A thermal management method, device and medium for a light truck cab.

Abstract

This application discloses a thermal management method, device, and medium for a light truck cab. The method includes: acquiring key component data and driver data of the cab; determining a set of equations for the corresponding key components based on the key component data and driver data; determining multiple hierarchical thermal resistance network models based on the set of equations; acquiring internal and external environmental parameters of the cab; training and integrating a pre-set neural network model based on the internal and external environmental parameters and the multiple hierarchical thermal resistance network models to obtain a thermal resistance network model; and predicting the heat flow of the cab using the thermal resistance network model. This application combines key component and driver data to construct a hierarchical thermal resistance network model, and then uses internal and external environmental parameters to train and integrate the neural network. This can accurately reflect the thermal characteristics of the cab, improve the accuracy of heat flow prediction, provide a reliable basis for thermal management, and help optimize design, improve energy efficiency, and enhance comfort.

Description

Technical Field

[0001] This application relates to the field of new energy vehicle technology, and in particular to a thermal management method, device and medium for a light truck cab. Background Technology

[0002] The thermal management of the cab of an electric light truck directly impacts driver comfort and vehicle energy consumption. Currently, light truck cabs on the market consist of multiple parts, including glass, outer sheet metal, air and felt layers, and interior trim. These components exhibit significantly different thermal characteristics under different operating conditions. Traditional cab thermal management modeling fails to fully consider the dynamic interactions between these components, leaving considerable room for optimization in temperature, humidity control, and energy consumption management. Furthermore, factors such as differences in driving speed, ambient temperature variations, vehicle load, solar radiation interference, nonlinear responses of the air conditioning system, and waste heat utilization from the electric light truck battery further complicate modeling. In practical applications, these factors lead to poor control accuracy in the thermal management system, failing to meet drivers' demands for a comfortable driving environment and potentially wasting energy. From a technological perspective, modeling methods based on pure physical equations require extensive experimentation to calibrate parameters and struggle to adapt to dynamic environmental changes; while purely data-driven AI models, lacking constraints on physical mechanisms, are prone to overfitting. Summary of the Invention

[0003] To address the aforementioned problems, this application proposes a thermal management method for a light truck cab, comprising: acquiring key component data and driver data of the cab, wherein the key components include glass, outer sheet metal, air layer, felt layer, and interior materials; determining a set of equations for the corresponding key components based on the key component data and driver data; determining multiple layered thermal resistance network models based on the set of equations; acquiring internal and external environmental parameters of the cab; training and integrating a pre-set neural network model based on the internal and external environmental parameters and the multiple layered thermal resistance network models to obtain a thermal resistance network model; and predicting the heat flow of the cab using the thermal resistance network model.

[0004] In one example, acquiring key component data and driver data for the cab specifically includes: collecting data on the light transmittance, thermal conductivity, specific heat capacity, geometry, and area of ​​the glass components; recording data on the thickness, thermal conductivity, specific heat capacity, geometry, and coverage area of ​​the outer sheet metal layer; measuring the thickness, air velocity, and heat exchange coefficient of the air layer; acquiring data on the thermal insulation performance, moisture absorption characteristics, and heat capacity of the felt layer; collecting data on the surface emissivity, thermal resistance, and geometry of the interior materials; determining the thermophysiological model corresponding to the driver; and acquiring node heat data based on multiple nodes of the thermophysiological model.

[0005] In one example, determining the corresponding set of equations for the key components based on the key component data and driver data specifically includes: determining an unsteady-state heat conduction equation based on the glass component data, the unsteady-state heat conduction equation including light transmittance, solar radiation absorptivity, and convective heat transfer between the inner and outer surfaces; determining a thermal resistance network model based on the outer sheet metal data, and determining a heat balance equation based on each node of the thermal resistance network model; determining an equivalent thermal resistance and heat capacity model based on the air layer and the felt layer data, and determining a heat transfer equation based on the equivalent thermal resistance and heat capacity model; determining a dynamic thermal response equation and a contact heat transfer equation based on the interior material data and the node heat data; and integrating the thermophysiological model, the unsteady-state heat conduction equation, the heat balance equation, the heat transfer equation, the dynamic thermal response equation, and the contact heat transfer equation to determine the set of equations.

[0006] In one example, the acquisition of internal and external environmental parameters of the cab includes: real-time monitoring of the ambient temperature, relative humidity, and solar radiation intensity of the external environment of the cab; recording the vehicle speed to determine the air convection coefficient based on the vehicle speed; collecting temperature and humidity data at multiple locations inside the cab; and acquiring the operating status of the air conditioning system, which includes the set temperature, fan speed setting, and air outlet temperature.

