Space thermal microclimate multi-target regulation and control method, device and system, medium and product

Abstract

The invention relates to the crossing field of biothermodynamics and dynamic environment control, and discloses a space thermal microclimate multi-target regulation and control method, device and system, a medium and a product. The method comprises the following steps: acquiring physiological parameters (including effective dressing thermal resistance and local sweat gland density of posture correction) of a passenger, environment space-time parameters and light environment parameters through a biological characteristic sensor, an environment sensor and an optical sensor; constructing a dynamic human body thermal comfort model library, and adopting a feature coding-parameter decoding mechanism to match a target thermal comfort model; generating a hierarchical regulation strategy based on the difference between the local comfortable temperature and humidity interval and the real-time thermal environment sensing parameter; and the refrigerant circulation system, the fluid circulation system and the optical radiation system are cooperatively controlled to realize local microclimate regulation and control. According to the method, the problems of inaccurate thermal comfort modeling and low multi-system cooperation efficiency in a transient environment are solved, the energy efficiency ratio is improved, and the comfort of passengers in a cabin is remarkably improved.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of biothermodynamics and dynamic environmental control, specifically involving multi-objective regulation methods, devices, systems, media, and products for space thermal microclimate. Background Technology

[0002] Traditional environmental control technologies have the following limitations in transient non-uniform environments: (1) Insufficient model generalization ability (universality): Static models do not consider the fluctuations in human body thermal perception caused by changes in human posture, the different thermal perception habits of people in different regions, and other factors affecting actual applications. (2) Weak collaborative control: The cabin system and power system of a vehicle are usually two independent systems with independent control. Traditional environmental control technologies can hardly meet the requirements of high energy efficiency, fast system response, uniform temperature and humidity, and personalization at the same time. The internal space of a vehicle meets the characteristics of the aforementioned transient non-uniform environment. Therefore, current control strategies for the cabin system and power system are difficult to adjust the thermal microclimate in the vehicle cabin to the most comfortable state for passengers, resulting in some parts of the passenger's body feeling too hot and some parts feeling too cold, thus leading to insufficient passenger comfort in the cabin. Summary of the Invention

[0003] In view of this, the present invention provides a method, device, system, medium and product for multi-objective control of space thermal microclimate, in order to solve the problems of low energy efficiency and insufficient passenger comfort caused by the application of traditional environmental control technology to transient non-uniform environments of vehicles.

[0004] In a first aspect, the present invention provides a multi-objective control method for space thermal microclimate, the method comprising: acquiring human thermal comfort model parameters and thermal environment sensing parameters, wherein the human thermal comfort model parameters are parameters required for matching the human thermal comfort model, the human thermal comfort model parameters include occupant physiological parameters and vehicle spatiotemporal parameters, and the thermal environment sensing parameters represent the current thermal environment state of the vehicle; matching a target human thermal comfort model from a model library based on the human thermal comfort model parameters, the target human thermal comfort model representing the comfortable temperature and humidity of different parts of the human body; determining a vehicle thermal management strategy based on the target human thermal comfort model and the thermal environment sensing parameters; and controlling the cabin domain thermal management system and the power domain thermal management system in the vehicle based on the vehicle thermal management strategy.

[0005] In some optional implementations, the step of matching the target human thermal comfort model from the model library based on the human thermal comfort model parameters includes: separating the corresponding human thermal comfort model parameters for each occupant from the human thermal comfort model parameters; matching the sub-human thermal comfort model corresponding to each occupant from the model library based on the human thermal comfort model parameters corresponding to each occupant, wherein the sub-human thermal comfort model is used to represent the comfortable temperature and humidity of different parts of an occupant's body, and the various sub-human thermal comfort models constitute the target human thermal comfort model.

[0006] In some optional implementations, the step of obtaining human thermal comfort model parameters includes: collecting the location information and number information of occupants in the cabin; based on the location information and the number information, collecting the thermal resistance information of the clothing, the posture information of the occupants, the temperature and humidity of the steering wheel, and the contact pressure information of the occupants in the cabin, wherein the location information, the thermal resistance information of the clothing, the posture information of the occupants, the temperature and humidity of the steering wheel, and the contact pressure information of the occupants in the cabin are used as the physiological parameters of the occupants; collecting the vehicle coordinates and the weather information of the environment in which the vehicle is located as the spatiotemporal parameters of the vehicle; and collecting the light illuminance and the light color temperature as the light environment parameters.

[0007] In some optional implementations, collecting the location and number information of occupants in the cabin includes: when the in-cabin vision device is turned on, acquiring image data collected by the in-cabin vision device, and identifying the location and number information of the occupants through the image data; when the in-cabin vision device is not turned on, inferring the location and number information of the occupants through seat pressure signals collected by seat pressure sensors.

[0008] In some optional implementations, the acquisition of thermal resistance information of the occupants' clothing in the cabin includes: acquiring initial thermal resistance information of the occupants' clothing in the cabin; and correcting the initial thermal resistance information of the clothing by occupant posture parameters and sweat gland density parameters to obtain effective thermal resistance information of the clothing.

[0009] In some optional implementations, the acquisition of initial clothing thermal resistance information of occupants in the cabin includes: acquiring an image of occupant clothing when the in-cabin vision device is activated; identifying the type of clothing of the occupant through the clothing image; determining the initial clothing thermal resistance information based on a first mapping relationship between the clothing type and standard thermal resistance; activating a seat piezoelectric array and measuring changes in occupant muscle tension through seat micro-vibration when the in-cabin vision device is not activated; inferring clothing thickness based on the changes in muscle tension; determining the initial clothing thermal resistance information based on a second mapping relationship between clothing thickness and standard thermal resistance; and inferring the initial clothing thermal resistance information based on the vehicle coordinates and weather information of the vehicle's environment when the in-cabin vision device is not activated and no seat piezoelectric array is installed.

[0010] In some optional implementations, the step of correcting the initial clothing thermal resistance information using occupant posture parameters and sweat gland density parameters to obtain effective clothing thermal resistance information includes:

[0011] The effective clothing thermal resistance information c is calculated using the following formula. eff :

[0012] clo eff =clo initial ×δ posture ×(1-ω·ρ sweat ·η evap )

[0013] Among them, clo initial For the initial clothing thermal resistance information, δ posture ρ is the attitude influence factor, ω is the coefficient, and ρ is the attitude influence factor. sweat η is the dynamic weighting coefficient for sweat gland density. evap δ represents the sweat evaporation efficiency. posture and ρ sweat The calculation method is as follows:

[0014] δ posture =1+k1·θ torso +k2·a arm

[0015]

[0016] Where, θ torso a is the angle of the occupant's torso leaning forward. arm The upper arm lifting angle is represented by k1 and k2, which are calibration coefficients; A i This represents the area of ​​the i-th partition after dividing the occupant's body surface into multiple partitions (in cm²). 2 ), D i The density of sweat glands in the i-th region (unit: glands / cm²) 2 ).

[0017] In some optional embodiments, the method further includes: determining the air temperature inside the vehicle, the relative humidity inside the vehicle, and the inner surface temperature of the glass using the vehicle's spatiotemporal parameters and the light environment parameters; calculating the difference between the dew point temperature and the inner surface temperature of the glass based on the air temperature inside the vehicle, the relative humidity inside the vehicle, and the inner surface temperature of the glass; determining that glass fogging is about to occur when the difference between the dew point temperature and the inner surface temperature of the glass is less than a preset temperature threshold and continues for a preset duration; and activating the vehicle's defogging device when it is determined that glass fogging is about to occur.

[0018] In some optional implementations, the step of obtaining the thermal environment sensing parameters includes: obtaining initial thermal parameters and thermal management mode status, wherein the initial thermal parameters are used to represent the temperature status parameters of the current thermal environment of the vehicle, and the thermal management mode status is used to represent the activation status of various functions in the vehicle's thermal management system; wherein, the initial thermal parameters include a temperature cloud map inside the vehicle, external temperature data of the vehicle, and humidity inside the vehicle, and the temperature cloud map inside the vehicle is obtained through the following steps: when the humidity inside the vehicle is less than a preset humidity threshold, the temperature cloud map inside the vehicle is calculated by using the audio field played inside the vehicle; when the humidity inside the vehicle is greater than or equal to the preset humidity threshold, the temperature cloud map inside the vehicle is generated by using a dummy model coupled with CFD simulation.

[0019] In some optional implementations, a sub-human thermal comfort model corresponding to each occupant is matched from the model library based on the human thermal comfort model parameters corresponding to each occupant. This includes: calling the regional thermal comfort model library based on the vehicle's spatiotemporal parameters and the light environment parameters to select a first human model family; performing a second selection on the first human model family based on the occupant's physiological parameters to determine a second human model family; one or more sub-human thermal comfort models in the second human model family constitute the target human thermal comfort model; and each sub-human thermal comfort model corresponds to the thermal comfort of an occupant.

[0020] In some optional implementations, the step of calling the regional thermal comfort model library based on the vehicle spatiotemporal parameters and the light environment parameters to select the first human body model family includes: extracting feature vectors from the vehicle spatiotemporal parameters and the light environment parameters using a large language model; obtaining a regional thermal comfort model decoder, wherein the regional thermal comfort model decoder is a mapping program pre-trained using vehicle spatiotemporal parameter samples, light environment parameter samples, and the regional thermal comfort model library; and inputting the feature vectors into the regional thermal comfort model decoder for calculation to obtain the first human body model family.

[0021] In some optional implementations, a vehicle thermal management strategy is determined based on the target human thermal comfort model and the thermal environment sensing parameters, including: when the occupant is not in the vehicle, a first vehicle thermal management strategy is determined based on the difference between the target thermal parameters and the thermal environment sensing parameters, wherein the target thermal parameters represent the default temperature state parameters of the vehicle; when the occupant is in the vehicle, multiple sub-human thermal comfort models in the second human model family are sorted, and then a second vehicle thermal management strategy is determined according to the order of the sub-human thermal comfort models, so as to regulate the temperature of different body parts of different occupants in the vehicle in a specified order.

[0022] In some optional implementations, the mapping relationship between the human thermal comfort model parameters of any occupant and the sub-human thermal comfort model is obtained through the following steps: calculating the total heat transfer scalar between the human body and the environment, the photothermal scalar of the human body's perception of the light environment, and the dynamic human metabolic heat production scalar using the current human thermal comfort model parameters of the current occupant; allocating the total heat transfer scalar, the photothermal scalar, and the dynamic human metabolic heat production scalar to each body part using physiological weighting coefficients; determining the corresponding comfortable temperature and humidity range based on the heat corresponding to each body part; and generating the sub-human thermal comfort model under the current human thermal comfort model parameter conditions based on the comfortable temperature and humidity range corresponding to the body part.

[0023] In some alternative implementations, the total heat transfer scalar between the human body and the environment is calculated using the following formula:

[0024]

[0025] Among them, Q total and Q total′ Q is the total heat transfer scalar. res and Q res′ For heat exchange during human respiration, E s and E s′ For evaporative heat exchange through human skin, Q convection and Q convection′ For other convective heat transfer between the human body and the environment, Q radiation and Q radiation′ For other radiative heat exchange between the human body and the environment, γ is the metabolic rate correction factor;

[0026] E s =M b ·η age +(0.8·ρ sweat ·RH 0.3 )

[0027] M b η is the basal metabolic rate of the crew. age ρ is the age correction factor. sweat RH represents the dynamic weighting coefficient for sweat gland density.

[0028] In some alternative implementations, the photothermal scalar quantity of human perception of the light environment is calculated using the following formula:

[0029] K = K rad (E,CCT)+β·K psych (E,CCT)

[0030] K represents the photothermal scalar quantity, E represents the illuminance, CCT represents the color temperature, β is the calibration coefficient, and K rad(E,CCT) represents the photothermal effect caused by infrared radiation from the light source, K. psych (E,CCT) represents psychological thermal sensation, used to quantify the weighted integral of psychological thermal sensation in the human eye's sensitive wavelength band V(λ) under specific color temperature illumination.

[0031]

[0032] Where E is the illuminance (in lx), S(λ) is the spectral power distribution of the light source (in W / nm), V(λ) is the spectral luminous efficiency function (dimensionless), λ represents the wavelength of light, and Ψ(CCT) is the color temperature-thermal conversion factor (dimensionless). If the occupants manually adjust the temperature, Ψ(CCT) is updated through deep learning.

[0033] K psych The derivation of the formula for (E,CCT) is as follows:

[0034] (1) Obtain the relative spectral distribution through spectral normalization.

[0035]

[0036] (2) Quantify the combined effect Γ(CCT) of spectral shape and color temperature by integrating the color temperature spectrum:

[0037]

[0038] (3) The psychological thermal sensation K was obtained by using illuminance weighting. psych (E,CCT):

[0039]

[0040] In some alternative implementations, the dynamic human metabolic heat production scalar is calculated using the following formula:

[0041] M dyn =M act +M env +M phy

[0042] Among them, M dyn M is the dynamic human metabolic heat production scalar value. act For real-time activity intensity metabolic heat production, M env M phy This represents the physiological baseline for metabolic heat production.

[0043] In some optional embodiments, the method further includes: acquiring the current ambient temperature and predicting the predicted ambient temperature when the vehicle travels to a preset position; determining a compensation temperature based on the current ambient temperature and the predicted ambient temperature; and adjusting the vehicle thermal management strategy based on the compensation temperature when the vehicle travels to the preset position.

[0044] In some optional implementations, the method further includes: when the human thermal comfort model parameters change beyond a preset event threshold, returning the step of matching the target human thermal comfort model from the model library based on the changed human thermal comfort model parameters.

