Air conditioner control method, vehicle, and electronic device

Abstract

Embodiments of the present application provide an air conditioner control method, a vehicle and electronic equipment, which comprises the following steps: obtaining an initial physiological characteristic sequence and an initial environment temperature characteristic sequence according to a preset time cycle; constructing an individual dynamic thermal comfort evaluation model based on the initial physiological characteristic sequence and the initial environment temperature characteristic sequence which are synchronized by a time stamp; calculating a real-time thermal comfort index of a target user by the individual dynamic thermal comfort evaluation model; determining a target control parameter for the air conditioner based on the real-time thermal comfort index, an instant physiological characteristic sequence and an instant environment temperature characteristic sequence which are synchronized by the time stamp; generating a control command for the air conditioner by using the target control parameter, and controlling the air conditioner based on the control command. Thus, by time sequence synchronization modeling of physiological and environmental characteristics, the hysteresis effect of human physiological feedback can be effectively eliminated, and precise dynamic matching of the air conditioner control parameter and the passenger transient thermal perception demand can be realized.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to an air conditioning control method, an air conditioning control vehicle, an electronic device, and a computer-readable storage medium. Background Technology

[0002] As vehicles become increasingly intelligent, in-vehicle air conditioning systems have become a key factor affecting passenger comfort. However, most existing in-vehicle air conditioning control methods rely primarily on in-vehicle environmental parameters (such as temperature and humidity), controlling these parameters through preset fixed thresholds or simple linear adjustment logic. This approach ignores the dynamic changes in individual passenger physiological states, leading to a disconnect between air conditioning control and actual passenger thermal comfort needs. During actual driving, different passengers exhibit significant differences in physiological responses; even at the same ambient temperature, different passengers may experience different thermal sensations. Furthermore, the physiological state of the same passenger changes dynamically over time at different stages of the journey. Traditional methods lack real-time data collection and dynamic response mechanisms for these physiological changes. When passenger physiological states change, the air conditioning system cannot adjust promptly and accurately, potentially resulting in some passengers being overheated or undercooled, leading to poor thermal comfort and unnecessary energy consumption. Summary of the Invention

[0003] The present invention provides an air conditioning control method, apparatus, electronic device, and computer-readable storage medium to overcome or at least partially solve the above-mentioned problems.

[0004] This invention discloses an air conditioning control method, comprising: According to a preset time period, an initial physiological feature sequence for characterizing the user's physiological changes within the time period and an initial environmental temperature feature sequence for characterizing the user's environment temperature changes within the time period are obtained cyclically. A timestamp synchronization operation is performed on the initial physiological characteristic sequence and the initial ambient temperature characteristic sequence, and an individual dynamic thermal comfort evaluation model is constructed based on the initial physiological characteristic sequence and the initial ambient temperature characteristic sequence after the timestamp synchronization operation. The real-time thermal comfort index of the target user is calculated using the individual dynamic thermal comfort evaluation model. Based on the real-time thermal comfort index, as well as the real-time physiological characteristic sequence and real-time ambient temperature characteristic sequence synchronized with timestamps, the target control parameters for the air conditioner are determined. The target control parameters are used to generate control commands for the air conditioner, and the air conditioner is controlled based on the control commands.

[0005] Optionally, the method further includes: Acquire image information of passengers inside the vehicle; A multi-dimensional extraction operation of clothing features is performed on the in-vehicle passenger image information to obtain clothing feature parameters used to characterize the heat exchange barrier performance of passengers.

[0006] Optionally, the step of performing multi-dimensional extraction of clothing features on the in-vehicle passenger image information to obtain clothing feature parameters characterizing the passenger's heat exchange barrier performance includes: Human key point recognition is performed on the in-vehicle passenger image information to locate and extract the passenger target area image; Perform semantic segmentation processing on the passenger target area image to determine the clothing coverage area image belonging to the clothing pixels; Texture frequency and edge gradient analysis are performed on the image of the clothing-covered area to generate material sensory feature information that characterizes the thickness of the clothing material; The sensory characteristics of the material are combined with the contour features of the image of the clothing-covered area to perform multi-dimensional feature matching, thereby determining the passenger's clothing category information; The clothing category information is matched with a preset physical thermal resistance comparison data table, and the thermal resistance reference value corresponding to the clothing category information is extracted from the preset physical thermal resistance comparison data table. The thermal resistance baseline value is combined with the area ratio of the clothing coverage area image in the passenger target area image for weighted calculation to generate clothing feature parameters that characterize the heat exchange barrier performance of the passenger.

[0007] Optionally, the step of constructing an individual dynamic thermal comfort evaluation model based on the initial physiological characteristic sequence and the initial environmental temperature characteristic sequence synchronized by the timestamp includes: A weighted summation calculation is performed on the skin temperature feature sequence used to characterize skin temperature changes in different parts of the body to generate a dynamic average skin temperature sequence. Obtain metabolic baseline information, as well as a thermal comfort state heart rate variability sequence for characterizing the ratio of low-frequency to high-frequency heart rate variability under thermal comfort conditions, and a thermal comfort state respiratory rate sequence for characterizing the ideal respiratory rate under thermal comfort conditions. Using the basal metabolic rate baseline value, the heart rate variability sequence used to characterize the ratio of low-frequency to high-frequency heart rate variability, the heart rate variability sequence of thermal comfort state, the respiratory rate sequence used to characterize respiratory rate change information, and the respiratory rate sequence of thermal comfort state, and combined with the clothing characteristic parameters, a nonlinear compensation calculation is performed to obtain a dynamic human metabolic rate sequence reflecting the real-time heat production capacity of the human body. Correction coefficients are determined based on the temperature change trend feature sequence used to characterize the trend of ambient temperature change and the clothing feature parameters, and the maximum fluctuation range of the skin temperature change information, the low-frequency to high-frequency ratio of the heart rate variability and the respiratory rate change information are determined. The comprehensive rate of change of core physiological parameters is determined based on the dynamic average skin temperature sequence, the correction factor, the maximum fluctuation range, the baseline value of basal metabolic rate, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, and the thermal comfort state respiratory rate sequence. The target skin temperature is determined based on the temperature change trend feature sequence, and weighting coefficients corresponding one-to-one with skin temperature, heart rate, and respiratory rate are determined. An individual dynamic thermal comfort evaluation model is constructed based on the clothing characteristic parameters, the weighting coefficients, the target skin temperature, the dynamic average skin temperature sequence, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, the thermal comfort state respiratory rate sequence, the maximum fluctuation range, and the comprehensive rate of change of the core physiological parameters.

[0008] Optionally, the step of calculating the real-time thermal comfort index of the target user through the individual dynamic thermal comfort evaluation model includes: According to a preset time cycle, the system cyclically acquires an instant physiological feature sequence to characterize the user's instant physiological changes within the current time cycle, and an instant environmental temperature feature sequence to characterize the user's environment's instant environmental temperature changes within the current time cycle. Acquire real-time images of passengers inside the vehicle; Perform multi-dimensional extraction of clothing features on the real-time in-vehicle passenger image information to obtain real-time clothing feature parameters that characterize the current passenger's heat exchange barrier performance; A timestamp synchronization operation is performed on the real-time physiological feature sequence and the real-time ambient temperature feature sequence, and based on the real-time physiological feature sequence and the real-time ambient temperature feature sequence after the timestamp synchronization operation, the real-time dynamic average skin temperature sequence, the real-time dynamic human metabolic rate sequence, and the comprehensive change rate of the real-time core physiological parameters of the target user are calculated. The real-time clothing characteristic parameters, the real-time dynamic average skin temperature sequence, the real-time dynamic human metabolic rate sequence, and the combined rate of change of the real-time core physiological parameters are input into the individual dynamic thermal comfort evaluation model to control the output of the real-time thermal comfort index of the individual dynamic thermal comfort evaluation model.

[0009] Optionally, the real-time physiological characteristic sequence includes a real-time heart rate variability sequence and a real-time respiratory rate sequence, the real-time ambient temperature characteristic sequence includes a temperature change trend characteristic sequence, and the step of determining the target control parameters for the air conditioner based on the real-time thermal comfort index and the real-time physiological characteristic sequence and the real-time ambient temperature characteristic sequence synchronized with timestamps includes: Based on the clothing characteristic parameters, the real-time thermal comfort index, the comprehensive rate of change of the instantaneous core physiological parameters, the instantaneous dynamic average skin temperature sequence, the instantaneous dynamic human metabolic rate sequence, the instantaneous heart rate variability sequence, and the instantaneous respiratory rate sequence, a multi-objective optimization function is constructed. Based on the temperature change trend characteristic sequence, determine the lower and upper limits of the target temperature inside the vehicle, the lower and upper limits of the target relative humidity inside the vehicle, and the lower and upper limits of the fresh air volume. The lower and upper limits of the target temperature inside the vehicle, the lower and upper limits of the target relative humidity inside the vehicle, and the lower and upper limits of the fresh air volume are determined as the constraints of the multi-objective optimization function. The multi-objective optimization function is solved based on the constraints to obtain the target supply air temperature, target relative humidity, target fresh air volume, supply air velocity, and supply air direction.

[0010] Optionally, the step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Obtain the current supply air temperature value and the compressor frequency conversion characteristic information of the air conditioner; the compressor frequency conversion characteristic information includes the compressor reference frequency; Calculate the difference between the target supply air temperature value and the current supply air temperature value; The target compressor frequency at the target time is calculated based on the difference, the preset temperature adjustment coefficient, and the compressor reference frequency. The compressor control command is generated using the target compressor frequency at the target time. Based on the compressor control commands, the compressor is driven to perform air supply temperature adjustment.

[0011] Optionally, the step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Get the current relative humidity value inside the vehicle; When it is determined that the target relative humidity value is lower than the current in-vehicle relative humidity value, the difference between the current in-vehicle relative humidity value and the target relative humidity value is calculated; Obtain information on air density, cooling space volume, and water specific heat capacity; The dehumidification power at the target time is calculated based on the difference, the air density information, the cooling space volume information, and the specific heat capacity information of water. The dehumidification power at the target time is used to generate control commands for the dehumidification device; The dehumidifier of the air conditioner is driven to adjust the humidity based on the control command of the dehumidifier.

[0012] Optionally, the step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Obtain the minimum and maximum fresh air volume; Calculate the first difference between the target fresh air volume value and the minimum fresh air volume value; Calculate the second difference between the maximum fresh air volume and the minimum fresh air volume; The percentage of the fresh air valve opening is calculated based on the first difference and the second difference; The opening percentage of the fresh air valve is used to adjust the opening of the fresh air valve of the air conditioner.

[0013] This invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method described in the embodiments of the present invention.

[0014] This invention also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this invention.

