjusted. At the same time, the real-time temperature and humidity status data of each accumulation area of ​​the target vehicle are monitored in each monitoring cycle, including surface temperature, temperature change rate, absolute humidity change rate and desired humidity gradient change. The surface temperature, temperature change rate, absolute humidity change rate and desired humidity gradient change of each accumulation area of ​​the target vehicle are subtracted from the surface temperature, desired temperature change rate, desired absolute humidity change rate and desired humidity gradient change of each accumulation area in the next monitoring cycle to obtain the temperature deviation, heat transfer deviation, humidity accumulation deviation and local moisture diffusion deviation of each accumulation area of ​​the target vehicle.

[0051] Feedforward control is performed on the future ventilation intensity of each accumulation area of ​​the target vehicle based on the temperature deviation, heat transfer deviation, humidity accumulation deviation, and local moisture diffusion deviation of each accumulation area.

[0052] It should be noted that the temperature deviation, heat transfer deviation, humidity accumulation deviation, and local moisture diffusion deviation of each accumulation zone reflect the degree to which the surface temperature of each zone exceeds or falls short of the predicted target, the acceleration or lag of heat transfer between zones, the deviation of the actual humidity change from the prediction, and the uneven diffusion of local moisture between adjacent zones, respectively. The controller comprehensively judges whether the current ventilation demand of each accumulation zone is met based on these deviations. If the temperature deviation or heat transfer deviation shows that the surface temperature is higher or lower than the predicted level (positive values ​​indicate a temperature higher than the predicted level, negative values ​​indicate a temperature lower than the predicted level; the same applies to heat transfer deviation, humidity accumulation deviation, and moisture diffusion deviation), or if the humidity accumulation deviation and local moisture diffusion deviation show significant moisture retention, then the ventilation in that zone is deemed insufficient. Conversely, if the deviation shows that the actual temperature and humidity levels are lower than the predicted target or that heat dissipation and dehumidification are excessive, then the ventilation in that zone is deemed excessive. After determining whether ventilation is insufficient or excessive, the controller uses a pre-built database to perform a lookup and matching based on the magnitude and direction of the deviations to obtain the target airflow and ventilation duration required for that zone at the current ventilation level. The database stores the airflow level increment / decrease values ​​and ventilation time correction values ​​corresponding to different deviation ranges. The controller retrieves the corresponding values ​​from the table and uses them as adjustment references. During actual adjustment, the controller adds or subtracts the increment / decrease value obtained from the database to obtain a new airflow setting. For example, if the original airflow is medium, the database query increases it to level one, changing medium to high. Similarly, the ventilation duration is adjusted by adding or subtracting the correction time provided by the database to extend or shorten the ventilation cycle. Fan speed adjustment is achieved by controlling the fan operating mode to match the new airflow setting, ensuring that the actual air delivery rate is consistent with the adjusted airflow. If a region has both temperature and humidity accumulation deviations, the controller will comprehensively consider the relative importance and trends of both, prioritizing the direction with higher heat dissipation or moisture removal requirements, while simultaneously adjusting the airflow and ventilation time accordingly. For example, based on the magnitude and direction of temperature deviation, heat transfer deviation, humidity accumulation deviation, and local moisture diffusion deviation, combined with the system's pre-set deviation levels, the corresponding airflow level increase / decrease value and ventilation time correction value are looked up in the database mapping table. Simultaneously, the priority adjustment direction for that area under the current deviation combination is determined. The database records the airflow and ventilation time adjustment strategies corresponding to different deviation levels or interval combinations, as well as the control decision of whether to prioritize heat dissipation or moisture removal when temperature and humidity deviations coexist. After obtaining the adjustment value from the table, the controller adds it to or reduces the original airflow level and ventilation time to obtain a new control setting. The fan speed is matched to the new airflow level by adjusting the fan operating mode.In practice, if a region has high temperature and heat transfer deviations but low humidity deviations, the airflow is increased and ventilation time is extended to accelerate heat dissipation, while humidity control remains a secondary objective. Conversely, if the temperature deviation is low but the humidity deviation is high, ventilation time is extended to remove moisture, while the airflow is kept at a medium level to avoid excessive cooling. During actual control, the controller continuously reads sensor data, updates deviations, looks up the latest target adjustment values, and dynamically corrects the existing airflow and ventilation time, thus forming a continuous feedforward closed-loop control. This ensures that the actual temperature and humidity in each region quickly approach the predicted target, guaranteeing that the seat ventilation system maintains optimal heat dissipation and dehumidification under different environmental conditions and occupant states.