[0045] Secondly, the present invention provides a multi-objective control device for space thermal microclimate, the device comprising: a parameter acquisition module for acquiring human thermal comfort model parameters and thermal environment sensing parameters, wherein the human thermal comfort model parameters are parameters required for matching the human thermal comfort model, and the human thermal comfort model parameters include at least one of occupant physiological parameters, vehicle spatiotemporal parameters, and light environment parameters, and the thermal environment sensing parameters represent the current thermal environment state of the vehicle; a model matching module for matching a target human thermal comfort model from a model library based on the human thermal comfort model parameters, wherein the target human thermal comfort model represents the comfortable temperature and humidity of different parts of the human body; a strategy determination module for determining a vehicle thermal management strategy based on the target human thermal comfort model and the thermal environment sensing parameters; and a system control module for controlling the cabin domain thermal management system and the power domain thermal management system in the vehicle based on the vehicle thermal management strategy.

[0046] Thirdly, the present invention provides a space thermal microclimate multi-objective control system, comprising: a memory, a thermal management control module, a cabin domain thermal management system, and a power domain thermal management system. The memory and the thermal management control module are communicatively connected to each other. The memory stores computer instructions. The thermal management control module executes the computer instructions to perform the method described in any one of the first aspects, thereby controlling the cabin domain thermal management system and the power domain thermal management system.

[0047] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the first aspect or any corresponding embodiment thereof.

[0048] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to perform the method described in the first aspect or any corresponding embodiment thereof.

[0049] The technical solution provided by this invention has the following advantages:

[0050] (1) Based on the above technical means, by integrating human thermal model parameters (occupant physiological parameters and vehicle spatiotemporal parameters) with thermal environment perception parameters, matching the target thermal comfort model and controlling related systems, and combining coordinated cabin and power domain thermal management, the problems of low energy efficiency, slow response, uneven temperature, and failure to meet the personalized needs of different passengers in traditional thermal management models are solved. Moreover, the target thermal comfort model is created based on the local comfort temperature differences of the human body, and coordinates the thermal management of the cabin and power domains, which not only improves the dynamic comfort of occupants, but also reduces overall energy consumption through collaborative control, overcoming the problems of over-adjustment or under-adjustment caused by single temperature control.

[0051] (2) Based on the above technical means, in the scenario where the vehicle can carry multiple occupants, physiological parameters that conform to their individual characteristics are collected for each occupant. Then, the total human thermal comfort model parameters are allocated to each occupant in combination with the vehicle's spatiotemporal parameters. The human thermal comfort model is matched using the human thermal comfort model parameters corresponding to each occupant, and multiple sub-human thermal comfort models can be matched. The subsequent vehicle thermal management strategy adjusts the temperature and humidity according to the sub-human thermal comfort model of each occupant at different times, in different areas, and in different body parts, which further improves the comfort of different occupants riding in the vehicle.

[0052] (3) Based on the above technical means, the collection content of occupant physiological, vehicle spatiotemporal and optical environment parameters has been refined. Compared with the traditional scheme that only relies on ambient temperature, details such as clothing thermal resistance, posture and steering wheel interaction have been added. These parameters can more comprehensively reflect the individual differences of occupants and environmental characteristics, provide accurate input for subsequent model matching, avoid the control deviation caused by missing parameters, and improve the pertinence and accuracy of thermal management strategies.

[0053] (4) Based on the above technical means, the limitations of a single device are solved by complementary acquisition of visual equipment and seat pressure sensors: the visual equipment can accurately identify when it is turned on, and infer from the pressure signal when it is not turned on. This ensures that the distribution of occupants can be accurately obtained in any scenario, providing a basis for zoned temperature control, avoiding ineffective adjustment of unoccupied areas, reducing energy consumption, and improving the control accuracy in multi-occupant scenarios.

[0054] (5) Based on the above technical means, the initial thermal resistance of clothing is first obtained, and then corrected by posture and sweat gland density parameters, which breaks through the limitations of traditional static thermal resistance estimation. Posture changes (such as leaning forward) and sweating status will dynamically affect the actual thermal resistance. The corrected data is more in line with the real heat exchange needs, making the human thermal comfort model more accurate and avoiding excessive cold or heat caused by clothing thermal resistance deviation.

[0055] (6) Based on the above technical means, full-scene coverage is achieved through progressive acquisition of three modes: visual recognition, seat piezoelectric array inference, and spatiotemporal parameter inference. When the visual device is available, the clothing type is directly identified; when it is unavailable, the thickness is inferred through muscle tension; and when neither is available, environmental inference is combined to ensure that an effective initial thermal resistance can be obtained under any hardware configuration, thereby improving the system's adaptability and data reliability.

[0056] (7) Based on the above technical means, a mathematical model is established to correct the initial thermal resistance by quantifying the influence factors of posture (forward tilt of the torso, upper arm angle) and the weighting coefficient of sweat gland density. This model is more scientific than empirical estimation. It accurately reflects the influence of posture changes on thermal resistance and the regulatory effect of sweat evaporation, making the thermal resistance data highly consistent with the actual heat dissipation needs of the human body, and providing accurate parameter support for subsequent temperature control strategies.

[0057] (8) Based on the above technical means, by calculating the difference between the dew point temperature of the windshield and the temperature of the inner surface of the glass, the risk of fogging can be determined in advance and the defogging equipment can be activated, overcoming the lag of traditional post-fogging treatment. This mechanism can provide a warning 10 seconds before fogging, and combined with stepped power control, it can ensure the safety of vehicle driving, avoid energy waste and temperature fluctuations caused by excessive defogging, and improve driving safety and comfort.

[0058] (9) Based on the above technical means, the embodiments of the present invention acquire initial thermal parameters and thermal management mode states to accurately represent the current thermal environment state of the vehicle, thereby enabling the vehicle to know from which state position to begin adjustment, providing a basis for the generation of thermal management strategies. Specifically, the complementary approach of audio field calculation (low humidity) coupled with CFD simulation of a dummy model (high humidity) overcomes the limitation of traditional sensors that can only acquire single-point temperatures. In low humidity, the temperature variation characteristics of the audio signal are used to generate accurate cloud maps; in high humidity, CFD ensures accuracy, providing global temperature data for thermal management strategies and improving the spatial accuracy of control.

[0059] (10) Based on the above technical means, the two-step method of "initial screening of regional thermal comfort model library and secondary screening of occupant physiological parameters" breaks through the universality defects of traditional general models. First, model families adapted to regional characteristics are screened according to region and light environment, and then refined by combining individual occupant physiological parameters, so that the target model is more in line with the specific environment and occupant needs, and the personalization and accuracy of thermal management are improved.

[0060] (11) Based on the above technical means, the regional thermal comfort model library is replaced with the regional thermal comfort model decoder. On the one hand, the occupation of vehicle hardware storage space is reduced. On the other hand, for regional thermal comfort models that are not configured in certain special scenarios, the corresponding regional thermal comfort models can be calculated and generated based on the regional thermal comfort model decoder, thereby improving the generalization ability of the regional thermal comfort model matching process.

[0061] (12) Based on the above technical means, this strategy distinguishes between the state of the occupants not boarding the vehicle and the state of boarding the vehicle, and realizes a refined transition of thermal management. When not boarding the vehicle, based on the difference between the target thermal parameters and the current environment, the cabin is pre-adjusted to approach a comfortable state in advance, avoiding excessive waiting time for the occupants after boarding the vehicle; after boarding the vehicle, the thermal comfort models of the sub-human bodies in the second human body model family are sorted, and the body parts of different occupants are adjusted step by step according to the interpolation logic of the key frame model. This design not only reduces transient energy consumption through pre-adjustment, but also avoids discomfort caused by sudden temperature changes by using the intermediate transition mechanism of the key frame model. At the same time, it ensures that the local needs of each occupant are accurately met according to priority in multi-occupant scenarios, which significantly improves dynamic comfort and system energy efficiency.

[0062] (13) Based on the above technical means, by calculating the total heat exchange, photothermal and dynamic metabolic scalars and allocating them to local parts of the body, corresponding comfortable temperature and humidity ranges are generated, which solves the problem that traditional models ignore local differences in the human body. This mechanism can accurately capture the different comfort needs of the head, torso, limbs and other parts, provide a basis for zoned temperature control, and realize local comfort regulation that is "personalized for each person".

[0063] (14) Based on the above technical means, a metabolic rate correction coefficient is introduced into the heat transfer equation to optimize heat transfer calculations for specific groups such as the elderly and children, overcoming the bias caused by the universality of traditional models. The formula also integrates details such as respiratory heat transfer and skin evaporation, and combines age and sweat gland density corrections to make the total heat transfer calculation more in line with the physiological characteristics of different passengers, improving the comfort of specific groups. In addition, the design of human skin evaporative heat transfer has been simplified to avoid data that is difficult to collect accurately in real time in dynamic vehicle scenarios and has a complicated calculation process. It can be calculated using only basal metabolic rate, age correction coefficient, dynamic weighting coefficient of sweat gland density, and relative humidity. The parameters are all readily available conventional data. The simplification retains the core influencing factors and significantly reduces the difficulty of data collection and computational complexity, ensuring that the total heat transfer scalar is updated in real time, improving the dynamic adaptability and response speed of thermal management strategies.

[0064] (15) Based on the above-mentioned technical means, the present invention provides a new method for calculating photothermal scalars, thereby integrating infrared radiation heat and psychological thermal sensation, and combining deep learning to update the color temperature-thermal sensation conversion factor, breaking through the limitation of traditional methods that ignore the subjective influence of the light environment. This mechanism can accurately quantify the psychological effect of color temperature on thermal sensation, and by manually adjusting the data to optimize the model, the photothermal compensation is more in line with the subjective feelings of passengers, improving the comfort experience.

[0065] (16) Based on the above technical means, metabolic heat production is calculated by comprehensively considering real-time activity intensity, environmental correction, and physiological baseline values, thus solving the lag problem of traditional static metabolic estimation. This formula can reflect the metabolic fluctuations caused by occupants' activities (such as adjusting their sitting posture) and environmental changes in real time, enabling the human thermal comfort model to dynamically adapt to the human body's state and improve the control accuracy in transient scenarios.

[0066] (17) Based on the above technical means, by predicting the ambient temperature of the vehicle when it travels to the preset position, the compensation temperature is determined only based on the difference between the current temperature and the predicted temperature to fine-tune the strategy. There is no need to frequently rematch the human thermal comfort model, which significantly optimizes the system response efficiency.

[0067] (18) Based on the above technical means, when the parameters of the human thermal comfort model change abruptly (such as changes in the number of occupants or posture), the model is automatically rematched, solving the problem of traditional control lag. This mechanism ensures that the human thermal comfort model is always synchronized with the current state, avoids control deviations caused by parameter changes, and improves the system's adaptability and response speed to dynamic scenarios. Attached Figure Description

[0068] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0069] Figure 1 This is a flowchart illustrating a multi-objective control method for space thermal microclimate according to an embodiment of the present invention.

[0070] Figure 2 This is another schematic flowchart of a multi-objective control method for space thermal microclimate according to an embodiment of the present invention;

[0071] Figure 3 This is another schematic flowchart of a multi-objective control method for space thermal microclimate according to an embodiment of the present invention;

[0072] Figure 4 This is a schematic diagram of the structure of a space thermal microclimate multi-objective control device according to an embodiment of the present invention;

[0073] Figure 5 This is a schematic diagram of the structure of a space thermal microclimate multi-objective control system according to an embodiment of the present invention;

[0074] Figure 6 This is a schematic diagram of a cabin thermal management system according to an embodiment of the present invention. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] The following description uses one of the most common vehicles, the new energy vehicle, as a specific example.

[0077] With the increasing number of electric vehicles on the road, "range anxiety" has become one of the main considerations for car buyers, especially in the winter in northern China, where driving range can drop by more than 30%. The energy consumption of vehicle thermal management systems has become one of the key factors affecting driving range. To reduce energy consumption, the industry generally adopts technologies such as intelligent zone temperature control and waste heat recovery. However, these technologies often overlook another major factor that users consider when buying a car—"comfort".

[0078] There are currently many theories on human thermal comfort, and two common evaluation standards are: (1) PMV (Predicted Mean Vote), which is a quantitative indicator of human thermal sensation defined by the International Organization for Standardization. It is one of the core parameters for evaluating thermal comfort, and its range is usually between -3 (cold) and +3 (hot). 0 indicates thermal neutrality (most comfortable). (2) PPD (Predicted Percentage Dissatisfied), which is an indicator related to PMV and represents the estimated percentage of people who are dissatisfied with the thermal environment. Even if PMV = 0 (best thermal neutrality), there are usually still about 5% of people who are dissatisfied (PPD ≈ 5%). The lower the PPD value, the higher the overall comfort.

[0079] Traditional methods for controlling the thermal management system within a vehicle based on human thermal comfort models still have the following major drawbacks.

[0080] The vehicle thermal management system in the aforementioned comparative application has the following defects:

[0081] (1) Isolated parameter control: Only environmental parameters are considered, without considering the physiological state parameters of the occupants, resulting in inaccurate zone control (over- or under-cooling / heating).

[0082] (2) Dynamic response lag: The physiological state of the occupants is dynamic. The response calculated using only historical data is somewhat different from the actual needs. At the same time, it is impossible to respond in real time to sudden situations (such as sudden changes in sunlight radiation or sudden changes in the physiological parameters of the occupants).

[0083] (3) Most of them adopt the standard PMV model. The standard PMV model is for uniform, steady-state (quasi-equilibrium) working conditions, such as a room. It is not adaptable to the actual working conditions of the vehicle (non-uniform cabin, transient environment) and does not take into account the short-term comfort of different sub-stages within the working conditions. Therefore, the actual energy efficiency optimization effect is limited and cannot achieve dynamic adaptation to the in-vehicle environment and human thermal perception.