[0015] This invention also discloses a vehicle and the electronic device described above.

[0016] The embodiments of the present invention have the following advantages: This invention, in its embodiments, acquires an initial physiological feature sequence characterizing a user's physiological changes within a preset time period, and an initial ambient temperature feature sequence characterizing the user's environment's temperature changes within the same time period. A timestamp synchronization operation is performed on the initial physiological feature sequence and the initial ambient temperature feature sequence, and an individual dynamic thermal comfort evaluation model is constructed based on these synchronized sequences. The real-time thermal comfort index of the target user is calculated using this model. Based on the real-time thermal comfort index, and the synchronized physiological feature sequence and the synchronized ambient temperature feature sequence, target control parameters for the air conditioner are determined. Control commands for the air conditioner are generated using these target control parameters, and the air conditioner is controlled based on these commands. This achieves precise dynamic matching between air conditioner control parameters and passengers' transient thermal perception needs by synchronously modeling physiological and environmental characteristics over time, effectively eliminating the lag in human physiological feedback. Attached Figure Description

[0017] Figure 1 This is a flowchart of the steps of an air conditioning control method provided in an embodiment of the present invention; Figure 2 This is a system block diagram of a vehicle air conditioning system provided in an embodiment of the present invention; Figure 3 This is a hardware structure block diagram of an electronic device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] Reference Figure 1 The diagram illustrates a flowchart of an air conditioning control method provided in an embodiment of the present invention, which may specifically include the following steps: Step 101: Circularly acquire an initial physiological feature sequence for characterizing the user's physiological changes within the time period, and an initial ambient temperature feature sequence for characterizing the ambient temperature changes of the user's environment within the time period, according to a preset time period. Step 102: Perform a timestamp synchronization operation on the initial physiological characteristic sequence and the initial ambient temperature characteristic sequence, and construct an individual dynamic thermal comfort evaluation model based on the initial physiological characteristic sequence and the initial ambient temperature characteristic sequence after the timestamp synchronization operation. Step 103: Calculate the real-time thermal comfort index of the target user using the individual dynamic thermal comfort evaluation model; Step 104: Based on the real-time thermal comfort index, and the real-time physiological characteristic sequence and real-time ambient temperature characteristic sequence synchronized by timestamp, determine the target control parameters for the air conditioner. Step 105: Generate a control command for the air conditioner using the target control parameters, and control the air conditioner based on the control command.

[0020] In this embodiment of the invention, initial physiological feature sequences characterizing the user's physiological changes within a preset time period, and initial environmental temperature feature sequences characterizing the user's environment temperature changes within the same time period, can be cyclically acquired to establish a basis for the system's perception of the user's physiological state and changes in the external environment. By cyclically acquiring "sequence data" within a specific time period, the system can capture the trend of parameters evolving over time (such as whether body temperature is rising or falling), rather than just acquiring isolated instantaneous values, thus providing raw material for subsequent dynamic trend analysis.

[0021] The preset time period refers to the fixed time interval at which the system executes data acquisition tasks, and is used to define the time length of the feature sequence.

[0022] Physiological change information refers to biological indicators that reflect the body's internal thermoregulation state, including skin temperature fluctuations, heart rate variability, and respiratory rate.

[0023] The initial physiological characteristic sequence refers to a set of physiological parameter values ​​arranged in chronological order within a preset time period.

[0024] The initial ambient temperature characteristic sequence refers to a set of numerical values ​​that record the evolution of the ambient temperature inside and outside the vehicle over time within the same period.

[0025] In this embodiment of the invention, a timestamp synchronization operation can be performed on the initial physiological characteristic sequence and the initial environmental temperature characteristic sequence. Based on the timestamp-synchronized initial physiological characteristic sequence and initial environmental temperature characteristic sequence, an individual dynamic thermal comfort evaluation model is constructed. This timestamp synchronization eliminates data misalignment caused by inconsistent sampling frequencies of different sensors, ensuring that physiological responses and environmental triggers accurately correspond on the timeline. The purpose of building this model is to characterize the dynamic response patterns of human physiological indicators to environmental changes and to address the lag problem in human thermal perception.

[0026] Timestamp synchronization refers to aligning and calibrating sequence data from different sources using a unified time reference, so that physiological data and environmental data at the same time can be matched.

[0027] An individual dynamic thermal comfort evaluation model is a mathematical model built based on a specific user's historical and real-time data to describe the individual's thermal perception logic in a dynamic environment.

[0028] In this embodiment of the invention, the real-time thermal comfort index of a target user can be calculated using the individual dynamic thermal comfort evaluation model. This transforms complex, multidimensional physiological and environmental sequence data into an intuitive digital indicator that can be used to generate control commands. The index calculated by this model accurately reflects the user's current level of physical comfort and serves as the direct logical basis for air conditioning adjustments.

[0029] The target users refer to the main users of the current air conditioning system, usually the occupants or driver in the vehicle.

[0030] The real-time thermal comfort index is a quantitative value that reflects the human body's satisfaction with the temperature of the environment at a given moment.

[0031] In this embodiment of the invention, target control parameters for the air conditioner can be determined based on the real-time thermal comfort index, as well as the real-time physiological characteristic sequence and the real-time ambient temperature characteristic sequence synchronized with timestamps, so as to transform "human subjective needs (thermal comfort index)" into "machine physical goals (control parameters)". By comprehensively considering real-time data and trend data, the combination of air conditioner operating parameters that can achieve a comfortable state most quickly and stably is calculated.

[0032] Real-time physiological feature sequence / real-time environmental temperature feature sequence refers to the latest feature data sequence generated at the current sampling time, reflecting transient changes.

[0033] Target control parameters refer to the physical quantity targets that an air conditioning system needs to set to achieve an ideal comfort state, such as target supply air temperature and target air velocity.

[0034] In this embodiment of the invention, the target control parameters can be used to generate control commands for the air conditioner, and the air conditioner can be controlled based on the control commands. This transforms the calculated abstract target into an electronic control command that the air conditioner hardware can recognize. By adjusting the operating status of the compressor, fan and other execution units, the thermal environment inside the vehicle is ultimately changed, completing a complete closed loop of "perception-decision-execution".

[0035] Control commands refer to specific electrical signals or bus instructions sent to air conditioning actuators (such as electronically controlled expansion valves and motor controllers).

[0036] Air conditioning operating status refers to the current physical operating parameters of each component of the air conditioning system, including compressor frequency, air valve opening degree, and airflow angle.

[0037] This invention, in its embodiments, acquires an initial physiological feature sequence characterizing a user's physiological changes within a preset time period, and an initial ambient temperature feature sequence characterizing the user's environment's temperature changes within the same time period. A timestamp synchronization operation is performed on the initial physiological feature sequence and the initial ambient temperature feature sequence, and an individual dynamic thermal comfort evaluation model is constructed based on these synchronized sequences. The real-time thermal comfort index of the target user is calculated using this model. Based on the real-time thermal comfort index, and the synchronized physiological feature sequence and the synchronized ambient temperature feature sequence, target control parameters for the air conditioner are determined. Control commands for the air conditioner are generated using these target control parameters, and the air conditioner is controlled based on these commands. This achieves precise dynamic matching between air conditioner control parameters and passengers' transient thermal perception needs by synchronously modeling physiological and environmental characteristics over time, effectively eliminating the lag in human physiological feedback.

[0038] Based on the above embodiments, modified embodiments of the above embodiments are proposed. It should be noted that, in order to keep the description brief, only the differences from the above embodiments are described in the modified embodiments.

[0039] In an optional embodiment of the present invention, the method further includes: Acquire image information of passengers inside the vehicle; A multi-dimensional extraction operation of clothing features is performed on the in-vehicle passenger image information to obtain clothing feature parameters used to characterize the heat exchange barrier performance of passengers.

[0040] Optionally, the step of performing multi-dimensional extraction of clothing features on the in-vehicle passenger image information to obtain clothing feature parameters characterizing the passenger's heat exchange barrier performance includes: Human key point recognition is performed on the in-vehicle passenger image information to locate and extract the passenger target area image; Perform semantic segmentation processing on the passenger target area image to determine the clothing coverage area image belonging to the clothing pixels; Texture frequency and edge gradient analysis are performed on the image of the clothing-covered area to generate material sensory feature information that characterizes the thickness of the clothing material; The sensory characteristics of the material are combined with the contour features of the image of the clothing-covered area to perform multi-dimensional feature matching, thereby determining the passenger's clothing category information; The clothing category information is matched with a preset physical thermal resistance comparison data table, and the thermal resistance reference value corresponding to the clothing category information is extracted from the preset physical thermal resistance comparison data table. The thermal resistance baseline value is combined with the area ratio of the clothing coverage area image in the passenger target area image for weighted calculation to generate clothing feature parameters that characterize the heat exchange barrier performance of the passenger.

[0041] The embodiments of the present invention can acquire in-vehicle passenger image information to obtain real-time images of passengers through a camera, enabling the system to have "visual" perception capabilities, thereby enabling non-contact acquisition of passengers' external environmental adaptation status (such as clothing).

[0042] In-vehicle passenger image information refers to digital image data containing the outline and surface features of passengers, captured by in-vehicle image sensors (such as infrared or visible light cameras).

[0043] The embodiments of the present invention can identify key points and extract regions of the human body. By identifying key points of the skeleton, the system can eliminate interference such as seats and interior decorations, and concentrate computing resources on the core areas related to the passenger's body, thereby improving the accuracy and efficiency of subsequent processing.

[0044] Human key point recognition refers to the technology of using computer vision algorithms to locate the coordinates of major human joints (such as shoulders, elbows, and wrists).

[0045] The passenger target area image refers to a local image cropped based on key point coordinates, which only contains the passenger's torso and limbs.

[0046] The embodiments of the present invention can perform semantic segmentation processing on the passenger target area image to determine the clothing coverage area image belonging to the clothing pixels. By separating the "person" and "clothing" in the image, the system can know exactly which parts of the passenger's body are covered by clothing and which parts are exposed, providing a spatial distribution basis for calculating thermal resistance.

[0047] Semantic segmentation refers to advanced image processing techniques that classify each pixel in an image (such as distinguishing skin, fabric, background, etc.).

[0048] A clothing-covered area image refers to an image of a collection of pixels extracted from a target area, consisting of clothing material.

[0049] The embodiments of the present invention can perform texture frequency and edge gradient analysis on the image of the clothing coverage area to generate material sensory feature information that characterizes the thickness of the clothing material. By analyzing the texture and edge gradient, the system can distinguish the difference between a thick down jacket (rounded edges, sparse texture) and a thin T-shirt (sharp edges, fine texture), thereby inferring its heat insulation ability.