[0007] In one example, a pre-set neural network model is trained and integrated based on the internal and external environmental parameters and the multiple hierarchical thermal resistance network models. Specifically, this includes: preprocessing the internal and external environmental parameters to determine key features, including ambient temperature, ambient humidity, solar radiation intensity, vehicle speed, temperature difference between the inside and outside of the glass, sheet metal temperature gradient, air layer heat flux density, and interior surface temperature; determining a pre-set multilayer perceptron and its network structure, which includes an input layer, multiple hidden layers, and an output layer; the number of nodes in the input layer is equal to the number of key features; the number of neurons in each hidden layer is set to a predetermined initial range, and ReLU is used as the activation function; the output layer is designed to display the temperature and humidity values ​​of multiple areas inside the driver's cab.

[0008] In one example, predicting the heat flow of the cab using the thermal resistance network model specifically includes: dividing the cab into multiple node systems, including series and parallel thermal resistance systems; calculating heat transfer using a pre-set set of formulas to obtain preliminary prediction results; inputting the input parameters of the current operating condition into the thermal resistance network model to output prediction residuals; adjusting the key parameters of the thermal resistance network model based on the prediction residuals, including equivalent thermal conductivity, convective heat transfer coefficient, and radiative heat transfer parameters; and recalculating using the adjusted thermal resistance network model to obtain corrected temperature prediction results.

[0009] In one example, after predicting the heat flow of the cab using the thermal resistance network model, the method further includes: verifying the prediction results of the thermal resistance network model, deploying the verified thermal resistance network model into the thermal management control system of the light truck, so as to predict the temperature and humidity values ​​of the cab through the thermal management control system, and to coordinate the control of thermal management components based on the temperature and humidity values, wherein the thermal management components include an air conditioning subsystem, a ventilation subsystem, and a battery cooling subsystem.

[0010] In one example, the method further includes: periodically collecting new data on the actual operation of light trucks, including thermal management-related data under different seasons, different regional climates, various driving conditions, and different load conditions; and updating and training the thermal resistance network model deployed in the thermal management control system based on the new data so that the thermal resistance network model adapts to the actual operating conditions of the vehicle.

[0011] On the other hand, this application also proposes a thermal management device for a light truck cab, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the thermal management device for a light truck cab to perform: the method described in any of the examples above.

[0012] On the other hand, this application also proposes a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to be the method described in any of the examples above.

[0013] This application involves comprehensive multi-dimensional data collection, covering key components, driver parameters, and internal and external environmental parameters, providing rich evidence for accurate modeling. Secondly, when constructing the hierarchical thermal resistance network model, corresponding equations are determined for different component characteristics, meticulously reflecting the thermal properties of each component and improving model accuracy. By training and integrating the neural network using internal and external environmental parameters and the hierarchical thermal resistance network model, the powerful learning and prediction capabilities of the neural network can be fully utilized, resulting in a more reliable thermal resistance network model. In terms of heat flow prediction, the cab is treated as a node system for calculation, and key parameters are adjusted based on the model output prediction residuals to obtain accurate, corrected prediction results. After verifying the prediction results, the model is deployed to the thermal management control system, enabling real-time prediction of temperature and humidity values. This guides the coordinated control of thermal management components such as air conditioning, ventilation, and battery cooling, achieving precise thermal comfort regulation and reducing energy consumption. Furthermore, regularly collecting new data to update and train the model ensures it adapts to actual vehicle operating conditions, maintains excellent predictive performance, and effectively responds to changes in different seasons, regions, operating conditions, and load conditions. This provides a scientific, efficient, and dynamic solution for thermal management of light truck cabs, improving overall vehicle performance and user experience. Attached Figure Description

[0014] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0015] Figure 1 This is a schematic flowchart illustrating a thermal management method for a light truck cab according to an embodiment of this application.

[0016] Figure 2 This is a schematic diagram of a thermal management device for a light truck cab according to an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0019] like Figure 1 As shown, in order to solve the above problems, this application provides a thermal management method for a light truck cab, the method comprising:

[0020] S101. Obtain key component data and driver data for the cab. The key components include glass, outer sheet metal, air layer, felt layer, and interior materials. Determine the corresponding set of equations for the key components based on the key component data and driver data, and determine multiple layered thermal resistance network models based on the set of equations.

[0021] Collect parameters of key components of the cab, including the geometry, material properties and relative positions of key components such as glass, outer sheet metal, air and felt layers, and interior trim, as shown in Table 1.