[0084] (4) Poor coordination and control. The cabin system and the power system are usually two independent systems with independent control, making it difficult to simultaneously meet high energy efficiency, fast system response, uniform temperature and humidity, and personalized requirements.

[0085] According to an embodiment of the present invention, a multi-objective regulation method for space thermal microclimate is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0086] This embodiment provides a multi-objective method for regulating space thermal microclimate. Figure 1 This is a flowchart of a multi-objective control method for space thermal microclimate according to an embodiment of the present invention. The flowchart includes the following steps:

[0087] Step S101: Obtain human thermal comfort model parameters and thermal environment perception parameters. The human thermal comfort model parameters are the parameters required to match the human thermal comfort model. The human thermal comfort model parameters include at least the occupant's physiological parameters and the vehicle's spatiotemporal parameters. The thermal environment perception parameters represent the current thermal environment state of the vehicle.

[0088] Step S102: Match the target human thermal comfort model from the model library according to the parameters of the human thermal comfort model. The target human thermal comfort model is used to represent the comfortable temperature and humidity of different parts of the human body.

[0089] Step S103: Determine the vehicle thermal management strategy based on the target human thermal comfort model and thermal environment perception parameters;

[0090] Step S104: Control the cabin thermal management system and the power thermal management system in the vehicle based on the vehicle thermal management strategy.

[0091] Specifically, this embodiment of the invention first obtains human thermal comfort model parameters and thermal environment perception parameters. The human thermal comfort model parameters refer to the parameters needed to match the human thermal comfort model in the model library. These parameters include at least occupant physiological parameters and vehicle spatiotemporal parameters, and optionally, light environment parameters. The thermal environment perception parameters are used to represent the current thermal environment state of the vehicle, such as the current vehicle temperature, the current vehicle air conditioning setting, etc.

[0092] There are several ways to collect parameters of the human thermal comfort model. For example, images can be captured by a wide-angle camera in the cabin, and image recognition technology can be used to determine that there are two occupants in the cabin, and that they are located in the driver's seat and the front passenger seat, respectively. Alternatively, seat pressure sensors can be used to verify the occupants' position information, or wearable devices can be used to collect data such as the occupants' heart rate and body temperature. This embodiment of the invention is only an example and is not limited thereto. Regarding the vehicle's spatiotemporal parameters, the current vehicle coordinates can be obtained through a GPS module, and local weather, ambient temperature, relative humidity, sunlight intensity, etc., can be obtained simultaneously through the network. This embodiment of the invention is only an example and is not limited thereto. Light environment parameters include illuminance and color temperature. The illuminance inside the vehicle can be measured by a rain sensor, and the color temperature can be obtained by converting images captured by the in-vehicle camera. This embodiment of the invention is only an example and is not limited thereto.

[0093] There are multiple ways to acquire thermal environment sensing parameters. For example, a combination of distributed NTC (Negative Temperature Coefficient) sensors and an ultrasonic temperature measurement network can be used to generate an in-vehicle temperature cloud map. This map could display the driver's seat temperature as 28°C, the front passenger seat as 29°C, the rear seats as 27°C, the outside temperature as 25°C, and the in-vehicle humidity as 45%. It can also confirm whether the current cabin thermal management system and powertrain thermal management system are in a switched-off, standby, or active state.

[0094] After completing the above steps to obtain the parameters, as follows: Figure 2 and Figure 3As shown, in this embodiment of the invention, the target human thermal comfort model is matched from the model library based on the aforementioned human thermal comfort model parameters. In this embodiment, the model library refers to a structured database stored in the vehicle thermal management system's memory or a cloud platform. It is used to centrally manage various human thermal comfort models and related parameter mapping relationships. The model library stores a large number of human thermal comfort models, and its core function is to provide callable human thermal comfort model resources for different scenarios (such as region, climate, and occupant characteristics). It supports the system in quickly matching and adapting models based on real-time collected human thermal comfort model parameters (occupant physiology, vehicle spatiotemporal factors, lighting environment, etc.). The human thermal comfort models stored in the model library are parameter models constructed based on the principles of human thermal balance, physiological characteristics, and environmental interaction laws. They are used to quantify the comfortable temperature and humidity ranges for different parts of the human body under different scenarios.

[0095] Unlike the traditional single standard PMV model, the human thermal comfort model of this invention has three major characteristics: "segmentation", "localization", and "dynamics".

[0096] The segmentation can be tailored to different regions (such as dry areas in the north and humid areas in the south), climate conditions (such as high temperature and humidity, low temperature and strong radiation), and occupant characteristics (such as age, clothing, and posture). For example, the northern winter model will emphasize the insulation of the torso and hands, while the southern summer model will focus on the heat dissipation needs of the head and back.

[0097] Localization allows each model to include independent comfort temperature and humidity parameters for different parts of the human body (head, torso, limbs, hands, etc.), rather than overall average temperature and humidity. For example, in the model for the driver, the comfortable temperature for the hands (touching the steering wheel) is 25-27℃, and the comfortable temperature for the head is 23-25℃, reflecting the differences in local thermal perception.

[0098] The dynamism is reflected in the model's dynamic matching, which allows the temperature comfort range to be dynamically adjusted according to environmental variables (such as light intensity and humidity) and physiological parameters (such as sweat gland activity and posture changes). For example, when an increase in occupant sweating is detected, the model will automatically lower the torso comfort temperature and humidity threshold to adapt to heat dissipation needs.

[0099] After matching the target human thermal comfort model, this embodiment of the invention determines the vehicle thermal management strategy based on the target human thermal comfort model and thermal environment perception parameters. For example, by comparing the comfortable temperature and humidity in the target thermal comfort level with the current in-vehicle temperature cloud map data, it is found that the current temperature of the driver's seat and the passenger seat is higher than the comfortable temperature and humidity of their respective torsos, so it is determined that the cooling mode needs to be activated. Considering the priority of driving safety, the strategy prioritizes the control of the driver's area, and then controls the passenger's area. In addition, the control valves, adjustment levels, etc. are selected according to the activation status of the cabin and power domain thermal management systems. For example, a collaborative control scheme is formulated based on the human thermal comfort model, where the heating system activates the cooling system of the cabin domain, the airflow of the driver's seat vent is adjusted to level 5 and the temperature is set to 23°C, the airflow of the passenger seat vent is adjusted to level 4 and the temperature is set to 24°C, and the seat ventilation system is activated to assist in heat dissipation; the battery thermal management system of the power domain switches to waste heat recovery mode, which introduces the waste heat generated during battery cooling into the cabin to reduce the load on the cabin domain thermal management system.

[0100] Finally, the cabin thermal management system and the powertrain thermal management system in the vehicle are controlled based on the above thermal management strategies.

[0101] The technical solution provided by this invention comprehensively captures the physiological characteristics of occupants and the internal and external environmental conditions of the vehicle through the collection of multi-dimensional parameters. This breaks through the limitations of traditional thermal management that relies on only a single environmental parameter, ensuring that the target model is highly adapted to the actual scenario and solving the problem of insufficient adaptability of the standard model to the non-uniform and transient environment of the vehicle. The strategy formulation is based on local comfort temperature and humidity and priority division to achieve precise control. Through precise matching of local comfort temperature and humidity of the human body, the PMV deviation is controlled within ±0.3, avoiding the energy waste and local discomfort of "one-size-fits-all" adjustment, and greatly improving the comfort of the occupants. The system control realizes the utilization of waste heat through the coordination of the cabin domain and the power domain, thereby improving the overall energy efficiency.

[0102] In some optional implementations, step S102 above includes:

[0103] Step I1: Separate the corresponding human thermal comfort model parameters for each occupant from the human thermal comfort model parameters;

[0104] Step I2: Match the sub-human thermal comfort model corresponding to each passenger from the model library according to the human thermal comfort model parameters corresponding to each passenger. The sub-human thermal comfort model is used to represent the comfortable temperature and humidity of different parts of a passenger's body. The sub-human thermal comfort models together form the target human thermal comfort model.

[0105] Specifically, based on the aforementioned technical means, and considering scenarios where a vehicle can carry multiple occupants, physiological parameters tailored to each occupant's individual characteristics are collected individually. Then, combined with the vehicle's spatiotemporal parameters, the overall human thermal comfort model parameters are allocated to each occupant. By using the human thermal comfort model parameters corresponding to each occupant, multiple sub-human thermal comfort models can be obtained. Subsequently, the vehicle's thermal management strategy adjusts temperature and humidity based on each occupant's sub-human thermal comfort model, adjusting it by time, zone, and body area, further improving the comfort of different occupants riding in the vehicle.

[0106] In some alternative implementations, the step of obtaining human thermal comfort model parameters includes:

[0107] Step a1: Collect the location and number information of the occupants in the cabin. Then, based on the location and number information, collect the thermal resistance information of the clothing, the occupant posture information, the temperature and humidity of the steering wheel, and the contact pressure information of the occupants in the cabin. Among them, the location information, thermal resistance information of the clothing, the occupant posture information, the temperature and humidity of the steering wheel, and the contact pressure information of the occupants in the cabin are used as occupant physiological parameters.

[0108] Step a2: Collect the vehicle coordinates and weather information of the vehicle's environment as the vehicle's spatiotemporal parameters;

[0109] Step a3: Collect light illuminance and light color temperature as light environment parameters.

[0110] Specifically, the physiological parameters of the occupants collected in step a1 of this embodiment of the invention include the occupant's position and number information, clothing thermal resistance information, occupant posture information, steering wheel temperature and humidity, and contact pressure information.

[0111] The occupant location and number information can be obtained through the occupant location-number-clothing perception module, and can be identified using any one of the following methods or a combination thereof: (1) Real-time video streams are collected by multiple wide-angle cameras in the cockpit (e.g., on the top of the cockpit), and then the head coordinates of the occupants are identified by an algorithm (e.g., YOLOv5 algorithm) to determine the number and location of the occupants. (2) The occupant location and number are identified by using a multispectral imaging unit (including near-infrared 850nm / thermal imaging 14μm dual bands) installed in the cockpit, combined with infrared images and target recognition algorithms. (3) When the cameras or multispectral imaging units of the occupant location-number-clothing perception module are not authorized to be turned on, the approximate location and number information of the occupants can be obtained by the seat pressure sensor of the seat sensing module. (4) The approximate location and number information of the occupants is scanned by the millimeter-wave radar sensor in the cabin.

[0112] Among them, clothing thermal resistance information can be collected through optical mode, physical sensing mode, spatiotemporal mode or a combination of the above modes. For example, infrared thermal imaging and pressure sensors can be used together to determine the thermal resistance of a user's clothing.

[0113] In an optional implementation, the technical means of collecting the initial thermal resistance of occupants in this embodiment of the invention can be achieved through the following steps:

[0114] Step a11: When the in-cabin vision equipment is turned on, acquire an image of the occupants' clothing;

[0115] Step a12: Identify the type of clothing worn by the occupants through images of their attire;

[0116] Step a13: Determine the initial clothing thermal resistance information based on the first mapping relationship between clothing type and standard thermal resistance;

[0117] Step a14: When the visual equipment in the cockpit is not turned on, activate the seat piezoelectric array and measure the occupant's muscle tone changes through seat micro-vibration.

[0118] Step a15: Infer clothing thickness based on muscle tone change information;

[0119] Step a16: Determine the initial clothing thermal resistance information through the second mapping relationship between clothing thickness and standard thermal resistance;

[0120] Step a17: When the visual equipment in the cockpit is not turned on and the seat piezoelectric array is not installed, infer the initial thermal resistance information of the clothing based on the vehicle coordinates and the weather information of the vehicle's environment.

[0121] Specifically, the thermal resistance of occupant clothing is acquired through optical methods. Images of occupant clothing are captured using a camera or multispectral imaging unit (including a dual-band near-infrared 850nm / thermal imaging 14μm band). Then, a clothing segmentation algorithm is used to identify the multi-layered clothing type of each occupant in the image. Specifically, the clothing type can be identified based on a ResNet-18 clothing classification model (the training dataset contains 200 common clothing categories). Then, based on the first mapping relationship between clothing type and standard thermal resistance, the initial clothing thermal resistance information is determined. For example, the first mapping relationship can be mapped to thermal resistance values ​​using the ISO7730 standard; for example, a short-sleeved T-shirt = 0.08clo, a down jacket = 0.7clo. The initial clothing thermal resistance information at this point is clo. initial For visual thermal recognition resistance value clo vis .

[0122] When optical sensing is disabled, the thermal resistance of occupant clothing can be acquired through physical sensing. Specifically, by activating the micro-vibration sensing mode of the seat's piezoelectric array, the body's muscle tension changes differently due to varying clothing thickness and external force. This allows for the inference of clothing thickness based on the mapping relationship between the magnitude of muscle tension changes and clothing thickness, a relationship that can be pre-trained using a large-scale data training model. Subsequently, a second mapping relationship between clothing thickness and thermal resistance is established based on standards such as GB / T 11048, ISO 11092, and ASTM F1868, allowing the initial clothing thermal resistance information (clo) to be retrieved by querying this second mapping relationship. initial That is, physical identification of thermal resistance.

[0123] When neither optical sensing nor physical sensing modes are available, embodiments of the present invention may employ a spatiotemporal mode, which uses the vehicle's GPS location information and corresponding weather information obtained from the spatiotemporal parameter unit to infer the user's clothing thickness and then infers the initial clothing thermal resistance information based on the second mapping relationship.

[0124] This invention achieves full-scene coverage through progressive data acquisition using three modes: visual recognition, seat piezoelectric array inference, and spatiotemporal parameter estimation. When visual devices are available, the clothing type is directly identified; when unavailable, thickness is inferred through muscle tension; and when neither is available, environmental inference is combined to ensure that effective initial thermal resistance can be obtained under any hardware configuration, thereby improving system adaptability and data reliability.