[0050] Texture frequency refers to the repeatability of the spatial distribution of pixel gray values ​​in an image, and is used to determine the roughness or density of a fabric.

[0051] Edge gradient refers to the rate of change of pixels in a spatial direction, and is often used to characterize the contour, thickness, and stacked appearance of clothing.

[0052] Material sensory characteristics refer to digital feature description vectors that characterize the thickness, fluffiness, or material type of clothing.

[0053] In this embodiment of the invention, the material sensory feature information can be combined with the contour features of the clothing coverage area image to perform multi-dimensional feature matching to determine the passenger's clothing category information. By combining material features with shape contours (such as long sleeves, short sleeves, coats, etc.), the visual perception result can be finally locked into the specific clothing type so as to call standard thermal parameters.

[0054] Outline features refer to the geometric shape characteristics of the clothing area, such as aspect ratio and boundary curve shape.

[0055] Multidimensional feature matching refers to the process of comparing multiple extracted feature points with known clothing features in a database and finding the most similar item.

[0056] Clothing category information refers to the specific clothing name or level determined after matching, such as "thick coat" or "short-sleeved T-shirt".

[0057] In this embodiment of the invention, the clothing category information can be matched with a preset physical thermal resistance reference data table. The thermal resistance reference value corresponding to the clothing category information can be extracted from the preset physical thermal resistance reference data table. By querying the standard physical data table, the corresponding experimental calibration thermal resistance value can be assigned to different clothing items, so that the calculation is at the physical level.

[0058] The physical thermal resistance comparison data table refers to a database of physical parameters that record the ability of various standard garments to impede heat flow under thermal equilibrium conditions.

[0059] The thermal resistance benchmark value refers to the measured value of thermal resistance per unit area of ​​a specific type of clothing under standard wearing conditions.

[0060] In this embodiment of the invention, the thermal resistance baseline value can be combined with the area ratio of the clothing coverage area image in the passenger target area image for weighted calculation to generate clothing characteristic parameters that characterize the heat exchange barrier performance of the passenger. By combining the actual proportion of clothing on the body to calculate the baseline thermal resistance, the overall thermal insulation status of the passenger can be more realistically reflected.

[0061] Area percentage refers to the proportion of the clothing pixel area to the entire passenger target area, reflecting the degree to which clothing covers the whole body.

[0062] Weighted conversion refers to a mathematical operation that proportionally adjusts the baseline thermal resistance value based on the size of the coverage area.

[0063] Clothing characteristic parameters refer to the final generated numerical indicators that represent the passenger's current overall heat insulation capacity, and are an important basis for correcting air conditioning settings.

[0064] This invention uses computer vision technology to automatically identify passenger clothing categories and calculate thermal resistance based on the coverage area, achieving seamless and automatic quantification of the human body's heat exchange barrier performance. This provides precise data support for improving the accuracy of air conditioning systems in adapting to individual comfort in different clothing scenarios.

[0065] In an optional embodiment of the present invention, the physiological change information includes facial skin temperature change information, chest skin temperature change information, neck skin temperature change information, arm skin temperature change information, heart rate variability low-frequency to high-frequency ratio, and respiratory rate change information. The initial physiological feature sequence includes a chest skin temperature feature sequence characterizing the chest skin temperature change information, a neck skin temperature feature sequence characterizing the neck skin temperature change information, an arm skin temperature feature sequence characterizing the arm skin temperature change information, a heart rate variability sequence characterizing the heart rate variability low-frequency to high-frequency ratio, and a respiratory rate sequence characterizing the respiratory rate change information. The initial ambient temperature feature sequence includes a temperature change trend feature sequence characterizing the ambient temperature change trend. The step of constructing an individual dynamic thermal comfort evaluation model based on the initial physiological feature sequence and the initial ambient temperature feature sequence synchronized by the timestamp includes: A weighted summation calculation is performed on the facial skin temperature feature sequence, the chest skin temperature feature sequence, the neck skin temperature feature sequence, and the arm skin temperature feature sequence to generate a dynamic average skin temperature sequence; Obtain metabolic baseline information, as well as a thermal comfort state heart rate variability sequence for characterizing the ratio of low-frequency to high-frequency heart rate variability under thermal comfort conditions, and a thermal comfort state respiratory rate sequence for characterizing the ideal respiratory rate under thermal comfort conditions. Using the basal metabolic rate baseline, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, and the thermal comfort state respiratory rate sequence, and combining the clothing characteristic parameters to perform nonlinear compensation calculations, a dynamic human metabolic rate sequence reflecting the real-time heat production capacity of the human body is obtained. Based on the temperature change trend feature sequence and the clothing feature parameters, a correction coefficient is determined, and the maximum fluctuation range of the facial skin temperature change information, the chest skin temperature change information, the neck skin temperature change information, the arm skin temperature change information, the low-frequency to high-frequency ratio of heart rate variability, and the respiratory rate change information is determined. The comprehensive rate of change of core physiological parameters is determined based on the dynamic average skin temperature sequence, the correction factor, the maximum fluctuation range, the baseline value of basal metabolic rate, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, and the thermal comfort state respiratory rate sequence. The target skin temperature is determined based on the temperature change trend feature sequence, and weighting coefficients corresponding one-to-one with skin temperature, heart rate, and respiratory rate are determined. An individual dynamic thermal comfort evaluation model is constructed based on the clothing characteristic parameters, the weighting coefficients, the target skin temperature, the dynamic average skin temperature sequence, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, the thermal comfort state respiratory rate sequence, the maximum fluctuation range, and the comprehensive rate of change of the core physiological parameters.

[0066] In this embodiment of the invention, a dynamic average skin temperature sequence can be calculated to fuse local skin temperature characteristics distributed across different parts of the human body. Since different parts of the human body have varying sensitivities to environmental heat stimuli, weighted calculations can yield a comprehensive dynamic index that more accurately reflects the overall body heat load, providing core temperature input for subsequent metabolic rate calculations and model construction.

[0067] Facial / chest / neck / arm skin temperature feature sequence refers to the sequence of temperature values ​​of specific anatomical parts of the human body over time during the sampling period, obtained through infrared or contact sensors.

[0068] The dynamic average skin temperature series refers to a time series that reflects the dynamic trend of the whole body average temperature, generated by weighting and summing multiple local skin temperatures according to their contribution weights to the overall thermal sensation.

[0069] In this embodiment of the invention, by acquiring physiological baselines and thermal comfort reference states, a "comfort reference system" can be provided to the system by introducing "baseline values" and "ideal reference state sequences." This enables subsequent steps not only to monitor physiological changes but also to quantify the degree to which the current physiological state deviates from the ideal state, forming the basis for individualized and precise evaluation.

[0070] The baseline value of basal metabolic rate refers to the energy metabolism level of the target user in a quiet, awake state, unaffected by the environment, and serves as the initial base for individual heat balance calculation.

[0071] Thermal comfort heart rate variability / respiratory rate sequence refers to a pre-calibrated reference sequence of heart rate variability (LF / HF) and respiratory rate values ​​corresponding to the user's most comfortable state.

[0072] In this embodiment of the invention, by calculating the dynamic human metabolic rate sequence, and by combining dynamic physiological signals such as heart rate and respiration, and by introducing the key thermal resistance variable of "clothing" for compensation, the actual metabolic heat generated by the human body due to activity or stress can be calculated in real time, thus solving the drawback of the traditional model that treats the metabolic rate as a constant value.

[0073] Nonlinear compensation calculation refers to a dynamic correction algorithm that takes into account the hindering effect of clothing thermal resistance on heat dissipation and the complex mapping relationship between physiological signals and heat generation.

[0074] Heart rate variability (LF / HF) sequences refer to the ratio of low-frequency to high-frequency power that reflects the variation in the interval between heartbeats. They are often used to characterize the balance between the sympathetic and parasympathetic nervous systems.

[0075] The dynamic human metabolic rate sequence refers to the time-fluctuating heat production power sequence calculated by combining the basal thermogenesis level and the immediate physiological stress feedback.

[0076] In this embodiment of the invention, by determining the correction coefficient and the maximum fluctuation range, the scaling ratio (correction coefficient) of the model parameters and the allowable variation range of physiological indicators can be dynamically adjusted based on the changing trend of the external ambient temperature and the thickness of the user's clothing, so as to ensure that the evaluation logic conforms to the current physical environment.

[0077] Temperature change trend characteristic sequence refers to the characteristic sequence that characterizes the speed and direction of the rise, fall or stagnation of ambient temperature.

[0078] The correction factor is a coefficient used to adjust the weight or scaling of physiological parameters in the evaluation model, and is affected by environmental trends and clothing.

[0079] The maximum fluctuation range refers to the maximum positive or negative deviation limit of various physiological parameters within the evaluable range under specific conditions.

[0080] This invention, through determining the comprehensive rate of change of core physiological parameters, moves beyond focusing on changes in a single dimension and instead couples the evolutionary trends of multiple dimensions such as temperature, respiration, and heart rate in a multidimensional manner. The results of this step directly reflect the human body's "response speed" and "adaptation intensity" to environmental changes, representing a key technical solution to the problem of delayed thermal perception.

[0081] The core physiological parameters refer to the set of indicators (skin temperature, heart rate, and respiration) selected in this protocol that have the highest weighting on thermal comfort.

[0082] The comprehensive rate of change refers to the evolution vector of multidimensional physiological indicators per unit time, reflecting the transient change trend of human thermal senses.

[0083] In this embodiment of the invention, by determining target values ​​and weight coefficients, an ideal "target state" can be mapped back through environmental trends, and the weight of each physiological dimension in the final model can be dynamically adjusted. For example, the weight of skin temperature can be increased during drastic temperature changes, and the weight of heart rate can be increased during resting conditions, enabling the model to have scene adaptation capabilities.

[0084] Target skin temperature refers to the ideal skin temperature value when the human body reaches dynamic equilibrium under the current environmental evolution trend.

[0085] The weighting coefficient refers to the proportion used to adjust the contribution of different physiological dimensions to the final thermal comfort evaluation index.

[0086] In this embodiment of the invention, an individual dynamic thermal comfort evaluation model can be constructed to establish a mathematical logic system capable of outputting individualized and dynamic thermal sensation scores in real time by nonlinearly correlating all the aforementioned static benchmarks, dynamic sequences, evolution trends, and weighting factors. Specifically, the individual dynamic thermal comfort evaluation model refers to a mathematical mapping model that takes multidimensional temporal physiological parameters and environmental correction parameters as inputs and outputs quantified thermal comfort perception values.

[0087] This invention achieves dynamic capture of individual metabolic levels and transient thermal perception deviations by weighted fusion of multidimensional skin temperature and synergistic characterization of the comprehensive change rate of physiological parameters. This significantly improves the real-time performance and accuracy of air conditioning systems in evaluating the thermal comfort status of different individuals under complex temperature variations.