[0053] Specifically, an executable airflow control signal is generated, and the executable range of the fan in the target vehicle is obtained. The specific process is as follows: Obtain the key performance parameters of the fan in the target vehicle, including the minimum and maximum output air volume of the fan, and use the minimum and maximum output air volume of the fan as the executable range of the fan in the target vehicle.

[0054] It should be noted that the key performance parameters of the wind turbine for the target vehicle can be obtained by consulting the technical specifications provided by the wind turbine manufacturer or the vehicle calibration manual.

[0055] Specifically, hardware constraint matching is performed on the executable airflow control signal. The specific process is as follows: extract the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle and the executable range of the target vehicle's fan. If the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle does not exceed the executable range of the target vehicle's fan, then the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle is connected to the current fan control signal to generate the final issued airflow control command. If the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle exceeds the executable range of the target vehicle's fan, then the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle is constrained to the maximum output airflow of the fan and then connected to the current fan control signal to generate the final issued airflow control command.

[0056] It should be noted that the instantaneous changes in future ventilation intensity in each accumulation area of ​​the target vehicle represent ideal seat ventilation parameters. However, the actual hardware airflow adjustment capability is limited: the fan can only operate at a few speeds (e.g., low, medium, high). Therefore, predicted values ​​exceeding the hardware range need to be clipped or mapped. The instantaneous changes in future ventilation intensity in each accumulation area of ​​the target vehicle are linearly mapped to the nearest executable speed, or the output is clipped through constraint prediction to avoid system overshoot or lag. When the prediction model gives an instantaneous change in future ventilation intensity of 15L / min for a certain accumulation area, and the upper limit of the target vehicle's fan's executable range is 20L / min, this predicted value is within the hardware capability and can be directly used as an executable airflow input to the current fan control signal to generate the final command. Conversely, when a prediction result is 28L / min, exceeding the fan's maximum output of 20L / min, the system will clip the instantaneous change according to the hardware capability, limiting 28L / min to the executable upper limit of 20L / min, before using it to generate the final command from the current fan control signal. This approach ensures that the prediction model responds to changes in ventilation demand while guaranteeing that the output ventilation intensity does not exceed the actual executable range of the fan hardware. It avoids airflow disturbances, system oscillations, or delays caused by excessive commands, thereby achieving a smooth connection between predicted airflow and fan execution capabilities.

[0057] like Figure 5 As shown, Figure 5 This is a schematic diagram of a seat ventilation parameter prediction and control system module based on dynamic environment perception. The system includes a multi-source sensing and accumulation area determination module, which is used to acquire multi-source asynchronous information from the target vehicle's sensors and perform spatiotemporal alignment determination to obtain the ventilation operation accumulation area of ​​each seat in the target vehicle. At the same time, based on the correspondence between temperature change and airflow level records, it obtains the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, and updates the ventilation operation accumulation area of ​​each seat in the target vehicle.

[0058] The active data acquisition and ventilation parameter compensation module is used to trigger the active data acquisition and compensation inference of the sensors based on the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, and to make preliminary adjustments to the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle.

[0059] The short-term prediction-driven feedforward control module is used to generate ventilation demand prediction results for the target vehicle based on the short-term prediction-driven ventilation demand inference model, and to dynamically monitor and compare the ventilation demand prediction results of the target vehicle to perform feedforward control on the future ventilation intensity of each accumulated area of ​​the target vehicle.

[0060] The air volume control signal generation and hardware constraint matching module is used to generate executable air volume control signals, obtain the fan's executable range for the target vehicle, and perform hardware constraint matching on the executable air volume control signals.

[0061] Optionally, in this embodiment, the storage medium may be located at at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0062] This invention also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform an audio processing method provided in the various optional embodiments described above.

[0063] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0064] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device, equipment, and storage medium embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0065] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0066] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for predicting and controlling seat ventilation parameters based on dynamic environmental perception, characterized in that, The method includes: The system acquires multi-source asynchronous information from the target vehicle's sensors and performs spatiotemporal alignment determination to obtain the ventilation operation accumulation area of ​​each seat in the target vehicle. Simultaneously, based on the correspondence between temperature change and airflow level records, it obtains the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle and updates the ventilation operation accumulation area of ​​each seat in the target vehicle. Based on the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, the active acquisition and compensation inference of the sensor triggers the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle to be initially adjusted. Based on the ventilation demand inference model driven by short-term prediction, the ventilation demand prediction results of the target vehicle are generated, and the ventilation demand prediction results of the target vehicle are dynamically monitored and compared to control the future ventilation intensity of each accumulation area of ​​the target vehicle in a feedforward manner. Generate an executable airflow control signal and obtain the fan's executable range for the target vehicle, then perform hardware constraint matching on the executable airflow control signal.

2. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The specific process for obtaining the ventilation operation accumulation area of ​​each seat in the target vehicle is as follows: A time-series confidence window is established based on the sensor sampling delay. Multi-source asynchronous information from the target vehicle's sensors is extracted from the time-series confidence window, including the seat surface temperature, heat dissipation rate, absolute humidity change rate, and cabin relative humidity change rate of each seat area of ​​the target vehicle. Extract the surface temperature of the seats in each seat area of ​​the target vehicle and collect the current ambient temperature and current solar radiation intensity; The current solar radiation intensity is compared with the solar radiation threshold stored in the database, and the surface temperature of the seats in each seat area of ​​the target vehicle is compared with the current ambient temperature. If the surface temperature of the seat in each seat area of ​​the target vehicle is higher than or equal to the current ambient temperature and the solar radiation intensity is higher than or equal to the solar radiation threshold, then the seat area of ​​the target vehicle is marked as the first heat accumulation area, thus obtaining each first heat accumulation area; otherwise, no marking is required. The absolute humidity change rate of each seat area in the target vehicle is compared with the relative humidity change rate of the vehicle cabin. If the absolute humidity change rate of each seat area in the target vehicle is higher than or equal to the relative humidity change rate of the vehicle cabin, then that seat area in the target vehicle is marked as the second moisture accumulation area, thus obtaining each second moisture accumulation area. Otherwise, no marking is required.

3. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle is obtained based on the correspondence between temperature changes and airflow level records. The specific process is as follows: Obtain the ventilation setting of the target vehicle and calculate the temperature change per unit time for each first heat accumulation area and each second moisture accumulation area of ​​the target vehicle at that ventilation setting. If the temperature change of any accumulation area within a unit of time at this ventilation setting is lower than or equal to the temperature change threshold stored in the database, then the accumulation area is further marked as an area with insufficient heat dissipation; otherwise, no marking is required. If a certain accumulation area is simultaneously marked as the first heat accumulation area and the insufficient heat dissipation area, then the effective heat dissipation state of that area is determined to be a high heat load state. If a certain accumulation area is simultaneously marked as a second moisture accumulation area and a heat dissipation inadequate area, then the effective heat dissipation state of that area is determined to be a moisture retention state.

4. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The specific process for updating the accumulated ventilation operation areas of each seat in the target vehicle is as follows: Within a time-series confidence window, several monitoring periods are preset to acquire local environmental disturbance characteristic data of each first heat accumulation area and each second moisture accumulation area of ​​the target vehicle. The local environmental disturbance characteristic data of each first heat accumulation area of ​​the target vehicle includes the short-term change rate of the surface temperature of each first heat accumulation area and the temperature gradient sequence of each adjacent first heat accumulation area. The local environmental disturbance characteristic data of each second moisture accumulation area includes the absolute humidity change rate of each second moisture accumulation area and the humidity gradient change rate between each adjacent second moisture accumulation area. At the same time, the local disturbance is incrementally calculated by combining the influence of temperature and humidity interaction in the neighborhood to obtain the temperature disturbance increment of each first heat accumulation area of ​​the target vehicle and the humidity disturbance increment of each second moisture accumulation area of ​​the target vehicle. The timing confidence window length is dynamically adjusted based on the temperature disturbance increment of each first heat accumulation zone of the target vehicle and the humidity disturbance increment of each second moisture accumulation zone of the target vehicle, and the ventilation operation accumulation zone of each seat of the target vehicle is updated.

5. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The preliminary adjustment of ventilation parameters in the accumulated ventilation operation areas of each seat in the target vehicle is carried out as follows: In each monitoring cycle, the first heat accumulation area of ​​the target vehicle is statistically analyzed. Within the time confidence window, the sampling rate of the temperature sensor in each first heat accumulation area is increased. At the same time, the temperature change rate of each adjacent first heat accumulation area of ​​the target vehicle is collected in each monitoring cycle. The temperature change rate of each adjacent first heat accumulation area of ​​the target vehicle is subtracted to obtain the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle. The temperature compensation inference value of each adjacent first heat accumulation area of ​​the target vehicle is obtained based on the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle. Similarly, the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle is obtained, and the humidity compensation inference value of each adjacent second moisture accumulation area of ​​the target vehicle is obtained based on the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle. Based on the temperature compensation inference value of each adjacent first heat accumulation zone of the target vehicle and the humidity compensation inference value of each adjacent second moisture accumulation zone of the target vehicle, the ventilation parameters of the ventilation operation accumulation zone of each seat of the target vehicle are initially adjusted.

6. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The process of generating ventilation demand prediction results for the target vehicle based on the short-term prediction-driven ventilation demand inference model is as follows: After making preliminary adjustments to the ventilation parameters of the ventilation operation accumulation areas of each seat in the target vehicle, the temperature gradient sequence of each adjacent first heat accumulation area of ​​the target vehicle, the humidity gradient sequence of each adjacent second moisture accumulation area of ​​the target vehicle, and the local environmental disturbance features within the time series confidence window are input into a lightweight random forest model to output the ventilation demand prediction results of each accumulation area of ​​the target vehicle. The ventilation demand prediction results of each accumulation area of ​​the target vehicle include the desired air volume level and the desired ventilation mode.

7. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The process of dynamically monitoring and comparing the predicted ventilation demand of the target vehicle, and then performing feedforward control on the future ventilation intensity of each accumulated area of ​​the target vehicle, is as follows: The desired airflow level of each accumulation area of ​​the target vehicle is mapped to the fan output of the future control cycle. The airflow, wind speed and ventilation duration of each accumulation area of ​​the target vehicle are adjusted. At the same time, the real-time temperature and humidity status data of each accumulation area of ​​the target vehicle are monitored in each monitoring cycle, including surface temperature, temperature change rate, absolute humidity change rate and desired humidity gradient change. The surface temperature, temperature change rate, absolute humidity change rate and desired humidity gradient change of each accumulation area of ​​the target vehicle are subtracted from the surface temperature, desired temperature change rate, desired absolute humidity change rate and desired humidity gradient change of each accumulation area in the next monitoring cycle to obtain the temperature deviation, heat transfer deviation, humidity accumulation deviation and local moisture diffusion deviation of each accumulation area of ​​the target vehicle. Feedforward control is performed on the future ventilation intensity of each accumulation area of ​​the target vehicle based on the temperature deviation, heat transfer deviation, humidity accumulation deviation, and local moisture diffusion deviation of each accumulation area.

8. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The specific process of generating an executable airflow control signal and obtaining the fan's executable range for the target vehicle is as follows: Obtain the key performance parameters of the fan in the target vehicle, including the minimum and maximum output air volume of the fan, and use the minimum and maximum output air volume of the fan as the executable range of the fan in the target vehicle.

9. The method for predicting and controlling seat ventilation parameters based on dynamic environment perception as described in claim 1, characterized in that, The specific process of performing hardware constraint matching on the executable airflow control signal is as follows: Extract the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle and the executable range of the target vehicle's fan. If the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle does not exceed the executable range of the target vehicle's fan, then the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle is connected to the current fan control signal to generate the final issued air volume control command. If the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle exceeds the executable range of the target vehicle's fan, then the instantaneous change in future ventilation intensity of each accumulated area of ​​the target vehicle is constrained to the maximum output air volume of the fan before being connected to the current fan control signal to generate the final issued air volume control command.

10. A system applying the seat ventilation parameter prediction and control method based on dynamic environment perception as described in any one of claims 1-9, characterized in that, include: The multi-source sensing and accumulation area determination module is used to acquire multi-source asynchronous information from the target vehicle's sensors and perform spatiotemporal alignment determination to obtain the ventilation operation accumulation area of ​​each seat in the target vehicle. At the same time, based on the correspondence between temperature change and airflow level records, it obtains the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle and updates the ventilation operation accumulation area of ​​each seat in the target vehicle. The active acquisition and ventilation parameter compensation module is used to trigger the active acquisition and compensation inference of the sensor based on the effective heat dissipation status of the ventilation operation accumulation area of ​​each seat in the target vehicle, and to make preliminary adjustments to the ventilation parameters of the ventilation operation accumulation area of ​​each seat in the target vehicle. The short-term prediction-driven feedforward control module is used to generate ventilation demand prediction results for the target vehicle based on the short-term prediction-driven ventilation demand inference model, and to dynamically monitor and compare the ventilation demand prediction results of the target vehicle, and to perform feedforward control on the future ventilation intensity of each accumulation area of ​​the target vehicle. The air volume control signal generation and hardware constraint matching module is used to generate executable air volume control signals, obtain the fan's executable range for the target vehicle, and perform hardware constraint matching on the executable air volume control signals.

Images