[0125] In addition, in some optional embodiments, the present invention also corrects the deviation of the initial clothing thermal resistance information collected by the sensor device in dynamic vehicle scenarios. The correction incorporates multiple factors such as occupant posture parameters and sweat gland density parameters to finally obtain effective clothing thermal resistance information co. eff By correcting posture and sweat gland density parameters, the limitations of traditional static thermal resistance estimation are overcome. Posture changes (such as leaning forward) and sweating status dynamically affect the actual thermal resistance. The corrected data more closely matches the actual heat exchange requirements, making the human body thermal comfort model more accurate and avoiding excessive cold or heat caused by clothing thermal resistance deviations, thereby further improving the accuracy of user clothing thermal resistance.

[0126] Specifically, this is achieved through the following formula:

[0127] clo eff =clo initial ×δ posture ×(1-ω·ρ sweat ·η evap )

[0128] Where, δ posture ρ is the attitude influence factor, ω is the coefficient or coefficient function, and ρ is the attitude influence factor. sweatη is the dynamic weighting coefficient for sweat gland density. evap The sweat evaporation efficiency (obtained through joint calibration of wind speed and humidity, improving the model's calculation accuracy in high humidity environments), δ posture and ρ sweat The calculation method is as follows:

[0129] δ posture =1+k1·θ torso +k2·a arm

[0130]

[0131] In the formula, θ torso This refers to the torso lean angle, and the data comes from occupant posture information, which can be derived from seat motor data; a arm The upper arm elevation angle is obtained from steering wheel contact data; k1 and k2 are calibration coefficients; A i This represents the area of ​​the i-th partition after dividing the occupant's body surface into multiple partitions (in cm²). 2 This area is obtained by referring to tables in GB / T 10000-2023 "Anthropometric Dimensions of Chinese Adults", GB / T 2428-2024 "Anthropometric Dimensions of Adult Head and Face", GB / T 42746-2023 "Three-Dimensional Foot Model of Adults", GB / T16252-2023 "Hand Size Classification of Adults", GB / T 26158-2010 "Anthropometric Dimensions of Chinese Minors", GB / T 26159-2010 "Hand Size Classification of Chinese Minors", GB / T 26160-2010 "Anthropometric Dimensions of Chinese Minors", and GB / T 26161-2010 "Foot Size Classification of Chinese Minors". Therefore, the corresponding body surface area varies for occupants of different heights and weights. i The density of sweat glands in the i-th body surface region of the occupant (unit: glands / cm²) 2 In this embodiment of the invention, the distribution of sweat glands in different regions is derived from the "Table of Sweat Gland Distribution in Different Regions", and a schematic table of the "Table of Sweat Gland Distribution in Different Regions" is shown below.

[0132]

[0133] The specific values ​​for sweat gland density (*) in each body surface region in the table above can be found in the "Standard Dataset of Human Anatomy".

[0134] For example, when correcting the thermal resistance of the driver's clothing during vehicle operation, the cockpit vision device first identifies that the occupant is wearing a windproof jacket. Based on the first mapping relationship between clothing type and standard thermal resistance, the initial clothing thermal resistance is determined. Subsequently, corrections are made based on occupant posture parameters and sweat gland density parameters. The occupant's torso lean angle is obtained through the angle sensor of the seat adjustment motor, and the upper arm lift angle is inferred from the steering wheel pressure sensor distribution data. Assuming the calibration coefficients k1 and k2 are known, δ can be obtained by substituting them into the posture influence factor formula. posture The occupant's body surface was then divided into six regions: head, torso, upper arms, and hands. The area of ​​each region was determined using the "Anthropometric Dimensions of Chinese Adults," and the corresponding sweat gland density was obtained from the "Standard Anatomical Dataset for Human Anatomy." Finally, a dynamic weighting coefficient ρ for sweat gland density was calculated using the sweat gland density and the area of ​​each region. sweat Assuming the coefficient ω is known, the sweat evaporation efficiency η evap Substituting into the effective clothing thermal resistance formula c eff =clo initial ×δ posture ×(1-ω·ρ sweat ·η evap The effective clothing thermal resistance cLo can then be calculated. eff .

[0135] The method for correcting clothing thermal resistance provided in this invention quantifies the changes in clothing fit caused by forward leaning of the torso and raising of the arms (e.g., when leaning forward, the torso is in closer contact with the seat, reducing thermal resistance); the sweat gland density weighting coefficient reflects the differences in sweating capacity of different body parts, and combined with sweat evaporation efficiency, corrects the weakening effect of sweating on the actual heat insulation performance of clothing. The combined effect of these two factors transforms the clothing thermal resistance from a static initial value into a dynamic effective value that conforms to the real-time physiological state. Compared to the traditional method of calculating fixed clothing thermal resistance, effective clothing thermal resistance information can accurately reflect the impact of changes in occupant posture and sweating status on heat exchange, making the subsequent matching of the human thermal comfort model more closely match the actual heat dissipation needs of the human body, avoiding over- or under-adjustment of air conditioning due to thermal resistance estimation errors, thus improving occupant comfort and reducing energy consumption.

[0136] After obtaining the thermal resistance information of the occupant's clothing in the above steps, it is necessary to continue collecting the occupant's posture information. In this embodiment of the invention, the occupant's posture information includes the torso forward tilt angle, head and neck tilt angle, upper arm elevation angle, etc. This information can be obtained by reading the seat controller register via the CAN bus. For example, by reading the on / off status of the seat heater and ventilation motor, the direction, travel percentage, and stop / rotation information of the seat adjustment motor, information such as the backrest motor angle (0°~120°, accuracy ±1°) and leg rest extension (0-150mm) can be obtained; the seat cushion pressure distribution matrix (unit kPa) can be obtained through the parameter information of the 4×4 piezoelectric sensor array.

[0137] Furthermore, in this embodiment of the invention, steering wheel temperature and humidity and contact pressure information are collected. For example, the steering wheel temperature and humidity sensor data (optional NTC thermistor ±0.5℃ accuracy) is collected every 100ms. Contact pressure detection can be performed using a piezoresistive thin-film sensor (range 0-50N). When the pressure is continuously >5N and the temperature gradient is >2℃ / s, it is determined to be effective contact.

[0138] The spatiotemporal parameters of the vehicle collected in step a2 of this embodiment of the invention include the vehicle coordinates and the weather information of the environment in which the vehicle is located.

[0139] Specifically, in this embodiment of the invention, the vehicle is positioned using a primary positioning source of "dual-frequency GPS + BeiDou module (update frequency 10Hz)" and an auxiliary positioning source of "real-time dynamic differential positioning and inertial measurement unit". The data sources of GPS, real-time dynamic differential positioning and inertial measurement unit are weighted and fused using a fusion algorithm (such as extended Kalman filter EKF) (the weights of each source are preset calibration values). The fusion result is then used to obtain the vehicle's precise coordinates and dead reckoning.

[0140] Specifically, embodiments of the present invention acquire real-time meteorological data through a TSP (Telematics Service Provider) platform, including, for example, solar radiation intensity (unit: W / m²). 2 (Accuracy ±10%), current ambient temperature T current (Based on recent weather station data + calibration by vehicle exterior temperature sensor), relative humidity, and the duration of current sunshine intensity and ambient temperature. Additionally, it can be obtained via future GPS data. t At any given time, the vehicle speed v t Solar azimuth α sun and altitude H elev The predicted ambient temperature T at a certain location and time in the future is calculated. env (t), i.e., T env (t)=Φ(GPS t ,v t ,α sun Helev ).

[0141] The light environment parameters collected in step a3 of this embodiment include light illuminance and light color temperature.

[0142] The light intensity is obtained through a rain light sensor (RLS), which is usually installed on the inside of the windshield. The measurement range is 0-100klx, and it is scanned in 5 areas (sampling rate 1Hz). In particular, there is dynamic compensation at night, namely automatic gain at night ×10lx.

[0143] Specifically, the color temperature (CCT) can be calculated by acquiring an RGB image of the glass using an in-vehicle camera module (e.g., an IR-cut filter lens mount), and then calculating it based on the CIE XYZ color space conversion. The conversion formula is as follows:

[0144] CCT=449n 3 +3525n 2 +6823.3n+5520.33, where

[0145] in, X, Y, and Z are the tristimulus values ​​of the CIE XYZ color space, respectively.

[0146] The method for obtaining human thermal comfort model parameters provided in this invention comprehensively captures individual occupant characteristics based on multi-source sensor fusion (visual, pressure, temperature, and humidity). It obtains intuitive clothing and location information through visual recognition and corrects thermal resistance using pressure sensors and posture parameters to ensure the accuracy of physiological parameters. Step a2 combines high-precision positioning and meteorological data to place the vehicle in a specific spatiotemporal environment, providing regional and climatic background for subsequent model matching. Step a3 quantifies the physical characteristics of the light environment (illuminance, color temperature), laying the foundation for analyzing the impact of light and heat on human comfort. Through the synergistic effect of these three elements, multi-dimensional and high-precision input parameters are provided for matching the human thermal comfort model, overcoming the shortcomings of traditional methods that rely on single parameters and ignore the interaction between the individual and the environment. By collecting detailed parameters, the vehicle's thermal management system can accurately distinguish the thermal needs of different occupants (such as different heat dissipation capabilities due to clothing differences). Combined with spatiotemporal and light environment parameters, it provides rich scene information for subsequent model matching, ensuring that the matched human thermal comfort model is more in line with actual conditions. This improves the personalization and accuracy of vehicle thermal management, reduces energy waste, and enhances the comfort experience of occupants.

[0147] In some alternative implementations, the step of acquiring thermal environment sensing parameters includes:

[0148] Step b1: Obtain initial thermal parameters and thermal management mode status. Initial thermal parameters are used to represent the temperature status parameters of the current thermal environment of the vehicle, and thermal management mode status is used to represent the activation status of various functions in the vehicle's thermal management system.

[0149] The initial thermal parameters include the vehicle's internal temperature cloud map, external temperature data, and internal humidity. The internal temperature cloud map is obtained through the following steps:

[0150] Step b11: When the humidity inside the vehicle is less than the preset humidity threshold, calculate the temperature cloud map inside the vehicle by using the audio field played inside the vehicle.

[0151] Step b12: When the humidity inside the vehicle is greater than or equal to the preset humidity threshold, a temperature cloud map inside the vehicle is generated by coupling a dummy model with CFD simulation.

[0152] Specifically, the thermal environment sensing parameters obtained in the embodiments of the present invention include initial thermal parameters and thermal management mode status.

[0153] Taking new energy vehicles as an example, the initial thermal parameters include the in-vehicle temperature cloud map, the outside temperature data, and the in-vehicle humidity. Specifically, the in-vehicle temperature cloud map can be acquired by several (e.g., eight) NTC sensors, such as those positioned on the dashboard, headliner, seats, and footwell. Then, an algorithm (e.g., the Kriging spatial interpolation algorithm) can be used to interpolate the temperature values ​​collected by the temperature sensors into a three-dimensional in-vehicle temperature cloud map.

[0154] However, the accuracy of the three-dimensional temperature cloud map generated by the interpolation algorithm is limited. In an optional implementation, this embodiment of the invention further selects different schemes to measure the temperature inside the vehicle based on humidity.

[0155] When the in-vehicle humidity is below a preset humidity threshold, the speaker components of the door audio modules, center console audio modules, headrest audio modules, and A / B / C pillar audio modules can be activated to send audio signals (ultrasonic transducers). These signals are received by several microphone components (ultrasonic transducers), and then digital signal analysis is performed on the audio signals at specific frequencies within the audio field. Based on the audio signal-temperature mapping relationship (obtained through pre-calibration), the temperature information corresponding to the audio path is obtained, thereby generating a three-dimensional temperature cloud map. The specific distributed ultrasonic temperature measurement network includes: a transmitter (speaker components, acting as ultrasonic transducers) emitting pulse-modulated waves of a specific frequency (kHz) (duty cycle 1:4); and a receiver (microphone components acting as ultrasonic transducers) analyzing the temperature variation characteristics of the sound velocity c = 331.5 + 0.6T. The humidity influence factor δ = 0.12e can be utilized. 0.03RH(RH represents humidity value) The compensated temperature calculation results are used to output a spatial thermal field matrix with an accuracy of 0.1℃. The specific implementation method of acoustic temperature measurement is existing technology, and the specific details will not be described in this embodiment.

[0156] When the humidity inside the vehicle is greater than or equal to a preset humidity threshold, the humidity affects the propagation speed of sound waves, thus significantly interfering with ultrasonic temperature measurement. In this embodiment of the invention, a virtual dummy model coupled with CFD (Computational Fluid Dynamics) simulation can be used to replace ultrasonic temperature measurement, generating a high-precision three-dimensional temperature / airflow field cloud map. The virtual dummy can simulate complex airflow structures (such as vortices and dead zones) with an accuracy of ±0.5℃. Specific details regarding the virtual dummy model coupled with CFD simulation temperature measurement are existing technologies and will not be elaborated upon in this embodiment.

[0157] In this embodiment of the invention, the humidity inside the vehicle can be obtained by a capacitive sensor (accuracy ±3%RH), which is commonly installed in the air intake duct of the HVAC (Heating, Ventilating and Air Conditioning) system, the seat ventilation duct, the rearview mirror cavity, etc.

[0158] In this embodiment of the invention, the outside temperature data can be obtained through an ambient temperature sensor.

[0159] Specifically, the thermal environment sensing parameters obtained in the embodiments of the present invention include initial thermal parameters and thermal management mode status.