[0088] In an optional embodiment of the present invention, the step of calculating the real-time thermal comfort index of the target user through the individual dynamic thermal comfort evaluation model includes: According to a preset time cycle, the system cyclically acquires an instant physiological feature sequence to characterize the user's instant physiological changes within the current time cycle, and an instant environmental temperature feature sequence to characterize the user's environment's instant environmental temperature changes within the current time cycle. Acquire real-time images of passengers inside the vehicle; Perform multi-dimensional extraction of clothing features on the real-time in-vehicle passenger image information to obtain real-time clothing feature parameters that characterize the current passenger's heat exchange barrier performance; A timestamp synchronization operation is performed on the real-time physiological feature sequence and the real-time ambient temperature feature sequence, and based on the real-time physiological feature sequence and the real-time ambient temperature feature sequence after the timestamp synchronization operation, the real-time dynamic average skin temperature sequence, the real-time dynamic human metabolic rate sequence, and the comprehensive change rate of the real-time core physiological parameters of the target user are calculated. The real-time clothing characteristic parameters, the real-time dynamic average skin temperature sequence, the real-time dynamic human metabolic rate sequence, and the combined rate of change of the real-time core physiological parameters are input into the individual dynamic thermal comfort evaluation model to control the output of the real-time thermal comfort index of the individual dynamic thermal comfort evaluation model.

[0089] In this embodiment of the invention, real-time physiological and environmental feature sequences can be acquired cyclically. By using high-frequency cyclic sampling, the data source for subsequent calculations can be ensured to have extremely high timeliness, and the minute perturbations of physiological indicators and physical environment within a very short time can be captured.

[0090] The preset time period refers to the fixed sampling frequency set in the air conditioning control system, which is used to define the time interval for data updates.

[0091] The user refers to the direct recipient of the vehicle's air conditioning system adjustment service.

[0092] The current time period refers to the most recent data acquisition cycle that the system is currently in.

[0093] Real-time physiological change information refers to the real-time changes in various bioelectrical signals or body surface parameters generated by the user's body at the moment of sampling.

[0094] Real-time physiological feature sequence refers to a set of real-time values ​​containing physiological parameters such as heart rate and respiration, arranged in chronological order within the current time period.

[0095] The user's environment refers to the physical field formed by the interior space of the carriage that directly affects the user's perceived heat exchange.

[0096] Real-time ambient temperature change information refers to the temperature fluctuation value of a local space inside the vehicle at the current sampling moment.

[0097] The real-time ambient temperature feature sequence refers to a set of ambient temperature values ​​that are continuously changing over time and are collected within the current period.

[0098] According to the embodiments of the present invention, real-time in-vehicle passenger image information can be obtained to acquire real-time external morphological information of passengers through image sensors, providing raw visual material for identifying the thickness and coverage of their clothing.

[0099] Real-time in-vehicle passenger image information refers to digital video frames or still images captured by the vehicle's onboard camera at the current moment, which include the torso and limb features of passengers.

[0100] The embodiments of the present invention can generate clothing feature extraction and parameters, so as to quantify the external resistance (thermal resistance) of human body heat dissipation by analyzing the properties of clothing in the image, so that the system can understand the different thermal sensations caused by the difference in clothing.

[0101] Real-time clothing feature parameters refer to numerical values ​​that quantify the heat exchange barrier capacity (i.e., thermal resistance) of the clothing layer on the human body surface, based on the current image recognition.

[0102] Multi-dimensional clothing feature extraction refers to the process of comprehensively extracting features such as clothing category, thickness, material, and coverage area through various algorithms such as image recognition, semantic segmentation, and texture analysis.

[0103] Passenger heat exchange barrier performance refers to the ability of the clothing layer on the surface of the human body to prevent body heat from dissipating into the environment, and is usually proportional to the thermal resistance of the clothing.

[0104] In this embodiment of the invention, time-stamp synchronization and real-time dynamic index calculation can eliminate spatiotemporal asynchronous errors between data. By aligning the time nodes of physiological and environmental data, deep dynamic indicators reflecting the body's internal energy production, surface heat dissipation status, and physiological stress trends can be calculated.

[0105] Timestamp synchronization refers to aligning data collected by different sensors using a unified time base to ensure that physiological data and environmental data at the same point in time correspond one-to-one.

[0106] Real-time dynamic average skin temperature sequence refers to a set of time-series values ​​that are calculated in real time based on synchronized data, reflecting the changes in the average surface temperature of the whole body at that moment.

[0107] Real-time dynamic human metabolic rate sequence refers to the sequence of energy metabolism intensity of a user at the current moment, calculated in real time based on dynamic physiological signals such as heart rate and respiration.

[0108] The combined rate of change of real-time core physiological parameters refers to the rate of coordinated change of multiple key physiological dimensions such as temperature and metabolism within the current cycle, reflecting the intensity of physiological state transitions.

[0109] In this embodiment of the invention, by controlling the model input and index output, all processed real-time dynamic data (including human physiological state, heat production state, heat dissipation resistance and evolution trend) are input into the model to complete the final mapping from physical quantities to subjective comfort index.

[0110] Input refers to the data transmission process of passing the calculated feature parameters to the algorithm model interface.

[0111] An individual dynamic thermal comfort evaluation model refers to a mathematical logic model constructed based on a preset algorithm that can infer the real-time subjective thermal sensation of the human body from multidimensional physiological dynamic parameters.

[0112] The real-time thermal comfort index refers to a value output by the model that can accurately quantify the user's subjective thermal sensation (such as hot, cold, or comfortable) at the current instant.

[0113] This invention, by introducing real-time clothing feature parameters on the basis of real-time physiological and environmental feature sequences and combining them with multi-dimensional dynamic index calculations after timestamp synchronization, enables the air conditioning system to achieve comprehensive perception and real-time quantitative assessment of individual dynamic thermal sensation under different clothing coverage environments.

[0114] In an optional embodiment of the present invention, the real-time physiological feature sequence includes a real-time heart rate variability sequence and a real-time respiratory rate sequence, the real-time ambient temperature feature sequence includes a temperature change trend feature sequence, and the step of determining the target control parameters for the air conditioner based on the real-time thermal comfort index, and the real-time physiological feature sequence and the real-time ambient temperature feature sequence synchronized with timestamps includes: Based on the clothing characteristic parameters, the real-time thermal comfort index, the comprehensive rate of change of the instantaneous core physiological parameters, the instantaneous dynamic average skin temperature sequence, the instantaneous dynamic human metabolic rate sequence, the instantaneous heart rate variability sequence, and the instantaneous respiratory rate sequence, a multi-objective optimization function is constructed. Based on the temperature change trend characteristic sequence, determine the lower and upper limits of the target temperature inside the vehicle, the lower and upper limits of the target relative humidity inside the vehicle, and the lower and upper limits of the fresh air volume. The lower and upper limits of the target temperature inside the vehicle, the lower and upper limits of the target relative humidity inside the vehicle, and the lower and upper limits of the fresh air volume are determined as the constraints of the multi-objective optimization function. The multi-objective optimization function is solved based on the constraints to obtain the target supply air temperature, target relative humidity, target fresh air volume, supply air velocity, and supply air direction.

[0115] In this embodiment of the invention, a multi-objective optimization function can be constructed to deeply couple objective physiological parameters affecting thermal sensation (such as heart rate, metabolism, and skin temperature), subjective perception indicators (thermal comfort index), and external physical barrier factors (clothing thermal resistance). Through this function, the system can comprehensively consider the dynamic balance between heat production and heat dissipation of the human body, thereby finding an air conditioning regulation combination scheme that can simultaneously satisfy multiple physiological indicators within the ideal range.

[0116] Clothing characteristic parameters refer to numerical values ​​that quantify the ability of a passenger's current clothing to impede heat exchange (such as thermal resistance).

[0117] The real-time thermal comfort index is a numerical value output by an evaluation model that quantifies the user's current subjective thermal sensation (degree of hotness or coldness).

[0118] The rate of change of real-time core physiological parameters refers to the rate at which multiple key physiological dimensions (such as skin temperature and metabolism) evolve synergistically within the current cycle.

[0119] Real-time dynamic average skin temperature sequence refers to the set of data collected so far that reflects the change of average surface temperature of the whole body over time.

[0120] Real-time dynamic human metabolic rate sequence refers to a set of values ​​calculated in real time that reflects the fluctuation of heat production intensity per unit time in the human body.

[0121] Real-time heart rate variability sequence / real-time respiratory rate sequence refers to time-series data that reflects the state of the autonomic nervous system and respiratory rhythm at the current moment.

[0122] A multi-objective optimization function is a mathematical formula that aims to simultaneously optimize multiple conflicting or related objectives (such as maximizing comfort or minimizing physiological deviation).

[0123] In this embodiment of the invention, the limits of in-vehicle environmental parameters can be determined. By determining the upper and lower limits of temperature, humidity and fresh air volume, it can be ensured that the air conditioner will not adjust the in-vehicle environment to an unreasonable or unhealthy range (such as too low fresh air volume or too large temperature fluctuation) while pursuing extreme user comfort, thus providing boundary conditions for optimization calculation.

[0124] The lower and upper limits of the target temperature / relative humidity / fresh air volume inside the vehicle refer to the lowest and highest range of physical values ​​that the air conditioning system is allowed to adjust to based on the current environmental background and human health standards.

[0125] This invention provides an embodiment of the invention that determines the constraints of a multi-objective optimization function, transforming physical constraints into mandatory criteria in a mathematical model. By using the determined environmental limits as constraints, it ensures that subsequent solution algorithms search for the optimal solution only within a legal space that satisfies occupant health and system capabilities.

[0126] Constraints refer to specific equality or inequality restrictions that variables must satisfy during mathematical optimization, used to filter out invalid or infeasible solutions.

[0127] In this embodiment of the invention, a solution operation can be performed to obtain the target control parameters. By mathematically solving a complex optimization function, high-dimensional physiological trend information can be transformed into five physical target parameters that the air conditioning system can directly execute. This process achieves a precise leap from "human physiological needs" to "machine execution parameters".

[0128] The solution operation refers to the process of using specific optimization algorithms (such as gradient descent, evolutionary algorithms, etc.) to calculate the optimal combination of variables that makes the function value optimal under the premise of satisfying constraints.

[0129] Target supply air temperature / humidity / fresh air volume / wind speed / wind direction refer to the specific physical parameter target values ​​that the air conditioning system needs to achieve at the air outlet in order to reach the optimal comfort state.