[0160] The thermal management mode status includes the air blowing mode and its parameters, the waste heat recovery function status, and the user-defined historical function settings status. The air blowing mode and its parameters can be obtained through the HVAC controller's CAN message, and there are seven modes in total: face / foot / defrost / mixed, etc. The open / closed status of the corresponding damper and the angle of the motorized air outlet blades can be obtained for each mode. The waste heat recovery function status can be obtained through the corresponding status word in the battery management system. The user-defined historical function settings status can be obtained by reading the historical settings group stored in memory; when user historical preferences are obtained, these preferences are prioritized for execution.

[0161] Based on the aforementioned technical means, this invention acquires initial thermal parameters and thermal management mode status to accurately represent the current thermal environment state of the vehicle, thereby enabling the vehicle to know from which temperature state to begin adjustment. Combined with the optimal temperature state represented by the human thermal comfort model, this provides a basis for generating thermal management strategies. Specifically, a complementary approach is adopted, combining audio field calculation (low humidity) with CFD simulation coupled with a dummy model (high humidity), overcoming the limitation of traditional sensors that can only acquire single-point temperature data. In low humidity conditions, the temperature variation characteristics of the audio signal are used to generate accurate cloud maps; in high humidity conditions, CFD ensures accuracy, providing global temperature data for thermal management strategies and improving the spatial accuracy of control.

[0162] In some alternative implementations, step I2 above includes:

[0163] Step c1: Based on the vehicle's spatiotemporal parameters and light environment parameters, call the regional thermal comfort model library to select the first human body model family;

[0164] Step c2: Based on the physiological parameters of the occupants, the first human body model family is screened a second time to determine the second human body model family. One or more sub-human body thermal comfort models in the second human body model family constitute the target human body thermal comfort model. The sub-human body thermal comfort model corresponds to the thermal comfort of one occupant.

[0165] Specifically, the human thermal comfort model provided in this embodiment of the invention is a thermal sensation characteristics and thermal comfort domain model that is segmented into different population groups and regions. The core of the model is the thermal comfort equation for different population groups. The equation is obtained based on the mapping relationship between thermal comfort and different regions (provinces / cities), different seasons (all four seasons), different weather (sunny, rainy, snowy, foggy, etc.), different genders, different ages, different reproductive statuses (null / pregnant / sterilized).

[0166] When matching the target human thermal comfort model from the model library based on the parameters of the human thermal comfort model, the system first calls the stored regional thermal comfort model library. This model library is pre-divided into multiple sub-libraries according to the climate characteristics and lighting conditions of different regions. At this time, combined with the acquired vehicle spatiotemporal parameters, such as the vehicle's current geographical coordinates, local weather conditions (including temperature, humidity, and solar intensity), and light environment parameters such as illuminance and color temperature, the system filters out the set of models that match these parameters from the regional thermal comfort model library, forming the first human model family. For example, first, the system reads the vehicle's current location information, opens the model library of the corresponding region (province / city), then reads the current weather information, illuminance, and color temperature, and filters the model library of the region (province / city) according to the parameters, selecting those that meet the conditions to obtain the first human model family. This filtering step is mainly based on macroscopic environmental characteristics to ensure the basic adaptability of the selected models. For example, in southern regions with high solar radiation and high color temperature, models that focus more on heat dissipation regulation will be prioritized.

[0167] Based on this, if the occupants have not yet entered the vehicle, the vehicle thermal management system can generate a rough control strategy to pre-regulate the temperature inside the vehicle based on the first human body model family and the target temperature parameters initially set for the vehicle.

[0168] If the occupants have already entered the vehicle, the process proceeds to step c2. The system then uses the acquired occupant physiological parameters to further refine and filter the first human body model family. These physiological parameters include the number of occupants in the cabin, their individual positions, the effective thermal resistance of their clothing, body posture (such as torso lean angle and upper arm elevation angle), and the temperature, humidity, and pressure information related to the driver's contact with the steering wheel. By comparing these specific individual characteristic parameters with the models in the first human body model family, models that better match the actual physiological state of each occupant are selected to form the second human body model family. Since there may be multiple occupants in the vehicle, and each person's physiological parameters differ, the second human body model family may contain sub-human thermal comfort models corresponding to each occupant. These sub-models are set for the localized comfort temperature and humidity of different occupants. Combining them constitutes a target human thermal comfort model that accurately reflects the thermal comfort needs of all occupants in the vehicle.

[0169] In one specific implementation, a second human body model family is selected from the first human body model family. First, a first human body model category (used to control the temperature and humidity at different locations of the vehicle) is selected based on the number of occupants. Simultaneously, within the first human body model category, a corresponding first human body model genus (used to control the temperature and humidity at different parts of the human body) is selected according to the physiological parameter characteristics of different occupants. Then, a first human body model species (used to control the temperature and humidity in different areas of different parts of the human body) is further determined to obtain the aforementioned sub-human thermal comfort model, thereby pre-regulating the thermal management system.

[0170] The second human model family was determined through a screening process involving family, genus, and species.

[0171] For example, in "location-part-area", location refers to the position where the occupant sits (generally there are 5 or 7 fixed positions). Parts include, but are not limited to, one classification method. One possible classification method is to divide into 4 parts, including the head (which can be further divided into the cranium and face), neck, trunk (which can be further divided into chest, abdomen and pelvis), and limbs. The regions include, but are not limited to: 1) the frontal region (from the forehead to the supraorbital margin, including the frontal bone and frontal sinuses), the parietal region (the central part of the top of the skull, mainly the parietal bone), the occipital region (the posterior part of the skull, including the occipital bone and external occipital protuberance), and the temporal region (the lateral part of the skull, including the temporal bone and auricle); 2) the orbital region (containing the eyeball and optic nerve), the nasal region (including the nasal bone, nasal cavity and paranasal sinuses), the oral region (including the maxilla and mandible, lips and oral cavity), and the cheek region (the lateral part of the face, including the buccinator muscle and parotid duct); 3) the anterior cervical region, sternocleidomastoid muscle region, and lateral cervical region of the neck; 4) the anterior thoracic region and posterior thoracic region of the chest; 5) the right hypochondrium, left hypochondrium, right lumbar region, left lumbar region, umbilicus region, right inguinal region, left inguinal region, epigastric region, and pubic region of the abdomen; 6) the pelvic region and perineal region of the pelvis; 7) the thoracic and lumbar regions of the back; and 8) the upper and lower limb regions of the limbs (including the bone-joint-muscle chain).

[0172] This invention first narrows down the model scope using macroscopic regional and environmental parameters to ensure model adaptability to the overall environment. Then, it combines this with microscopic individual occupant physiological parameters for precise selection, ensuring the model fits the specific circumstances of each occupant. This tiered selection method considers both the general impact of different regional environments on human thermal comfort and the specific needs arising from individual occupant differences, avoiding the compatibility issues that may arise when using general models. The target human thermal comfort model obtained through this two-step selection process can more accurately match the vehicle's environment and the actual state of the occupants, providing a reliable basis for subsequent thermal management strategy development. This allows for more personalized and precise thermal management control of the vehicle, effectively improving the thermal comfort experience of each occupant while avoiding unnecessary energy consumption and improving the operational efficiency of the thermal management system. Furthermore, this two-step method of "initial screening of regional thermal comfort model libraries and secondary screening of occupant physiological parameters" overcomes the universality limitations of traditional general models. First, model families suitable for regional characteristics are selected based on location and lighting environment (before vehicle loading). Then, the models are refined by combining individual occupant physiological parameters to make the target models more closely match the specific environment and occupant needs, thereby improving the personalization and accuracy of thermal management (after vehicle loading). This achieves three-level directional temperature control of "person-part-area", meeting the needs of vehicle energy efficiency balance optimization, faster response speed, improved temperature uniformity, and personalized user requirements.

[0173] In some alternative implementations, step c3 includes:

[0174] Step c31: Use a large language model to extract feature vectors from the vehicle's spatiotemporal parameters and the light environment parameters;

[0175] Step c32: Obtain the regional thermal comfort model decoder. The regional thermal comfort model decoder is a mapping program pre-trained using vehicle spatiotemporal parameter samples, light environment parameter samples, and the regional thermal comfort model library.

[0176] Step c33: Input the feature vector into the regional thermal comfort model decoder for calculation to obtain the first human body model family.

[0177] Specifically, in real-world applications, the parametric model data in regional thermal comfort model libraries is enormous, often requiring a large amount of vehicle hardware storage space. This results in high storage costs, and selecting only a few parametric models from the regional thermal comfort model library that meet the vehicle's spatiotemporal and lighting environment parameters is inefficient.

[0178] Based on this, this embodiment of the invention further provides an improved pre-matching method for the first human body model family. First, multi-source data of vehicle spatiotemporal parameters and light environment parameters are input into a large language model for feature vector extraction. For example, the vehicle's current coordinates, monthly average ambient temperature, maximum relative humidity, and solar radiation intensity are input into the large language model. The geographic coordinates are obtained in real-time via a GPS module, and historical meteorological datasets (time span ≥ 10 years) are downloaded from a meteorological platform. The large language model uses a multimodal Transformer architecture to process heterogeneous data, and the calculation formula is as follows:

[0179]

[0180] Among them, v region The feature vector represents the regional thermal environment, GPS represents the vehicle's current coordinates, and T represents the feature vector. avg The average monthly ambient temperature, RH max I represents the highest relative humidity, and I represents the solar radiation intensity.

[0181] Then obtain the regional thermal comfort model decoder. MLP Decoder MLP The lightweight multilayer perceptron decoder is pre-trained using training samples composed of vehicle spatiotemporal parameter samples, light environment parameter samples, and the regional thermal comfort model library, and is stored in the vehicle as a mapping program.

[0182] Then, the above feature vector v region Input the data into the regional thermal comfort model decoder and calculate the corresponding first human body model family θ. model .

[0183] θ model =Decoder MLP (v region )

[0184] Among them, the first human body model family θ model The parameter set for the generated thermal comfort model includes, but is not limited to, the photothermal compensation coefficient β and the basal metabolic rate M of the human body (occupant). b And some parameters used in conjunction with occupant physiological parameters to calculate the second family of human models. Where β∈[β1,β2], β is used to adjust the psychological thermal sensation K. psych (E,CCT) represents the contribution weight of the photothermal scalar K, where β1 is the value corresponding to the lower limit of the interval and β2 is the value corresponding to the upper limit of the interval.

[0185] Based on the above technical means, replacing the regional thermal comfort model library with a regional thermal comfort model decoder reduces the storage space occupied by the vehicle hardware. On the other hand, for certain special scenarios where the regional thermal comfort model is not configured according to the corresponding parameters, a new regional thermal comfort model can be generated based on the regional thermal comfort model decoder, thereby improving the generalization ability of the regional thermal comfort model matching process.

[0186] In some optional implementations, step S103 above includes:

[0187] Step d1: When the occupants are not on the vehicle, the first vehicle thermal management strategy is determined based on the difference between the target thermal parameters and the thermal environment sensing parameters. The target thermal parameters are used to represent the default temperature state parameters of the vehicle.

[0188] Step d2: When the occupants board the vehicle, the multiple sub-human thermal comfort models in the second human body model family are sorted, and then the thermal management strategy of the second vehicle is determined according to the order of the sub-human thermal comfort models, so as to regulate the temperature of different body parts of different occupants in the vehicle in a specified order.

[0189] Specifically, taking new energy vehicles as an example, when the system detects that no occupants have boarded the vehicle, it first acquires the vehicle's thermal environment sensing parameters, including real-time interior temperature and humidity measured by sensors, as well as exterior ambient temperature. Simultaneously, it retrieves the vehicle's preset target thermal parameters. These target thermal parameters are default values ​​set based on standard comfort requirements, or values ​​remotely set by the user via a mobile app, or values ​​stored after the user's last use of the vehicle (e.g., a target temperature set by the user of 24℃). Before the user boards the vehicle, the system compares the real-time thermal environment parameters with the target thermal parameters and calculates the difference between them. Based on the read parameter information, it generates an initial compensation temperature T, including the initial thermal environment parameters, the target thermal environment parameters (e.g., the maximum heating temperature Tmax or the minimum cooling temperature Tmin), and the thermal management mode status. Then, based on the initial compensation temperature T, it determines the first vehicle thermal management strategy and activates the corresponding components of the thermal management system until the absolute value of the difference between the target thermal environment temperature and the initial thermal environment temperature is less than 1℃. During this process, individual differences are not considered; only the default target parameters are used as the adjustment benchmark.

[0190] Once the occupants board the vehicle, step d2 is executed. At this point, the system has already filtered and obtained a second human body model family based on previously acquired occupant physiological parameters. This model family includes not only sub-human thermal comfort models for each occupant, but also all human body models under the current initial thermal parameters, all human body models corresponding to the target thermal parameters for each occupant, and thermal management control strategies matching these models. Simultaneously, based on these initial models, target models, and corresponding strategies, the system generates a series of intermediate thermal keyframe human body models for the sub-human thermal comfort models. The number and specific parameters of these keyframe models are determined by the control strategy, used to achieve a smooth transition from the current thermal state to the target comfort state. Next, the system sorts the multiple sub-human thermal comfort models in the second human body model family. The sorting can be based on the occupant's position priority, such as the driver taking precedence over the front passenger and rear passengers, or it can be based on the physiological differences of the occupants to determine the order of adjustment. After the sorting is completed, the second vehicle thermal management strategy is determined according to the order. For example, the temperature of the driver's head, torso and other core comfort areas is controlled first. According to the parameters of the first keyframe model, the temperature of the air outlet around the driver is set to 23°C and the air volume is adjusted to level 4. After approaching the comfort state corresponding to the keyframe, the corresponding parts of the front passenger's body are controlled in sequence according to the sorting. The parameters are gradually adjusted according to the subsequent keyframe models until all parts of the body of all passengers reach their corresponding comfortable temperature and humidity range, thereby achieving precise temperature control in a hierarchical and sequential manner.

[0191] The technical solution provided by this invention addresses the scenario before passengers board the vehicle. By adjusting the difference between the default target parameters and the current environment, a basic comfortable environment is created for passengers in advance, avoiding extreme temperatures upon boarding and reducing waiting time. After passengers board, a second human body model family and its included sub-models and keyframe models are used, combined with sub-model sorting, to achieve a dynamic and orderly transition from the initial state to the target state. Through sub-model sorting and the application of keyframe models, the local comfort needs of different passengers can be accurately adapted, and temperature is adjusted in an orderly manner according to priority. This ensures a smooth temperature change process, significantly improving passenger comfort. It considers the individual comfort needs of different passengers and ensures the smoothness of temperature adjustment through keyframe models, avoiding discomfort caused by sudden increases or decreases. Compared to a uniform control method, this approach is more targeted and humanized, and can also avoid ineffective energy consumption to some extent.

[0192] In some optional implementations, the mapping relationship between the human thermal comfort model parameters of any occupant and the sub-human thermal comfort model is obtained through the following steps:

[0193] Step e1: Calculate the total heat transfer scalar between the human body and the environment, the photothermal scalar of the human body's perception of the light environment, and the dynamic human metabolic heat production scalar using the current human thermal comfort model parameters of the current occupant.

[0194] Step e2 involves allocating the total heat transfer scalar, photothermal scalar, and dynamic human metabolic heat production scalar to various body parts using physiological weighting coefficients; wherein, in one embodiment, the physiological weighting coefficients are dynamically generated by a large language model based on the vehicle's spatiotemporal parameters and light environment parameters.

[0195] Step e3: Determine the corresponding comfortable temperature and humidity range based on the heat corresponding to each part of the body;

[0196] Step e4: Generate a sub-human thermal comfort model based on the comfortable temperature and humidity range corresponding to the local body area, under the current human thermal comfort model parameter conditions.

[0197] Specifically, when it is necessary to establish a mapping relationship between the human thermal comfort model parameters of any occupant and the sub-human thermal comfort model, this embodiment of the invention will call the current human thermal comfort model parameters of the occupant, including occupant physiological parameters, vehicle spatiotemporal parameters, and light environment parameters, and calculate three key scalars based on these parameters. Among them, the total heat transfer scalar between the human body and the environment is obtained by integrating respiratory heat transfer, skin evaporative heat transfer, convective heat transfer, and radiative heat transfer, and after combining the metabolic rate correction coefficient, the total heat transfer scalar is finally calculated; the photothermal scalar of the human body's perception of the light environment is obtained by calculating the sum of infrared radiation heat and psychological heat perception based on illuminance and color temperature; the dynamic human metabolic heat production scalar is calculated by comprehensively considering real-time activity intensity, environmental correction, and physiological baseline values.

[0198] Since the human thermal comfort model provided in this embodiment of the invention is used to characterize the comfortable temperature and humidity range of different parts of the human body, the human thermal comfort model is represented by a matrix, and the parameters at different positions in the matrix are used to represent the comfortable temperature and humidity range of different parts of the human body.

[0199] Based on this, to generate the matrix of the human thermal comfort model, embodiments of the present invention can allocate the above three scalars to various parts of the human body according to preset physiological weight coefficients or coefficients generated by a large language model. The body parts are divided into six regions: head, trunk, upper arm, hand, thigh, and calf. The physiological weight coefficients of each region are set according to the differences in thermal sensitivity, such as 0.25 for the head, 0.3 for the trunk, 0.1 for the upper arm, 0.15 for the hand, 0.1 for the thigh, and 0.1 for the calf. This is just an example and is not a limitation.

[0200] After obtaining the total heat in each area, the comfortable temperature and humidity range can be determined based on the corresponding total heat in each body part. In a specific scenario, for example, heat and comfortable temperature and humidity may be negatively correlated; that is, the higher the local heat, the lower the required comfortable temperature and humidity. For example, the total heat of the head is 48.25 W / m². 2 The corresponding comfortable temperature and humidity range is 23-25℃; the total body heat is 57.9 W / m². 2 The corresponding comfortable temperature and humidity range is 24-26℃.

[0201] Finally, the system integrates the comfortable temperature and humidity ranges of various body parts to generate a sub-human thermal comfort model under the current human thermal comfort model parameters. This sub-model clearly marks the comfortable temperature and humidity ranges for the head, torso, upper arm, hand, thigh, and calf, forming a complete set of quantitative standards that reflect the individual thermal comfort needs of the occupant, providing a precise basis for the formulation of subsequent thermal management strategies.

[0202] This invention, through its embodiments, deconstructs the heat exchange between the human body and the environment, the effects of photothermal activity, and metabolic heat generation, precisely distributing macroscopic heat to various local areas. It then combines this with the correlation between local heat and comfortable temperature and humidity, achieving a mapping from abstract parameters to specific comfort needs. This approach overcomes the limitations of traditional models that only focus on the overall thermal state, fully considering the physiological differences and thermal sensitivities of different parts of the body, ensuring a high degree of consistency between the sub-human thermal comfort model and the actual experience of the occupants. The established mapping relationship transforms complex human thermal comfort model parameters into intuitive local comfortable temperature and humidity ranges, ensuring the personalization and accuracy of the sub-human thermal comfort model. Compared to traditional general models, the sub-models generated by this method can more meticulously meet the needs of different body parts of the occupants, providing a reliable basis for zoned temperature control. This improves the local comfort experience of the occupants while avoiding energy waste caused by overall control, making the thermal management system more efficient and better suited to individual needs.

[0203] In an optional implementation, the human thermal comfort model provided by this invention can obtain the comfortable temperature and humidity range of different groups of people under different working conditions by analyzing the relationship between local thermal sensation and physiological and environmental parameters. The relationship between local thermal sensation and physiological and environmental parameters can be obtained through the dynamic total heat transfer equation, the photothermal correlation equation, and other equations.

[0204] Therefore, in an optional implementation, the total heat exchange scalar between the human body and the environment provided by the embodiments of the present invention is calculated by the following formula:

[0205]

[0206] Among them, Q total and Q total′ Q is the total heat transfer scalar. res and Q res′ For heat exchange during human respiration, E s and E s′ For evaporative heat exchange through human skin, Q convection and Q convection′ For other convective heat transfer between the human body and the environment, Q radiation and Q radiation′ For other radiative heat exchange between the human body and the environment, γ is the metabolic rate correction factor.

[0207] Specifically, Q res It is a function related to the partial pressure of water vapor in the environment (Pa) and the air temperature (°C), and can be derived by inversely calculating the CO2 concentration from the HVAC gas sensor. Q convection Q is obtained through CFD simulation or an algorithm combining the temperature gradient of the NTC array. radiation The temperature difference between the body surface and the interior can be calculated using an infrared thermal imager. In this embodiment of the invention, Q... res Q convection Q radiation The calculation method is existing technology and will not be described in detail. Specifically, if the passengers are designated groups such as the elderly, children, or pregnant women, the accuracy of the total heat transfer is further corrected using a metabolic rate correction coefficient γ (less than 1). If the passengers are non-designated groups such as normal adults, no correction is needed. In this embodiment of the invention, the designated group is identified by the age range of the passengers and the parameter range into which their physiological parameters fall. The specific age range and parameter range are determined based on actual conditions and expert experience; this embodiment of the invention does not impose any special limitations. Furthermore, the specific value of the metabolic rate correction coefficient γ is defined according to actual conditions; this invention does not impose any special limitations.

[0208] This invention has modified the calculation method for evaporative heat transfer through human skin. The modified formula is as follows:

[0209] E s =M b ·η age +(0.8·ρ sweat ·RH 0.3 )

[0210] M b Basal metabolic rate (W / m 2 When the aforementioned first human body model family has no output values, the values ​​are taken with reference to the ISO 8996 standard. η age This is an age correction factor, set by the user according to an age range, for example, η for ages 18-30. age For a calibration value of 1, η is for individuals aged 60 and older. age The calibration value is 2. ρ sweat The dynamic weighting coefficient for sweat gland density has been calculated through the aforementioned steps, and RH represents the relative humidity of the human body.

[0211] Specifically, the human body information in the equation is based on the national standards "Anthropometric Dimensions of Chinese Adults", "Head and Face Dimensions of Adults", "Hand Dimensions Classification of Adults", "Foot Dimensions Analysis of Adults", "Anthropometric Dimensions of Chinese Minors", "Hand Dimensions Classification of Chinese Minors", "Head and Face Dimensions of Chinese Minors" and "Foot Dimensions Classification of Chinese Minors".

[0212] Based on the simplified formula for calculating evaporative heat transfer from human skin described above, the improved simplified model adopted in this proposal not only considers the influence of sweat glands but also takes into account that many physiological parameters are difficult to obtain in real time at the vehicle end. Therefore, the core principle is to use as few measurement parameters as possible. Considering specific common practical operating scenarios, such as seat heating and steering wheel heating, this model is more suitable for use in vehicles compared to conventional methods for calculating evaporative heat transfer from human skin.

[0213] Heat dissipation optimization driven by sweat gland distribution, E is calculated based on anatomical standard sweat gland density table partitioning. s This allows the strategy provided in the embodiments of the present invention to support local microclimate control at the 0.5°C level. Furthermore, the parameters used in the embodiments of the present invention, such as the number of sweat glands, are standard fixed values ​​obtained through anatomical measurement, eliminating the need for measurement and further reducing the difficulty of parameter acquisition.

[0214] Based on the aforementioned technical means, this embodiment of the invention introduces a metabolic rate correction coefficient into the heat transfer equation, optimizing heat transfer calculations for special groups such as the elderly and children, overcoming the bias caused by the universality of traditional models. The formula also integrates details such as respiratory heat transfer and skin evaporation, combined with age and sweat gland density corrections, making the total heat transfer calculation more closely aligned with the physiological characteristics of different passengers, thus improving the comfort of special groups. Furthermore, a simplified design for human skin evaporative heat transfer has been implemented, avoiding data that is difficult to collect accurately in real-time and involves cumbersome calculations in dynamic vehicle scenarios. Calculations can be performed using only basal metabolic rate, age correction coefficient, dynamic weighting coefficient of sweat gland density, and relative humidity—all parameters being readily available conventional data. This simplification retains core influencing factors (with sweat gland density additionally considered) while significantly reducing data collection difficulty and computational complexity, ensuring real-time updates of the total heat transfer scalar, and improving the dynamic adaptability and response speed of thermal management strategies.

[0215] In an optional implementation, the photothermal scalar quantity of human perception of the light environment provided by the embodiments of the present invention is calculated by the following formula:

[0216] K = K rad (E,CCT)+β·K psych (E,CCT)

[0217] K represents the photothermal scalar quantity, E represents the illuminance, CCT represents the color temperature, and β is the calibration coefficient. rad (E,CCT) represents the photothermal effect caused by infrared radiation from the light source, K. psych (E,CCT) represents psychological thermal sensation, which quantifies the weighted integral of psychological thermal sensation in the human eye's sensitive wavelength band V(λ) under specific color temperature illumination.

[0218]

[0219] Where E is the illuminance (in lx), S(λ) is the spectral power distribution of the light source (in W / nm), V(λ) is the spectral luminous efficiency function (dimensionless), λ represents the wavelength of light, and ψ(CCT) is the color temperature-thermal sensitivity conversion factor (dimensionless). If the occupants manually adjust the temperature, ψ(CCT) is updated through deep learning.

[0220] K psych The detailed calculation process for (E,CCT) is as follows:

[0221] (1) Obtain the relative spectral distribution through spectral normalization.

[0222]

[0223] (2) Quantify the combined effect Γ(CCT) of spectral shape and color temperature by integrating the color temperature spectrum:

[0224]

[0225] (3) The psychological thermal sensation K was obtained by using illuminance weighting. psych (E,CCT):

[0226]

[0227] Specifically, this invention proposes a mathematical mapping model of spectrum → color temperature → psychological thermal sensation, thereby further improving the accuracy of calculating local thermal sensation in the human body. This model is based on biophysics. Specifically, visible light stimulation of the retina → suprachiasmatic nucleus of the hypothalamus → adjustment of the body temperature set point leads to:

[0228] Low color temperature (warm light): promotes vasodilation → increases heat loss from the body surface → requires lowering the air conditioning temperature to compensate.

[0229] High color temperature (cold light): promotes vasoconstriction → reduces heat dissipation from the body surface → requires increasing air conditioning temperature to compensate.

[0230] Based on the above mechanism, this invention provides a method for calculating the photothermal scalar K. Here, E represents light illuminance, CCT represents light color temperature, and β is a calibration coefficient, typically 0.05-0.1. Because the photothermal correlation equation is a complex function, the photothermal scalar K needs to be calculated based on the spectral data of the actual light source (integrating the radiant power of a specific infrared band). Therefore, this invention's method for calculating the photothermal scalar K includes K... rad (E,CCT) and K psych (E,CCT) consists of two parts. K rad (E,CCT) is obtained from the actual heat load caused by infrared radiation from the light source (mainly the sun). It is related to the light illuminance, the light source spectrum (implied in CCT or requiring a separate parameter), and the occupant exposure. The calculation method of this parameter is existing technology and will not be described in detail in this embodiment.

[0231] K psych (E,CCT) represents psychological thermal sensation, which is related to the subjective thermal sensation shift caused by visible light (illuminance and color temperature). It characterizes the amount of thermal sensation shift induced by the color temperature-psychological effect in the human visible spectrum (380-780nm). Therefore, in this embodiment of the invention, it is qualitatively described as follows: at low color temperatures (e.g., <3000K), K... psych Increased K leads to a stronger subjective feeling of heat; at high color temperatures (e.g., >6000K), K... psych Decreasing the temperature reduces the subjective feeling of heat, thus yielding the aforementioned K. psych The expression.