[0130] This invention, through the construction of a multi-objective optimization function based on multidimensional physiological evolution trends and combined with environmental boundary constraints, realizes the transformation of regulation parameters from single empirical control to collaborative decision-making based on "physiological trends and environmental constraints," significantly improving the system coordination and response accuracy of air conditioning regulation in meeting individual dynamic comfort needs.

[0131] Optionally, the step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Obtain the current supply air temperature value and the compressor frequency conversion characteristic information of the air conditioner; the compressor frequency conversion characteristic information includes the compressor reference frequency; Calculate the difference between the target supply air temperature value and the current supply air temperature value; The target compressor frequency at the target time is calculated based on the difference, the preset temperature adjustment coefficient, and the compressor reference frequency. The compressor control command is generated using the target compressor frequency at the target time. Based on the compressor control commands, the compressor is driven to perform air supply temperature adjustment.

[0132] In this embodiment of the invention, the air supply status and frequency conversion characteristic information can be obtained. By obtaining the actual temperature of the current air outlet of the air conditioner and the frequency conversion characteristics of the hardware itself, the system can understand the current status and the hardware capability boundary, ensuring that the frequency adjustment amount calculated subsequently conforms to the physical logic and can give full play to the performance of the frequency conversion compressor.

[0133] The current supply air temperature value refers to the actual air temperature at the air conditioner outlet measured by the temperature sensor at the current moment.

[0134] Compressor variable frequency characteristic information refers to the set of physical properties of a compressor that adjust its speed according to the input frequency to regulate its cooling / heating capacity.

[0135] The compressor reference frequency refers to the reference frequency value of the compressor when it is operating stably under standard operating conditions, and it serves as the starting base for frequency regulation calculations.

[0136] In this embodiment of the invention, the supply air temperature difference can be calculated. By calculating the algebraic difference between the target value and the current value, the absolute amount that the air conditioning system needs to compensate in the temperature dimension can be determined. This difference directly determines the magnitude of subsequent compressor frequency increase or decrease.

[0137] The target supply air temperature value refers to the ideal supply air temperature that meets the thermal comfort requirements of the occupants, calculated by the aforementioned multi-objective optimization.

[0138] The difference refers to the numerical offset between the target supply air temperature and the current actual supply air temperature.

[0139] In this embodiment of the invention, the target frequency of the compressor at a target time can be calculated. By introducing an adjustment coefficient, the system can dynamically adjust the target speed of the compressor according to the size of the temperature difference, thereby achieving rapid and stable temperature convergence and avoiding large temperature fluctuations.

[0140] The preset temperature regulation coefficient is a proportional factor used to balance regulation speed and stability, which determines the sensitivity of temperature difference in frequency change.

[0141] The target compressor frequency at the target time refers to the optimal operating frequency that the compressor should achieve in order to eliminate the temperature difference at a specific point in the future.

[0142] In this embodiment of the invention, compressor control commands can be generated and executed. By generating standard low-level control protocol commands and driving the hardware, the system completes the transition from digital decision-making to mechanical execution. Ultimately, by changing the compressor speed, the circulating refrigerant flow rate is adjusted to achieve the purpose of changing the air supply temperature.

[0143] Compressor control commands refer to specific electronic control commands that conform to vehicle communication protocols (such as CAN or LIN bus) and are sent to the compressor controller.

[0144] The air supply temperature regulation action refers to the process in which the compressor changes its speed according to the instruction, causing a change in the state of the refrigerant and ultimately resulting in a change in the temperature of the air at the outlet.

[0145] This invention combines the inverter characteristics of the compressor with temperature difference feedback to calculate the target frequency, thereby achieving a smooth response of the air conditioner's outlet temperature to the dynamic thermal comfort needs of the human body, ensuring the efficiency and stability of the environmental regulation process.

[0146] Optionally, the step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Get the current relative humidity value inside the vehicle; When it is determined that the target relative humidity value is lower than the current in-vehicle relative humidity value, the difference between the current in-vehicle relative humidity value and the target relative humidity value is calculated; Obtain information on air density, cooling space volume, and water specific heat capacity; The dehumidification power at the target time is calculated based on the difference, the air density information, the cooling space volume information, and the specific heat capacity information of water. The dehumidification power at the target time is used to generate control commands for the dehumidification device; The dehumidifier of the air conditioner is driven to adjust the humidity based on the control command of the dehumidifier.

[0147] In this embodiment of the invention, the current relative humidity value inside the vehicle can be obtained. By measuring the current physical state through sensors, the system can obtain a data basis for determining whether the dehumidification function needs to be turned on, ensuring that the triggering of the humidity adjustment logic has an objective basis.

[0148] The current relative humidity value inside the vehicle refers to the measured percentage of the partial pressure of water vapor in the air inside the vehicle at the current moment compared to the saturated water vapor pressure at the same temperature.

[0149] This invention can determine dehumidification needs and calculate humidity differences to logically determine whether the current environment is more humid than ideal, and quantify the excess. This difference is a core variable in subsequent energy calculations, determining how much moisture the system needs to remove from the air.

[0150] The target relative humidity value refers to the ideal air humidity target that enables the human body to achieve a state of thermal comfort balance, calculated by the aforementioned multi-objective optimization.

[0151] The difference refers to the absolute numerical deviation between the current measured humidity value and the target humidity value.

[0152] This invention provides an embodiment of physical environment parameter information to collect various environmental constants and spatial parameters required for performing thermodynamic calculations. This information defines the physical properties of the target to be adjusted (such as air mass, heat capacity, and space size), and is a necessary boundary condition for achieving accurate power calculations.

[0153] Air density information refers to the mass of air per unit volume, and is used to convert volume parameters into mass parameters.

[0154] The refrigeration space volume information refers to the effective enclosed space volume inside the vehicle that requires air conditioning.

[0155] The specific heat capacity of water refers to the amount of heat absorbed or released when a unit mass of water changes its temperature by a unit amount. In this context, it is used to measure the energy conversion characteristics during the condensation dehumidification process.

[0156] This invention allows for the calculation of dehumidification power at a target time. By mathematically coupling humidity deviation with physical environmental parameters, the rate of energy change required to reduce the vehicle's interior humidity to a target level can be calculated. This result achieves a quantitative conversion from "environmental condition requirements" to "equipment power requirements."

[0157] The dehumidification power at target time refers to the power output required by the dehumidification device at a specific time to achieve the humidity regulation target.

[0158] In this embodiment of the invention, control instructions for a dehumidification device can be generated and executed to convert abstract power values ​​into low-level control signals that can be recognized by the dehumidification device (such as the evaporator of an air conditioning system or a dedicated dehumidification unit). By driving the hardware to run, the physical regulation of the humidity inside the vehicle can be ultimately achieved.

[0159] Dehumidifier control commands refer to electrical control commands sent to actuators (such as compressor duty cycle signals or electronic expansion valve adjustment signals).

[0160] Humidity regulation refers to the process by which the water vapor content in the air is changed through physical means (such as condensation) after the actuator is in operation.

[0161] This invention combines real-time humidity feedback inside the vehicle with spatial physical characteristics to accurately calculate dehumidification power, enabling on-demand adjustment of the humidity environment inside the vehicle. This ensures that passengers feel dry and comfortable while avoiding excessive operation of the dehumidification device.

[0162] Optionally, the step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Obtain the minimum and maximum fresh air volume; Calculate the first difference between the target fresh air volume value and the minimum fresh air volume value; Calculate the second difference between the maximum fresh air volume and the minimum fresh air volume; The percentage of the fresh air valve opening is calculated based on the first difference and the second difference; The opening percentage of the fresh air valve is used to adjust the opening of the fresh air valve of the air conditioner.

[0163] The embodiments of the present invention can obtain the limit value of fresh air volume, so as to provide a range benchmark for subsequent percentage conversion by obtaining the minimum value of fresh air volume (usually the minimum air volume to meet basic breathing needs) and the maximum value (the maximum air volume that the system hardware can provide), and ensure that the fresh air adjustment command is within the physical capability range of the equipment.

[0164] Minimum fresh air volume refers to the minimum amount of outside air that the air conditioning system is allowed to introduce in order to maintain the freshness and oxygen content of the air inside the vehicle.

[0165] The maximum fresh air volume refers to the maximum amount of external air that the system can provide when the air conditioner's fresh air valve is fully open and the fan is operating under the corresponding conditions.

[0166] The embodiments of the present invention can calculate the first difference in fresh air volume, so as to determine how much additional fresh air supply is needed compared to the minimum guaranteed level under the current environmental and physiological needs by calculating the difference between the target air volume and the minimum reference air volume.

[0167] The target fresh air volume refers to the ideal fresh air intake calculated by the aforementioned multi-objective optimization, which can take into account both passenger thermal comfort and air quality requirements.

[0168] The first difference refers to the linear difference between the target fresh air volume value and the minimum fresh air volume value.

[0169] The embodiments of the present invention can calculate the second difference of fresh air volume, so as to obtain the total air volume change space corresponding to the fresh air valve from fully closed (or minimum adjustment position) to fully open by calculating the difference between the maximum value and the minimum value, and use it as the denominator for calculating the proportional weight.

[0170] The second difference refers to the total adjustment range of the fresh air system, that is, the difference between the maximum fresh air volume and the minimum fresh air volume.

[0171] This invention can calculate the percentage of the opening of the fresh air valve to convert the abstract "air volume requirement" into a specific "valve mechanical position". By calculating the ratio of the first difference to the second difference, an opening value between 0% and 100% is obtained, realizing a standardized conversion from physical quantity target to hardware execution target.

[0172] The percentage of the opening of the fresh air valve refers to the relative displacement ratio between the fully closed and fully open positions of the fresh air regulating valve, which is a direct input parameter for controlling the valve drive motor.

[0173] This invention embodiment can adjust the opening of the fresh air valve to enable the system to drive the fresh air door motor to move according to the calculated percentage command, thereby changing the cross-sectional area of ​​the physical channel through which external air enters the vehicle, and thus changing the amount of fresh air introduced in real time, completing the closed-loop regulation of the fresh air freshness in the vehicle.

[0174] A fresh air valve is a mechanical damper located at the air inlet of an air conditioner, used to control the ratio of fresh external air to recirculated internal air.

[0175] Adjustment refers to the process by which the actuator changes the physical position of the damper according to the opening command.

[0176] This invention achieves precise proportional control of the amount of fresh air introduced by dynamically calculating the valve opening percentage based on the fresh air volume range. This ensures the air quality inside the vehicle while avoiding ambient temperature fluctuations and energy consumption caused by excessive fresh air introduction.

[0177] To enable those skilled in the art to better understand the embodiments of the present invention, an example is used below to illustrate the embodiments of the present invention.