[0232] Wherein, S(λ) represents the spectral power distribution of the light source, obtained through an RLS sensor and camera module, and V(λ) is the CIE spectral luminous efficiency function (dimensionless), which can be obtained with reference to the ISO / CIE 11664-1:2019 standard. Furthermore, the CIE 1931 standard colorimetric system indicates that the human eye has different sensitivities to different wavelengths of light; therefore, in this embodiment of the invention, S(λ)·V(λ) is used to characterize the subjective brightness stimulation of the human eye in the visible light band.

[0233] Wherein, Ψ(CCT) is the color temperature-thermal sensitivity conversion factor, which is a factor used to convert color temperature into thermal sensitivity. It is calculated as follows:

[0234]

[0235] In some alternative implementations, unreasonable output values ​​at extreme color temperatures (<2000K or >10000K) can be avoided by defining a constraint range (e.g., constraint range: 0.6≤Ψ(CCT)≤1.0).

[0236] Additionally, in some alternative implementations, when K psych After compensation, if the passenger manually adjusts the air conditioning temperature by more than ±0.5℃, it is recorded as a compensation deviation event. In this embodiment of the invention, Ψ(CCT) is further updated through deep learning:

[0237] Ψ new (CCT)=Ψ old (CCT)+α·(T manual -T auto )

[0238] Where α is the learning rate, and its value includes, but is not limited to, 0.01; Ψ old (CCT) represents the color temperature-thermal conversion factor before the update, Ψ new (CCT) is the updated color temperature-thermal conversion factor, T manual For manually adjustable temperature values, T auto The temperature value is automatically matched by the system.

[0239] In some alternative implementations, this can be achieved by defining a constraint range (e.g., constraint range: 0.5 ≤ Ψ). new ≤1.2), limiting the range of variation, preventing model divergence caused by a single abnormal adjustment, and improving model robustness.

[0240] Through the aforementioned technical means, the calculation method for photothermal scalars, by integrating objective infrared radiation heat and subjective psychological thermal sensation, significantly improves the accuracy of the human thermal comfort model in capturing the influence of the light environment. From a mechanistic perspective, based on the principle of biophysics (visible light influences the hypothalamic temperature set point via the retina), a quantitative mapping model of "spectrum → color temperature → psychological thermal sensation" is established. For the first time, the implicit influence of color temperature on subjective thermal sensation (such as low color temperature promoting vasodilation and high color temperature promoting vasoconstriction) is transformed into a calculable Kx. psych The parameters break through the limitations of traditional thermal comfort models that only focus on physical environmental parameters (temperature, humidity), making thermal sensation calculation more in line with the actual physiological response of the human body.

[0241] From the perspective of computational accuracy, K psych The expression accurately quantifies the shift in subjective thermal sensation in the 380-780nm visible band by integrating the spectral power distribution S(λ), the spectral luminous efficiency function V(λ), and the color temperature-thermal sensation conversion factor ψ(CCT). The specific formula for ψ(CCT) (combined with the hyperbolic tangent function) intuitively reflects the nonlinear relationship between color temperature and thermal sensation (e.g., thermal sensation is enhanced at <3000K and weakened at >6000K), upgrading the calculation of psychological thermal sensation from qualitative description to quantitative analysis, and significantly improving the accuracy of calculating local thermal sensation in the human body.

[0242] In terms of dynamic adaptability, updating Ψ(CCT) through a deep learning mechanism enables the model to continuously adapt to individual subjective differences in light and heat (such as different people having different sensitivities to warm / cold light), avoiding the universality bias of fixed parameter models and achieving personalized light and heat compensation.

[0243] In terms of practical application, this method allows the thermal management strategy to respond to both objective infrared radiation and subjective psychological thermal sensation, reducing the problem of "discrepancy between perceived and actual temperature" caused by the light environment and improving passenger comfort. At the same time, precise photothermal compensation can avoid over-adjustment and reduce the energy consumption of the cabin thermal management system.

[0244] In some alternative implementations, the dynamic human metabolic heat production scalar is calculated using the following formula:

[0245] M dyn =M act +M env +M phy

[0246] Among them, M dyn M is a scalar measure of dynamic human metabolic heat production. act To determine real-time activity intensity and metabolic heat production, data was obtained through joint calibration of steering wheel pressure, sweat, and seat pressure. M envTo refine the metabolic heat production correction for the environment, the current ambient temperature and humidity information obtained from internet weather forecasts, along with the predicted ambient temperature and dynamic response compensation temperature calculated through other steps, are used to obtain M. phy The baseline values ​​for metabolic heat production were obtained by referring to a table based on information such as the age, gender, and fertility status of the occupants (driver + passengers) (the data in the table were obtained from calibration experiments).

[0247] Based on the aforementioned technical methods, metabolic heat production is calculated by integrating real-time activity intensity, environmental correction, and physiological baseline values, thus overcoming the lag in traditional static metabolic estimation. This formula can reflect metabolic fluctuations caused by occupant activity (such as adjusting posture) and environmental changes in real time, enabling the human thermal comfort model to dynamically adapt to the human body's state and improve the control accuracy in transient scenarios.

[0248] Finally, the total heat transfer scalar Q was obtained. total Scalar of photothermal energy K and Scalar of dynamic human metabolic heat production M dyn By combining other parameters and comparing them with a model library, the optimal target human thermal comfort model is matched, and the corresponding thermal management strategy is executed. The model mapping relationship can be expressed by the following formula:

[0249]

[0250] T-HUMS is a matrix representation where each parameter represents the comfortable temperature and humidity range of a local part of the human body, thus representing the user's local thermal sensation.

[0251] In some alternative implementation methods, the approach also includes:

[0252] Step f1: Obtain the current ambient temperature and predict the ambient temperature when the vehicle travels to the preset position;

[0253] Step f2: Determine the compensation temperature based on the current ambient temperature and the predicted ambient temperature;

[0254] Step f3: Adjust the vehicle thermal management strategy based on the compensated temperature when the vehicle travels to the preset position.

[0255] Specifically, during vehicle operation, the system continuously acquires the current ambient temperature and, combined with the vehicle's GPS coordinates, speed, solar azimuth, and altitude, uses a pre-defined prediction model to calculate the predicted ambient temperature when the vehicle reaches a predetermined location. For example, if the vehicle is currently in a suburban area with an ambient temperature of 28°C, and navigation information indicates that it will enter a warmer industrial area in 10 minutes, the system predicts the ambient temperature to be 32°C based on historical meteorological data and real-time sunshine variations for that area.

[0256] After acquiring and predicting the ambient temperature, the system determines the compensation temperature based on the difference between the current ambient temperature and the predicted ambient temperature. The compensation temperature is calculated using the following formula:

[0257]

[0258] In the formula, ΔT dyn Indicates the compensation temperature, K p and K d These are the compensation coefficients (calibration values) for PID control; The temperature change rate is obtained by updating RLS sensor data in real time with a period of 0.5 seconds. ΔT is then calculated. dyn Then, it is input into the thermal management system controller as the basis for transient compensation control strategy regulation.

[0259] The system then records the coordinates of the preset location and the estimated arrival time. When the vehicle reaches that location, it adjusts the current thermal management strategy based on a determined compensation temperature. For example, if the HVAC system is originally set to 25°C and the airflow is at level 3, upon reaching the preset location, the system lowers the set temperature to 23°C and increases the airflow to level 4 based on a 2°C compensation temperature. At the same time, it adjusts the direction of the air outlets to enhance airflow to the upper body of the occupants, ensuring that the cabin temperature does not exceed the comfort range due to the increase in the ambient temperature.

[0260] Through the above-mentioned technical means, on the one hand, adjusting the strategy in advance can avoid drastic fluctuations in cabin temperature caused by sudden changes in ambient temperature, reducing the discomfort of passengers caused by sudden temperature changes; on the other hand, without the need to re-match the model, the system's computational load is reduced and the response speed is improved by only adjusting the temperature compensation strategy. At the same time, it avoids the energy waste that may be caused by frequent model adjustments, thus improving the operating efficiency of the thermal management system while ensuring comfort.

[0261] In some alternative implementation methods, the approach also includes:

[0262] Step g1: When the human thermal comfort model parameters change beyond a preset event threshold, return the step of matching the target human thermal comfort model from the model library based on the changed human thermal comfort model parameters.

[0263] Specifically, this invention also provides a dynamic response mechanism, including a transient compensation algorithm for sudden change scenarios. For example, when an occupant's sitting posture changes, the corresponding human thermal comfort model can be adjusted within 5 seconds based on the posture, and the corresponding thermal management strategy can be executed. Another example is a sudden increase in solar radiation (such as at a tunnel exit). By combining the windshield tilt angle / transmittance parameters (obtained through vehicle domain communication), the air supply strategy can be adjusted within 5 seconds to avoid localized overheating (measured radiation temperature rise > 10℃ / min).

[0264] Based on the aforementioned technical means, when the parameters of the human thermal comfort model change abruptly (such as changes in the number of occupants or posture), the model is automatically re-matched, solving the problem of traditional control lag. This mechanism ensures that the human thermal comfort model is always synchronized with the current state, avoiding control deviations caused by parameter changes, and improving the system's adaptability and response speed to dynamic scenarios.

[0265] In some alternative implementations, the method further includes:

[0266] Step h1: Determine the air temperature, relative humidity, and inner surface temperature of the glass inside the vehicle using the vehicle's spatiotemporal parameters and light environment parameters.

[0267] Step h2: Calculate the difference between the dew point temperature and the inner surface temperature of the glass based on the air temperature inside the vehicle, the relative humidity inside the vehicle, and the inner surface temperature of the glass.

[0268] Step h3: When the difference between the dew point temperature and the inner surface temperature of the glass is less than the preset temperature threshold and continues for a preset duration, it is determined that glass fogging is about to occur.

[0269] Step h4: When it is determined that glass fogging is about to occur, activate the vehicle's defogging equipment.

[0270] Based on the above-mentioned technical means, this embodiment of the invention also provides a method for preventing fogging in vehicles. During vehicle operation, the system combines the vehicle's spatiotemporal parameters and light environment parameters. An NTC temperature sensor deployed inside the vehicle measures the air temperature, a capacitive humidity sensor acquires the relative humidity, and an infrared sensor monitors the temperature of the inner surface of the windshield in real time. These parameters provide basic data for subsequent calculations.

[0271] The system then uses the improved Magnus formula to calculate an engineering approximation of the difference between the windshield's dew point temperature and the temperature of the glass's inner surface, as shown in the following formula:

[0272]

[0273] In the formula, ΔT dp T represents the difference between the dew point temperature and the temperature of the inner surface of the glass. a T represents the air temperature inside the vehicle, RH represents the relative humidity inside the vehicle, and T represents the air temperature inside the vehicle. glass The inner surface temperature of the glass. An engineering approximation of the dew point temperature.

[0274] The system will calculate ΔT dp Compared with the preset temperature threshold, when ΔT dpIf the state remains unchanged after <T℃ (T is the characteristic temperature) and for a preset duration (such as 3 s or more), it is determined that the windshield will fog up within 10 seconds. Generally, T can be 0 (℃) or 2 (℃).

[0275] Immediately trigger the defogging mechanism, control the blend door of the HVAC system to switch to the defogging mode, direct the air flow to the windshield, and adjust the air volume to 18 m 3 / h (meeting the requirement of ≥15 m 3 / h); meanwhile, if the ambient temperature <5℃, synchronously activate the PTC (Positive Temperature Coefficient) heater, and the power of the PTC heater is increased step by step to avoid temperature overshoot. The entire process from parameter acquisition to defogging start takes about 2 seconds, and the anti-fog preparation is completed 8 seconds in advance, effectively avoiding the impact of glass fogging on the driving vision.

[0276] Through the above technical means, based on accurate temperature and humidity parameter acquisition and coupling calculation, combined with the dew point temperature difference determination method to achieve fogging pre-judgment. Compared with the traditional method of defogging after fogging, the anti-fogging measures are started 10 seconds in advance, significantly improving driving safety; at the same time, the PTC heater is controlled in grades according to the ambient temperature, avoiding unnecessary energy consumption waste while ensuring the defogging effect, and taking into account both safety and economy.

[0277] In this embodiment, a spatial thermal microclimate multi-objective regulation device is also provided. This device is used to implement the above embodiment and preferred implementation manners, and those that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0278] This embodiment provides a spatial thermal microclimate multi-objective regulation device, as Figure 4 shown, the device includes:

[0279] A parameter acquisition module 501, configured to acquire human thermal comfort model parameters and thermal environment perception parameters. The human thermal comfort model parameters are the parameters required to match the human thermal comfort model. The human thermal comfort model parameters include occupant physiological parameters and vehicle spatio-temporal parameters, and the thermal environment perception parameters represent the current thermal environment state of the vehicle;

[0280] A model matching module 502, configured to match a target human thermal comfort model from the model library according to the human thermal comfort model parameters. The target human thermal comfort model is used to represent the comfortable temperature and humidity of different body parts of the human body;

[0281] The strategy determination module 503 is used to determine the vehicle thermal management strategy based on the target human thermal comfort model and thermal environment perception parameters.

[0282] System control module 504 is used to control the cabin domain thermal management system and the power domain thermal management system in the vehicle based on the vehicle thermal management strategy.

[0283] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0284] In this embodiment, a multi-objective space thermal microclimate control device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a thermal management control module and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0285] This invention also provides a multi-objective control system for space thermal microclimate, such as... Figure 5 As shown, it includes: a memory, a thermal management control module, a cockpit thermal management system, and a powertrain thermal management system.