[0178] This invention provides a vehicle air conditioning thermal comfort intelligent control method based on dynamic physiological parameters, applied to a vehicle air conditioning system. The vehicle air conditioning system includes a physiological parameter acquisition unit, an environmental parameter acquisition unit, a control center, and an execution unit. The method includes the following steps: Step 1: Use the physiological parameter acquisition unit to collect at least one set of dynamic physiological parameters for each passenger in the vehicle in real time, and use the environmental parameter acquisition unit to collect environmental parameters inside and outside the vehicle. Then, perform timestamp synchronization and preprocessing on the collected physiological and environmental parameters. Step 2: Based on the preprocessed dynamic physiological parameters, construct an individual dynamic thermal comfort evaluation model and calculate the real-time thermal comfort index for each passenger. Step 3: Based on the real-time thermal comfort index, dynamic physiological parameter change rate, and environmental parameters of all passengers, construct a multi-objective optimization function and solve for the target control parameters of the vehicle air conditioning system. Step four: The control center generates an execution command based on the target control parameters and sends it to the execution unit. The execution unit responds to the command and adjusts the operating status of the air conditioning system.

[0179] In summary, this invention utilizes a physiological parameter acquisition unit to collect at least one set of dynamic physiological parameters for each passenger in real time, and an environmental parameter acquisition unit to collect in-vehicle and external environmental parameters. The collected physiological and environmental parameters are time-stamped and preprocessed. Based on the preprocessed dynamic physiological parameters, an individual dynamic thermal comfort evaluation model is constructed to calculate the real-time thermal comfort index for each passenger. Based on the real-time thermal comfort index, the rate of change of dynamic physiological parameters, and environmental parameters of all passengers, a multi-objective optimization function is constructed to solve for the target control parameters of the vehicle's air conditioning system. The control center generates execution commands based on the target control parameters and sends them to the execution unit. The execution unit responds to the commands and adjusts the operating state of the air conditioning system. This invention establishes a precise mapping model between physiological parameters and air conditioning control parameters, improving control accuracy and real-time performance, and meeting the personalized and dynamic thermal comfort needs of passengers.

[0180] In some embodiments, the time-stamp synchronization and preprocessing of the collected physiological and environmental parameters includes: removing outliers and smoothing the time series of physiological parameters (using moving average filtering with a window size of 5s); normalizing environmental parameters and mapping parameters of different dimensions to the [0,1] interval; and associating and matching physiological and environmental parameters at the same time based on the timestamp to form a spatiotemporally synchronized parameter dataset.

[0181] In some embodiments, dynamic physiological parameters include core physiological parameters and auxiliary physiological parameters: core physiological parameters include skin temperature distribution, heart rate variability (HRV), and respiratory rate (RR), which directly reflect the body's thermoregulatory state and metabolic level; auxiliary physiological parameters include clothing thermal resistance estimates, which are used to correct thermal comfort evaluation results; environmental parameters include in-vehicle temperature, in-vehicle relative humidity, outside-vehicle temperature, solar radiation intensity, and in-vehicle CO2 concentration.

[0182] In some embodiments, Figure 2This is a system block diagram of a vehicle air conditioning system provided in an embodiment of the present invention, such as... Figure 2 As shown, the system includes: a physiological parameter acquisition unit, an environmental parameter acquisition unit, a control center, and an execution unit. The physiological parameter acquisition unit includes a distributed infrared temperature sensor, a wearable physiological monitoring module, and an image recognition module. The distributed infrared temperature sensor is deployed in the headrests and armrests of the vehicle seats for non-contact acquisition of skin temperature distribution on various parts of the passenger's body. The wearable physiological monitoring module communicates with the control center via Bluetooth to collect passenger heart rate variability and respiratory rate data. The image recognition module calculates the estimated value of clothing thermal resistance based on images acquired by the in-vehicle camera. The environmental parameter acquisition unit includes a temperature sensor, a humidity sensor, a sunlight sensor, and a CO2 sensor, which are deployed in different areas inside the vehicle and outside the vehicle, respectively.

[0183] In some embodiments, calculating the estimated value of clothing thermal resistance based on images captured by an in-vehicle camera specifically includes the following process: Step 1: Extract clothing features from multiple dimensions to obtain clothing feature parameters; Extract clothing feature parameters from the image, such as clothing area, material type, thickness distribution, and coverage area: Semantic segmentation of clothing regions: An improved U-Net++ network was selected as the segmentation model, and the network structure was optimized for the vehicle scene: The encoder uses ResNet50 as the backbone network, and the feature extraction capability is enhanced through residual connections. At the same time, the CBAM attention mechanism is introduced to highlight the features of the clothing region and suppress background interference. The decoder adopts an adaptive upsampling module, combined with multi-scale feature fusion (fusion of 1 / 2, 1 / 4, and 1 / 8 scale feature maps) to improve the segmentation accuracy of clothing edges (such as cuffs, collars, and trouser hems). The loss function adopts a weighted fusion of Dice loss and cross-entropy loss (weight ratio 0.7:0.3) to solve the problem of the imbalance of pixel ratio between clothing region and background region. Model Training: A vehicle-mounted clothing segmentation dataset was constructed, containing 100,000 image samples with different lighting, poses, and clothing types. Each sample was labeled with three categories: clothing region, skin region, and background region. The dataset was divided into training, validation, and test sets in an 8:1:1 ratio. The AdamW optimizer was used for 150 epochs of training with corresponding learning rates. An early stopping mechanism was triggered when the Intersection over Union (IoU) on the validation set showed no improvement for 10 consecutive epochs. The final model achieved an IoU ≥ 0.92 for clothing segmentation on the test set. Post-Segmentation Processing: Morphological operations were performed on the segmentation results. Dilation-erosion was used to remove small holes (regions with an area < 50 pixels), and a contour smoothing algorithm was used to optimize the clothing edges to obtain a complete clothing region mask.

[0184] Clothing Material Classification and Recognition: Based on the segmented clothing regions, a two-branch CNN model is used for material recognition: Feature Extraction Branch: MobileNetV3-Small is used as the base network, with an added SE attention module to enhance the extraction of material texture features (such as the fiber texture of cotton, the smooth texture of synthetic fibers, and the fluff texture of wool), and multi-scale pooling (1×1, 3×3, 5×5) is introduced to capture material information at different scales; Classification Branch: Through fully connected layers and the Softmax activation function, the probability distribution of 10 common clothing materials in automotive scenarios is output, including cotton, linen, synthetic fibers, wool, silk, leather, down, knitwear, denim, and blended fabrics; Model Training: A clothing material dataset is constructed, containing 80,000 images of clothing of the above 10 materials, with each image labeled with material category and texture feature labels; a transfer learning strategy is adopted, loading MobileNetV3-Small... The pre-trained weights on ImageNet were used, with the parameters of the first 8 network layers frozen, training only the last 5 layers and the classification layer. During training, data augmentation techniques such as random flipping, rotation (±10°), and brightness adjustment (±15%) were employed to avoid overfitting. The final model achieved a material classification accuracy ≥0.89 on the test set. Material confidence calibration: The classification results were post-processed, calculating the probability difference between the two materials with the highest confidence scores. If the probability difference was <0.2, further calibration was performed using auxiliary features such as clothing color and texture complexity, outputting the final material type and confidence factor C. MAT (The confidence factor ranges from [0,1], with higher confidence levels resulting in larger values).

[0185] Refined estimation of clothing thickness: Based on a combination of binocular vision principles and deep learning, the thickness values ​​of different areas of clothing are estimated. Stereo matching and disparity map calculation: For the stereo images after epipolar correction, the SIFT algorithm is used to extract feature points of the clothing region. Initial matching is performed using the FLANN matcher, and then the RANSAC algorithm is used to remove mismatched points (the proportion of inliers retained is ≥0.85). The semi-global block matching (SGBM) algorithm is used to calculate the dense disparity map of the clothing region. The block size is set to 9×9 and the disparity range is 0-64 pixels. The disparity consistency is optimized by dynamic programming.

[0186] Initial thickness calculation: Based on the principle of binocular vision ranging, the distance from each point on the clothing surface to the camera is calculated using the disparity map. Combined with the reference distance on the human body surface (obtained through human posture estimation), the initial thickness value is obtained, as shown in the following formula: Based on the binocular vision ranging formula, the distance from a point on the clothing surface to the camera is calculated using parallax: ; The distance from the surface of the clothing to the camera; Based on the distance from the human body surface to the camera, the initial thickness is the difference between the two: ; in, For the initial thickness, The distance from the human body surface to the camera (calculated by detecting the human body contours through key point detection). For the surface of clothing Parallax, For camera focal length, The baseline length of the binocular camera (pre-calibrated to 120mm).

[0187] Clothing coverage area quantification: Precisely calculate the area and proportion of clothing covering the human body. Human Keypoint Detection and Contour Extraction: The YOLO-Pose algorithm was used to detect 17 key human landmarks (top of head, acromion, elbow, wrist, hip, knee, ankle, etc.), with a detection confidence threshold set to 0.7. Occluded keypoints were filled in using prior knowledge of human pose. Based on the keypoints, the Delaunay triangulation algorithm was used to construct the human surface contour. Combined with the skin region mask obtained from semantic segmentation, the human body surface regions were distinguished. and exposed skin areas ; Coverage area and proportion calculation: Two-dimensional coverage area Considering the three-dimensional curved surface characteristics of the human body, the three-dimensional surface area is calculated by constructing a simplified three-dimensional human body model (the torso is a cylinder and the upper limbs are frustums). Then the three-dimensional coverage area is obtained. Then the overall clothing coverage ratio is .

[0188] Step 2: Based on the clothing feature parameters and material thermal property database, calculate the estimated value of clothing thermal resistance through a fusion model driven by physical mechanisms; Construction of a database of thermal properties of clothing materials: This involves collecting thermal property parameters of 10 common clothing materials, including thermal conductivity, specific heat capacity, and density. It also records thermal resistance correction factors for different thicknesses (e.g., for cotton materials, thermal resistance increases by 1 mm for every 1 mm increase in thickness). Establish a dynamic database update mechanism; if the material identification confidence factor C... MAT If the sample is not labeled, it will be added to the dataset to be labeled, and the database parameters will be updated after manual labeling.

[0189] Multi-parameter confidence quantification: Thickness confidence factor RMSE and regional thickness standard deviation based on thickness estimation The calculation formula is as follows: ,in, Global average thickness, C h ∈[0,1], the higher the value, the more reliable the thickness estimation.

[0190] Coverage area confidence factor : Success rate based on key node detection The calculation formula is as follows: ,in, Let CS be the area of ​​the missing contour region, ∈ [0,1], where the success rate of key node detection is... The calculation is as follows: Seventeen key human body nodes are preset (top of head, brow bone, eyes, nose, mouth, acromion, elbow, wrist, hip, knee, ankle, etc.). Successfully detected nodes must meet the condition that "the Euclidean distance between the detected coordinates and the actual coordinates is ≤ a preset threshold" (the preset threshold is set to 20 pixels in the vehicle scene, corresponding to an actual physical distance of 2mm). ;in, To meet the distance threshold for the number of critical nodes, The preset total number of key nodes is succeeds∈[0,1]. A higher value indicates a better detection effect.

[0191] Based on the initial thickness at various points on the clothing surface The average thickness of the area containing each point on the clothing surface was calculated. ; Calculate the estimated value of clothing thermal resistance : Among them, K MAT This is the thermal conductivity of the material to which the clothing is made.

[0192] In some embodiments, in step two, based on the preprocessed dynamic physiological parameters, an individual dynamic thermal comfort evaluation model is constructed, and the real-time thermal comfort index for each passenger is calculated, specifically including the following process: ; in, The real-time thermal comfort index of passengers at time t. Let t be the passenger's metabolic rate at time t. This is the coefficient of the human metabolic rate constant. It is a constant. , and These are the thermal comfort index weighting coefficients; in this embodiment, those skilled in the art will determine based on... Set as ,Will The value is set to 0.028, which uses the core coefficient of the traditional PMV to ensure the model's compatibility with the classic thermal comfort evaluation system. , and respectively , and Skin temperature is the most direct physiological indicator of thermal comfort, so it has a high weighting coefficient. Heart rate variability (HRV) and respiratory rate (RR) are indirect indicators of heat stress, and their weights are relatively low (0.3 each).

[0193] in, ; It's worth noting that the basal metabolic rate baseline: "80" corresponds to the baseline basal metabolic rate value under sedentary conditions (the unit is usually W / m²; the human metabolic rate under sedentary conditions is approximately 70~90 W / m², and 80 is a typical value for this scenario). The relationship between physiological signals and metabolic rate: Heart rate variability: Decreased HRV usually reflects stress states (including heat stress), therefore... This indicates that "a decrease in HRV corresponds to an increase in metabolic rate"; respiratory rate (RR): the resting respiratory rate is approximately 15 breaths / minute. This represents the change in metabolic rate when the respiratory rate deviates from the resting value. These coefficients (2.5, 1.8) were obtained by fitting the correlation analysis between measured metabolic rates of passengers under different thermal environments (such as indirect calorimetry) and physiological signals, and are suitable for dynamic estimation of the metabolic rate of seated passengers in the vehicle.

[0194] The average skin temperature at time t is calculated by weighting the skin temperatures of different areas. ;in, , , and All are weighted coefficients for skin temperature at different locations. Let t be the passenger's facial temperature. Let t be the passenger's chest temperature. Let t be the passenger's neck temperature. Let t be the temperature of the passenger's arm. The optimal skin temperature is set at 33.5 ± 0.5℃. This range represents a crucial threshold for achieving balance between heat dissipation through skin radiation, convection, and evaporation, and metabolic heat production. Temperatures below 33℃ tend to create a feeling of "coolness," while temperatures above 33.5℃ tend to create a feeling of "heat." Therefore, The settings are determined by those skilled in the art based on actual circumstances; it is worth noting that the percentage of skin area of ​​each part of the human body in the total body is one of the core bases for traditional "average skin temperature calculation": the face and chest are areas with a relatively large body surface area (approximately 3%~5% for the face and 10%~15% for the chest), and are highly exposed (usually unobstructed in passenger scenarios), thus contributing more significantly to overall thermal perception, and therefore have higher weights (0.3 each); the neck and arms have relatively smaller body surface area (approximately 2%~3% for the neck and 10%~12% for the arms), and may be covered by clothing in some scenarios, so their weights are relatively lower (0.2 each). In this embodiment, , , and Can be set to , , and .

[0195] Let be the low-frequency / high-frequency ratio of the passenger's heart rate variability at time t. The low-frequency / high-frequency ratio of heart rate variability under thermal comfort conditions; Let t be the passenger's breathing rate. The ideal respiratory rate under thermal comfort conditions, ΔTmax, ΔHRVmax, and ΔRRmax, are the maximum fluctuation values ​​of skin temperature, heart rate variability, and respiratory rate, respectively, determined by those skilled in the art based on human physiological limits: normal human skin temperature is approximately between 32 and 35°C, while the safe fluctuation range under heat / cold stress is usually no more than 5°C. The fluctuation range of heart rate variability 2.0 (from comfort value 1.0~1.5 to stress value 3.0~3.5) covers the typical changes under heat stress while avoiding exceeding the critical range of "autonomic nervous system dysfunction." The respiratory rate (RR) at rest is approximately 12~18 breaths / minute. Breathing becomes deeper and faster under heat stress, and under mild heat stress, RR usually does not exceed 25~30 breaths / minute (an increase of 10~15 breaths compared to the resting value). The fluctuation range of 10 breaths / minute (from comfort value 15 to stress value 25) conforms to the physiological response under mild heat stress while avoiding exceeding the risk range of "hyperventilation" (usually RR>30 breaths / minute will cause discomfort). In this embodiment, ΔTmax, ΔHRVmax, and ΔRRmax are 5℃, 2.0, and 10 (times / minute), respectively. This is a correction function for dynamic changes in physiological parameters. The overall rate of change of core physiological parameters: ; ,in, In this embodiment, the correction factor can be set. It is 0.5.

[0196] In some embodiments, based on the real-time thermal comfort index, dynamic physiological parameter change rate, and environmental parameters of all passengers, a multi-objective optimization function is constructed, and the target control parameters of the vehicle air conditioning system are obtained by solving the function. The specific process includes the following steps: ; The constraints are: ; in, For multi-objective optimization functions, The number of passengers in the vehicle. , and The weights are the weight coefficients of the multi-objective optimization function. The core function of vehicle air conditioning is to ensure passenger thermal comfort. Therefore, "minimizing the mean absolute value of passenger PMV" (i.e., making most passengers close to thermal neutrality) is the highest priority objective, with the largest weight (0.5). The maximum value of the rate of change of passenger physiological parameters has the second highest weight (0.3). This is because drastic fluctuations in physiological parameters (such as skin temperature and HRV) can cause discomfort and even affect health. In the vehicle scenario, air conditioning control needs to avoid sudden changes in physiological parameters (such as stress caused by sudden temperature rises and falls). Therefore, "maintaining stable physiological parameters" is the second most important objective after thermal comfort. The weight is the proportion of air conditioning energy consumption to the maximum energy consumption, with the lowest weight (0.2). This is because the primary requirement of vehicle air conditioning is passenger experience, and energy consumption is a secondary constraint (especially in passenger car scenarios, where energy consumption priority is usually lower than comfort). This weight setting reflects the scenario logic of "minimizing energy consumption while ensuring comfort and physiological stability." In this embodiment, , and They were set to 0.5, 0.3, and 0.2 respectively. Let be the real-time thermal comfort index of the i-th passenger at time t. Let be the comprehensive rate of change of the physiological parameters of the i-th passenger at time t. Let t be the real-time power consumption of the vehicle's air conditioning system. The maximum rated power consumption of the vehicle's air conditioning system. and These are the lower and upper limits of the comfortable temperature range (in this embodiment, the comfortable temperature range is set to 20℃-28℃). Let t be the target temperature inside the vehicle. and These are the lower and upper limits of the comfortable humidity range, respectively (in this embodiment, the comfortable humidity range is set to 30%-60%). Let be the target relative humidity inside the vehicle at time t. and These are the lower and upper limits of the fresh air volume range that meets air quality requirements (in this embodiment, the fresh air volume range that meets air quality requirements is set to...). ), Let t be the fresh air volume at time t; The Particle Swarm Optimization (PSO) algorithm is used to solve the multi-objective optimization function. Output the target control parameters of the vehicle's air conditioning system, including target supply air temperature, target relative humidity, target fresh air volume, supply air velocity, and air direction: Initialize the particle swarm: Set the number of particles to 50 and the number of iterations to 30. Each particle corresponds to a set of control parameters (supply air temperature 20-28℃, humidity 30%-60%, fresh air volume 30-120m³ / h, wind speed 0.8-1.8m / s, wind direction 0-180°). Randomly generate the initial particle position and velocity.

[0197] Fitness calculation: Substitute the particle parameters into the multi-objective optimization function F, combine the real-time thermal comfort index, the rate of change of physiological parameters and energy consumption, calculate the fitness value of each particle, and select the individual optimal solution (pbest) and the global optimal solution (gbest).

[0198] Iterative update: Particle position and velocity are updated according to the PSO formula to balance global search and local exploitation and avoid getting trapped in local optima; fitness is recalculated in each iteration and pbest and gbest are dynamically updated.

[0199] Constraint verification: Ensure that the updated particle parameters meet environmental and physiological constraints (such as temperature and humidity thresholds); if not, correct them to the range of constraints.

[0200] Output results: After the iteration terminates, the parameters corresponding to gbest are used as the target control parameters, including precise air supply temperature, relative humidity, fresh air volume, and continuous wind speed. The directional wind direction is determined by combining the passenger position and the real-time thermal comfort index distribution.

[0201] Furthermore, the process by which the control center generates and executes commands based on the target control parameters includes the following steps: Supply air temperature adjustment: Based on the target supply air temperature With current supply air temperature The difference, combined with the compressor's variable frequency characteristics, is used to calculate the compressor's target frequency at time t. : ;in, The compressor's reference frequency, This is the temperature regulation coefficient; Humidity adjustment: When the target relative humidity is lower than the current relative humidity inside the vehicle, the dehumidification function is activated, and the dehumidification power at time t is calculated. : ; where ρ air air density, For the volume of the carriage, The specific heat capacity of water, This represents the current relative humidity inside the vehicle. For the target relative humidity, The dehumidification efficiency of the vehicle's air conditioning system; Fresh air volume adjustment: Adjust the opening of the fresh air valve according to the target fresh air volume. : ;in, Let t be the target fresh air volume; Wind speed and direction adjustment: Based on the passenger location distribution and physiological parameter distribution, a directional air supply strategy is adopted to adjust the air outlet angle to point to the area. The passenger location distribution and physiological parameter distribution are obtained by the physiological parameter acquisition unit.