[0286] The iTMS (Intelligent Thermal Management System) thermal management control module may further include hardware chips. These hardware chips can be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. The programmable logic devices can be complex programmable logic devices (CLPs), field-programmable gate arrays (FPGAs), general-purpose array logic (GDAs), or any combination thereof.

[0287] The memory stores instructions executable by at least one thermal management control module to cause the thermal management control module to perform the method shown in the above embodiments.

[0288] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory may include high-speed random access memory and non-transient memory, such as at least one disk storage device, flash memory device, or other non-transient solid-state storage device. In some alternative embodiments, the memory may include memory remotely located relative to the thermal management control module, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0289] The memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory may also include a combination of the above types of memory.

[0290] The cabin thermal management system includes a refrigerant circulation loop and a water circulation loop. For example... Figure 6 As shown, the refrigerant circulation loop includes: an evaporator (plus a blower), a pressure sensor, a compressor, a condenser, a liquid receiver, and an electronic expansion valve. These components are connected in series to form the refrigerant circulation loop. The water circulation loop includes: a condenser, a first three-way valve, a first two-way valve, a heater core (plus a blower), a multi-way valve, a first expansion tank, and a first water pump. These components are connected in series to form a water circulation loop for heating the passenger compartment. In some optional embodiments, the cold air core and the heater core are connected in parallel. When the cold air core is connected to the first three-way valve and the multi-way valve before and after, a water circulation loop can also be formed for cooling the passenger compartment. In some optional embodiments, the first three-way valve and the first two-way valve can be replaced by a multi-way valve. Figure 6 (middle dashed box).

[0291] Among them, the power domain thermal management system is used for cooling of electric drive / engine / battery, battery heating or waste heat recovery functions, and together with the crew cabin heating, it improves the working efficiency of the thermal management system and reduces energy consumption.

[0292] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated thermal management control module, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that the computer, thermal management control module, micro thermal management control module controller, or programmable hardware includes storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, thermal management control module, or hardware, the methods shown in the above embodiments are implemented.

[0293] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0294] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A multi-objective method for regulating space thermal microclimate, characterized in that, The method includes: Acquire human thermal comfort model parameters and thermal environment perception parameters. The human thermal comfort model parameters are the parameters required to match the human thermal comfort model. The human thermal comfort model parameters include occupant physiological parameters and vehicle spatiotemporal parameters. The thermal environment perception parameters represent the current thermal environment state of the vehicle. The target human thermal comfort model is obtained by matching the parameters of the human thermal comfort model from the model library. The target human thermal comfort model is used to represent the comfortable temperature and humidity of different parts of the human body. Based on the target human thermal comfort model and the thermal environment perception parameters, determine the vehicle thermal management strategy; The vehicle thermal management strategy controls the cabin thermal management system and the powertrain thermal management system within the vehicle.

2. The method according to claim 1, characterized in that, The step of matching the target human thermal comfort model from the model library based on the parameters of the human thermal comfort model includes: Separate the corresponding human thermal comfort model parameters for each occupant from the human thermal comfort model parameters; Based on the human thermal comfort model parameters corresponding to each occupant, a sub-human thermal comfort model corresponding to each occupant is matched from the model library. The sub-human thermal comfort model is used to represent the comfortable temperature and humidity of different parts of an occupant's body. All the sub-human thermal comfort models constitute the target human thermal comfort model.

3. The method according to claim 2, characterized in that, The steps to obtain parameters for a human thermal comfort model include: Collect information on the location and number of occupants inside the cabin; Based on the location information and the quantity information, the thermal resistance information of the occupants' clothing, the occupant's posture information, the temperature and humidity of the steering wheel, and the contact pressure information are collected, wherein the location information, the thermal resistance information of the occupants' clothing, the occupant's posture information, the temperature and humidity of the steering wheel, and the contact pressure information are used as the occupant's physiological parameters. The vehicle's coordinates and the weather information of its environment are collected as the vehicle's spatiotemporal parameters; Light illuminance and color temperature are collected as parameters of the light environment.

4. The method according to claim 3, characterized in that, Collect the location and number of occupants inside the cabin, including: When the in-cabin vision device is activated, image data collected by the in-cabin vision device is acquired, and the location and number information of the occupants are identified through the image data. When the in-cabin vision equipment is not activated, the occupant's position and number information are inferred from the seat pressure signals collected by the seat pressure sensor.

5. The method according to claim 3, characterized in that, Collect thermal resistance information of the clothing worn by occupants in the cabin, including: Collect initial thermal resistance information of the occupants' clothing inside the cabin; The initial clothing thermal resistance information is corrected by occupant posture parameters and sweat gland density parameters to obtain effective clothing thermal resistance information.

6. The method according to claim 5, characterized in that, The initial thermal resistance information of the occupants' clothing collected in the cockpit includes: When the in-cabin vision equipment is activated, it acquires images of the occupants' attire; The type of clothing worn by the occupants is identified from the images of their attire. The initial clothing thermal resistance information is determined based on the first mapping relationship between the clothing type and the standard thermal resistance; When the in-cabin vision equipment is not turned on, the seat piezoelectric array is activated, and the occupant's muscle tone changes are measured through seat micro-vibration. Infer clothing thickness based on the muscle tone change information; The initial clothing thermal resistance information is determined by the second mapping relationship between the clothing thickness and the standard thermal resistance; When the visual equipment in the cockpit is not turned on and the seat piezoelectric array is not installed, the initial clothing thermal resistance information is inferred based on the vehicle coordinates and the weather information of the vehicle's environment.

7. The method according to claim 5, characterized in that, The process of correcting the initial clothing thermal resistance information using occupant posture parameters and sweat gland density parameters to obtain effective clothing thermal resistance information includes: The effective clothing thermal resistance information c is calculated using the following formula. eff : clo eff =clo initial ×d posture ×(1-h·r sweat ·or evap ) Among them, clo initial For the initial clothing thermal resistance information, δ posture ρ is the attitude influence factor, ω is the coefficient, and ρ is the attitude influence factor. sweat η is the dynamic weighting coefficient for sweat gland density. evap δ represents the sweat evaporation efficiency. posture and ρ sweat The calculation method is as follows: d posture =1+k1·θ torso +k2·a arm Where, θ torso a is the angle of the occupant's torso leaning forward. arm The upper arm lifting angle is represented by k1 and k2, which are calibration coefficients; A u D represents the area of ​​the i-th partition after dividing the occupant's body surface into multiple partitions. u Let be the sweat gland density of the i-th partition.

8. The method according to claim 3, characterized in that, The method further includes: The air temperature, relative humidity, and inner surface temperature of the glass inside the vehicle are determined by the vehicle's spatiotemporal parameters and the light environment parameters. The difference between the dew point temperature and the inner surface temperature of the glass is calculated based on the air temperature inside the vehicle, the relative humidity inside the vehicle, and the inner surface temperature of the glass. When the difference between the dew point temperature and the inner surface temperature of the glass is less than a preset temperature threshold and continues for a preset duration, it is determined that glass fogging is about to occur. When it is determined that glass fogging is about to occur, activate the vehicle's defogging equipment.

9. The method according to claim 1, characterized in that, The steps for obtaining the thermal environment sensing parameters include: Acquire initial thermal parameters and thermal management mode status. The initial thermal parameters are used to represent the temperature status parameters of the current thermal environment of the vehicle, and the thermal management mode status is used to represent the activation status of various functions in the vehicle's thermal management system. The initial thermal parameters include the vehicle's internal temperature cloud map, external temperature data, and internal humidity. The vehicle's internal temperature cloud map is obtained through the following steps: When the humidity inside the vehicle is less than a preset humidity threshold, the temperature cloud map inside the vehicle is calculated by the audio field played inside the vehicle. When the humidity inside the vehicle is greater than or equal to a preset humidity threshold, a temperature cloud map inside the vehicle is generated using a dummy model coupled with CFD simulation.

10. The method according to claim 3, characterized in that, The step of matching the sub-human thermal comfort model corresponding to each occupant from the model library based on the human thermal comfort model parameters corresponding to each occupant includes: Based on the vehicle's spatiotemporal parameters and the light environment parameters, the regional thermal comfort model library is invoked to select the first human body model family; Based on the occupant's physiological parameters, the first human body model family is screened a second time to determine the second human body model family. One or more sub-human body thermal comfort models in the second human body model family constitute the target human body thermal comfort model.

11. The method according to claim 10, characterized in that, The process involves calling a regional thermal comfort model library based on the vehicle's spatiotemporal parameters and the light environment parameters to select a first family of human body models, including: Feature vectors are extracted from the spatiotemporal parameters of the vehicle and the optical environment parameters using a large language model; A regional thermal comfort model decoder is obtained, which is a mapping program pre-trained using vehicle spatiotemporal parameter samples, light environment parameter samples, and the regional thermal comfort model library. The feature vector is input into the regional thermal comfort model decoder for calculation to obtain the first human body model family.

12. The method according to claim 10, characterized in that, Based on the target human thermal comfort model and the thermal environment perception parameters, a vehicle thermal management strategy is determined, including: When the occupants are not on the vehicle, a first vehicle thermal management strategy is determined based on the difference between the target thermal parameter and the thermal environment sensing parameter. The target thermal parameter is used to represent the vehicle's default temperature state parameter. When occupants board the vehicle, multiple sub-human thermal comfort models in the second human body model family are sorted, and then the thermal management strategy of the second vehicle is determined according to the order of the sub-human thermal comfort models, so as to regulate the temperature of different body parts of different occupants in the vehicle in a specified order.

13. The method according to claim 2 or 10, characterized in that, The mapping relationship between the parameters of any occupant's human thermal comfort model and the sub-occupant's thermal comfort model is obtained through the following steps: The total heat transfer scalar between the human body and the environment, the photothermal scalar of the human body's perception of the light environment, and the dynamic human metabolic heat production scalar are calculated using the current human thermal comfort model parameters of the current occupants. The total heat transfer scalar, the photothermal scalar, and the dynamic human metabolic heat production scalar are allocated to various parts of the body using physiological weighting coefficients. The corresponding comfortable temperature and humidity range is determined based on the heat generated in each part of the body. A sub-human thermal comfort model is generated based on the comfortable temperature and humidity range corresponding to a local part of the body, under the current human thermal comfort model parameters.

14. The method according to claim 13, characterized in that, The total heat transfer scalar between the human body and the environment is calculated using the following formula: Among them, Q total and Q total′ Q is the total heat transfer scalar. res and Q res′ For heat exchange during human respiration, E s and E s′ For evaporative heat exchange through human skin, Q convection and Q convection′ For other convective heat transfer between the human body and the environment, Q radiation and Q radiation′ For other radiative heat exchange between the human body and the environment, γ is the metabolic rate correction factor; E s =M b ·or age +(0.8·ρ sweat ·RH 0.3 ) M b η is the basal metabolic rate of the crew. age ρ is the age correction factor. seeat RH represents the dynamic weighting coefficient for sweat gland density.

15. The method according to claim 13, characterized in that, The photothermal scalar quantity perceived by the human body in relation to the light environment is calculated using the following formula: K=K rad (E,CCT)+β·K psych (E,CCT) K represents the photothermal scalar quantity, E represents the illuminance, CCT represents the color temperature, β is the calibration coefficient, and K rad (E,CCT) represents the photothermal effect caused by infrared radiation from the light source, K. psych (E,CCT) represents psychological heat; Where S(λ) is the spectral power distribution of the light source, V(λ) is the spectral luminous efficiency function, λ represents the wavelength of the light, and Ψ(CCT) is the color temperature-thermal conversion factor. If the occupant manually adjusts the temperature, Ψ(CCT) is updated through deep learning.

16. The method according to claim 13, characterized in that, The dynamic scalar of human metabolic heat production is calculated using the following formula: M dyn =M act +M env +M phy Among them, M dyn M is the dynamic human metabolic heat production scalar value. act For real-time activity intensity metabolic heat production, M env M phy This represents the physiological baseline for metabolic heat production.

17. The method according to claim 3, characterized in that, The method further includes: Obtain the current ambient temperature and predict the ambient temperature when the vehicle travels to the preset position; The compensation temperature is determined based on the previous ambient temperature and the predicted ambient temperature. The vehicle's thermal management strategy is adjusted based on the compensation temperature when the vehicle travels to a preset position.

18. The method according to claim 1, characterized in that, The method further includes: When the changes in the parameters of the human thermal comfort model exceed a preset event threshold, the step of matching the target human thermal comfort model from the model library based on the changed human thermal comfort model parameters is returned.

19. A multi-objective control device for space thermal microclimate, characterized in that, The device includes: The parameter acquisition module is used to acquire human thermal comfort model parameters and thermal environment perception parameters. The human thermal comfort model parameters are the parameters required to match the human thermal comfort model. The human thermal comfort model parameters include occupant physiological parameters and vehicle spatiotemporal parameters. The thermal environment perception parameters represent the current thermal environment state of the vehicle. The model matching module is used to match the target human thermal comfort model from the model library according to the parameters of the human thermal comfort model. The target human thermal comfort model is used to represent the comfortable temperature and humidity of different parts of the human body. The strategy determination module is used to determine the vehicle thermal management strategy based on the target human thermal comfort model and the thermal environment perception parameters. The system control module is used to control the cabin thermal management system and the power thermal management system in the vehicle based on the vehicle thermal management strategy.

20. A multi-objective control system for space thermal microclimate, characterized in that, include: The system includes a memory, a thermal management control module, a cockpit-domain thermal management system, and a powertrain-domain thermal management system. The memory and the thermal management control module are interconnected. The memory stores computer instructions. The thermal management control module executes the computer instructions to perform the method described in any one of claims 1 to 18 to control the cockpit-domain thermal management system and the powertrain-domain thermal management system.

21. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method of any one of claims 1 to 18.

22. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the method of any one of claims 1 to 18.

Images