[0202] In some embodiments, auxiliary physiological parameters, including estimated skin moisture and clothing thermal resistance, are used to correct thermal comfort evaluation results, specifically including the following processes: As an indirect correction factor for human metabolic rate: clothing thickness is indirectly related to the amount of human activity (e.g., wearing thin clothes during exercise results in a high metabolic rate; wearing thick clothes during rest results in a low metabolic rate). The estimated value of clothing thermal resistance can help calibrate the metabolic rate calculated from heart rate variability (HRV) and respiratory rate (RR), avoiding the deviation in metabolic rate estimation caused by fluctuations in physiological parameters. When clothing has high thermal resistance, even if the skin temperature is slightly higher, it may be a "false heat signal" caused by the insulation of the clothing, rather than actual thermal discomfort. Conversely, when the thermal resistance is low, a slightly lower skin temperature may be a "false cold signal" caused by excessive heat dissipation. The estimated thermal resistance of clothing is adjusted by modifying the weighting coefficients of the thermal comfort model to make the thermal comfort index more closely match the actual feeling of passengers (for example, at the same skin temperature of 33.5℃, the thermal comfort index needs to be lowered by 0.1~0.2 when the thermal resistance is high (1.2clo) to avoid excessive cooling by the air conditioner; when the thermal resistance is low (0.3clo), the thermal comfort index needs to be raised by 0.1~0.2 to avoid excessive heating by the air conditioner).

[0203] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0204] In addition, this application also provides an electronic device 30, please refer to... Figure 3It includes a processor 310 and a memory 320, wherein the memory 310 is used to store computer programs; the processor 320 is used to execute the programs stored in the memory 310 to implement the air conditioning control method described in any embodiment of this application.

[0205] like Figure 4 As shown, in another embodiment of the present invention, a computer-readable storage medium 401 is also provided, which stores instructions that, when executed on a computer, cause the computer to perform the air conditioning control method described in the above embodiment.

[0206] This invention also discloses a vehicle and the electronic device described above.

[0207] In this application, "multiple" refers to two or more.

[0208] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0209] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if any) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0210] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0211] Unless otherwise specified, all steps in this application may be performed sequentially or randomly. For example, if the method includes steps A and B, it means that the method may include steps A and B performed sequentially, or it may include steps B and A performed sequentially. For example, if the method may also include step C, it means that step C may be added to the method in any order. For example, the method may include steps A, B, and C, or it may include steps A, C, and B, or it may include steps C, A, and B, etc.

[0212] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An air conditioning control method, characterized in that, include: According to a preset time period, an initial physiological feature sequence for characterizing the user's physiological changes within the time period and an initial environmental temperature feature sequence for characterizing the user's environment temperature changes within the time period are obtained cyclically. A timestamp synchronization operation is performed on the initial physiological characteristic sequence and the initial ambient temperature characteristic sequence, and an individual dynamic thermal comfort evaluation model is constructed based on the initial physiological characteristic sequence and the initial ambient temperature characteristic sequence after the timestamp synchronization operation. The real-time thermal comfort index of the target user is calculated using the individual dynamic thermal comfort evaluation model. Based on the real-time thermal comfort index, as well as the real-time physiological characteristic sequence and real-time ambient temperature characteristic sequence synchronized with timestamps, the target control parameters for the air conditioner are determined. The target control parameters are used to generate control commands for the air conditioner, and the air conditioner is controlled based on the control commands.

2. The method according to claim 1, characterized in that, The method further includes: Acquire image information of passengers inside the vehicle; A multi-dimensional extraction operation of clothing features is performed on the in-vehicle passenger image information to obtain clothing feature parameters used to characterize the heat exchange barrier performance of passengers.

3. The method according to claim 2, characterized in that, The step of performing multi-dimensional extraction of clothing features on the in-vehicle passenger image information to obtain clothing feature parameters used to characterize the passenger's heat exchange barrier performance includes: Human key point recognition is performed on the in-vehicle passenger image information to locate and extract the passenger target area image; Perform semantic segmentation processing on the passenger target area image to determine the clothing coverage area image belonging to the clothing pixels; Texture frequency and edge gradient analysis are performed on the image of the clothing-covered area to generate material sensory feature information that characterizes the thickness of the clothing material; The sensory characteristics of the material are combined with the contour features of the image of the clothing-covered area to perform multi-dimensional feature matching, thereby determining the passenger's clothing category information; The clothing category information is matched with a preset physical thermal resistance comparison data table, and the thermal resistance reference value corresponding to the clothing category information is extracted from the preset physical thermal resistance comparison data table. The thermal resistance baseline value is combined with the area ratio of the clothing coverage area image in the passenger target area image for weighted calculation to generate clothing feature parameters that characterize the heat exchange barrier performance of the passenger.

4. The method according to claim 2 or 3, characterized in that, The steps for constructing an individual dynamic thermal comfort evaluation model based on the initial physiological characteristic sequence and the initial environmental temperature characteristic sequence synchronized by the timestamp include: A weighted summation calculation is performed on the skin temperature feature sequence used to characterize skin temperature changes in different parts of the body to generate a dynamic average skin temperature sequence. Obtain metabolic baseline information, as well as a thermal comfort state heart rate variability sequence for characterizing the ratio of low-frequency to high-frequency heart rate variability under thermal comfort conditions, and a thermal comfort state respiratory rate sequence for characterizing the ideal respiratory rate under thermal comfort conditions. Using the basal metabolic rate baseline value, the heart rate variability sequence used to characterize the ratio of low-frequency to high-frequency heart rate variability, the heart rate variability sequence of thermal comfort state, the respiratory rate sequence used to characterize respiratory rate change information, and the respiratory rate sequence of thermal comfort state, and combined with the clothing characteristic parameters, a nonlinear compensation calculation is performed to obtain a dynamic human metabolic rate sequence reflecting the real-time heat production capacity of the human body. Correction coefficients are determined based on the temperature change trend feature sequence used to characterize the trend of ambient temperature change and the clothing feature parameters, and the maximum fluctuation range of the skin temperature change information, the low-frequency to high-frequency ratio of the heart rate variability and the respiratory rate change information are determined. The comprehensive rate of change of core physiological parameters is determined based on the dynamic average skin temperature sequence, the correction factor, the maximum fluctuation range, the baseline value of basal metabolic rate, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, and the thermal comfort state respiratory rate sequence. The target skin temperature is determined based on the temperature change trend feature sequence, and weighting coefficients corresponding one-to-one with skin temperature, heart rate, and respiratory rate are determined. An individual dynamic thermal comfort evaluation model is constructed based on the clothing characteristic parameters, the weighting coefficients, the target skin temperature, the dynamic average skin temperature sequence, the heart rate variability sequence, the thermal comfort state heart rate variability sequence, the respiratory rate sequence, the thermal comfort state respiratory rate sequence, the maximum fluctuation range, and the comprehensive rate of change of the core physiological parameters.

5. The method according to claim 4, characterized in that, The step of calculating the real-time thermal comfort index of the target user using the individual dynamic thermal comfort evaluation model includes: According to a preset time cycle, the system cyclically acquires an instant physiological feature sequence to characterize the user's instant physiological changes within the current time cycle, and an instant environmental temperature feature sequence to characterize the user's environment's instant environmental temperature changes within the current time cycle. Acquire real-time images of passengers inside the vehicle; Perform multi-dimensional extraction of clothing features on the real-time in-vehicle passenger image information to obtain real-time clothing feature parameters that characterize the current passenger's heat exchange barrier performance; A timestamp synchronization operation is performed on the real-time physiological feature sequence and the real-time ambient temperature feature sequence, and based on the real-time physiological feature sequence and the real-time ambient temperature feature sequence after the timestamp synchronization operation, the real-time dynamic average skin temperature sequence, the real-time dynamic human metabolic rate sequence, and the comprehensive change rate of the real-time core physiological parameters of the target user are calculated. The real-time clothing characteristic parameters, the real-time dynamic average skin temperature sequence, the real-time dynamic human metabolic rate sequence, and the combined rate of change of the real-time core physiological parameters are input into the individual dynamic thermal comfort evaluation model to control the output of the real-time thermal comfort index of the individual dynamic thermal comfort evaluation model.

6. The method according to claim 5, characterized in that, The real-time physiological characteristic sequence includes a real-time heart rate variability sequence and a real-time respiratory rate sequence; the real-time ambient temperature characteristic sequence includes a temperature change trend characteristic sequence; the step of determining the target control parameters for the air conditioner based on the real-time thermal comfort index, and the real-time physiological characteristic sequence and the real-time ambient temperature characteristic sequence synchronized with timestamps includes: Based on the clothing characteristic parameters, the real-time thermal comfort index, the comprehensive rate of change of the instantaneous core physiological parameters, the instantaneous dynamic average skin temperature sequence, the instantaneous dynamic human metabolic rate sequence, the instantaneous heart rate variability sequence, and the instantaneous respiratory rate sequence, a multi-objective optimization function is constructed. Based on the temperature change trend characteristic sequence, determine the lower and upper limits of the target temperature inside the vehicle, the lower and upper limits of the target relative humidity inside the vehicle, and the lower and upper limits of the fresh air volume. The lower and upper limits of the target temperature inside the vehicle, the lower and upper limits of the target relative humidity inside the vehicle, and the lower and upper limits of the fresh air volume are determined as the constraints of the multi-objective optimization function. The multi-objective optimization function is solved based on the constraints to obtain the target supply air temperature, target relative humidity, target fresh air volume, supply air velocity, and supply air direction.

7. The method according to claim 6, characterized in that, The step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Obtain the current supply air temperature value and the compressor frequency conversion characteristic information of the air conditioner; the compressor frequency conversion characteristic information includes the compressor reference frequency; Calculate the difference between the target supply air temperature value and the current supply air temperature value; The target compressor frequency at the target time is calculated based on the difference, the preset temperature adjustment coefficient, and the compressor reference frequency. The compressor control command is generated using the target compressor frequency at the target time. Based on the compressor control commands, the compressor is driven to perform air supply temperature adjustment.

8. The method according to claim 6, characterized in that, The step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Get the current relative humidity value inside the vehicle; When it is determined that the target relative humidity value is lower than the current in-vehicle relative humidity value, the difference between the current in-vehicle relative humidity value and the target relative humidity value is calculated; Obtain information on air density, cooling space volume, and water specific heat capacity; The dehumidification power at the target time is calculated based on the difference, the air density information, the cooling space volume information, and the specific heat capacity information of water. The dehumidification power at the target time is used to generate control commands for the dehumidification device; The dehumidifier of the air conditioner is driven to adjust the humidity based on the control command of the dehumidifier.

9. The method according to claim 6, characterized in that, The step of generating a control command for the air conditioner using the target control parameters and controlling the air conditioner based on the control command includes: Obtain the minimum and maximum fresh air volume; Calculate the first difference between the target fresh air volume value and the minimum fresh air volume value; Calculate the second difference between the maximum fresh air volume and the minimum fresh air volume; The percentage of the fresh air valve opening is calculated based on the first difference and the second difference; The opening percentage of the fresh air valve is used to adjust the opening of the fresh air valve of the air conditioner.

10. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-7.

11. A vehicle, characterized in that, It includes the electronic device as described in claim 8.

Images