[0022] Table 1

[0023]

[0024]

[0025] In addition, it is necessary to collect parameters such as light transmittance and thermal conductivity of different types of glass, such as windshield and side window glass; thickness, thermal conductivity, and specific heat capacity of the outer sheet metal; thickness, air velocity, and heat exchange coefficient of the air layer; thermal insulation performance, moisture absorption characteristics, and heat capacity of the felt layer; and surface emissivity and thermal resistance of the interior materials, as shown in Table 2.

[0026] Table 2

[0027]

[0028]

[0029] In one embodiment, the modeling of the cab model begins with modeling the glass layer within the main body of the cab. This involves considering factors such as the glass's light transmittance, solar radiation absorptivity, and convective heat transfer between the inner and outer surfaces, establishing an unsteady-state heat conduction equation. According to the law of conservation of energy, the rate of change of heat per unit volume within the glass layer equals the net rate of heat input into that volume. Specifically, when considering heat conduction, based on Fourier's law, the heat flux density... Where k is the thermal conductivity of the glass. This represents the temperature gradient. For solar radiation, let the solar radiation absorptivity be α and the solar radiation intensity be I, then the amount of solar radiation absorbed per unit volume is αI. For convective heat transfer between the inner and outer surfaces, let the convective heat transfer coefficient of the inner surface be h1, the convective heat transfer coefficient of the outer surface be h2, the indoor temperature be T1, and the outdoor temperature be T2, then the heat transfer per unit area of ​​the inner surface is h1(T-T1), and the heat transfer per unit area of ​​the outer surface is h2(T2-T). Let the glass layer thickness be L and the transmittance be τ. When establishing the heat conduction equation, transmittance mainly affects the amount of solar radiation reaching the glass, because the radiation passing through the glass is reduced. Considering all these factors, a one-dimensional unsteady-state heat conduction equation can be obtained:

[0030]

[0031] Where ρ is the density of the glass, c is the specific heat capacity of the glass, t is time, and x is the coordinate along the glass thickness direction. The parameters required for modeling include the glass's thermal properties ρ, c, and k; solar radiation-related parameters I, α, and τ; convective heat transfer coefficients h1 and h2; boundary temperatures T1 and T2; the geometric parameter glass layer thickness L; and the initial temperature distribution T(x,0) of the glass layer. By solving the above equations, the temperature distribution T(x,t) within the glass layer as a function of time t and location x can be obtained. This leads to information such as the temperature changes of the inner and outer surfaces of the glass layer over time, and the heat flow through the glass layer over time, thereby enabling the analysis of the glass layer's heat transfer characteristics and insulation performance.

[0032] In one embodiment, sheet metal layer modeling aims to calculate the temperature of thermal resistance network nodes based on the material's thermal conductivity and thickness. First, a geometric model is constructed. Based on the actual size and shape of the cab, a precise geometric model of the sheet metal layer is built using 3D modeling software, clearly defining the dimensions, shapes, and spatial relationships of each part, paying particular attention to details such as thickness variations in different areas. Next, the material model is determined, clarifying the thermophysical properties of the materials used in the sheet metal layer, primarily the thermal conductivity, as different materials have different thermal conductivity. Then, a thermal resistance network model is established, discretizing the sheet metal layer into a large number of nodes. Each node represents a small region within the sheet metal layer, and adjacent nodes are connected by thermal resistance, thus simulating the heat conduction process within the sheet metal layer. In this model, nodes are the basic units, distributed according to a certain pattern within the sheet metal layer space. Their temperature reflects the thermal condition of their location. The node distribution density is determined based on the complexity of the sheet metal layer structure and the required calculation accuracy; in areas with large structural changes or drastic changes in heat flux density, the node distribution is denser. The thermal resistance connecting adjacent nodes is determined by the material's thermal conductivity and the distance between nodes, i.e., the sheet metal layer thickness. The calculation formula is R = d / kA. It is usually assumed that the heat transfer area A is a unit area, so R = d / k. Thermal resistance reflects the ease or difficulty of heat transfer between nodes; the greater the thermal resistance, the more difficult the heat transfer. The calculation process for the node temperature in a thermal resistance network is as follows: First, determine the boundary conditions, clarifying the boundary temperature or heat flux density of the sheet metal layer. For example, it is known that one side of the sheet metal layer is in contact with a high-temperature environment, with a temperature of T. h The other side is in contact with a low-temperature environment, with a temperature of T. c The temperatures on both sides can be used as known temperature conditions for the boundary nodes; then, the thermal resistance network equation is established. According to the principle of thermal balance, for each internal node, the heat flowing into the node is equal to the heat flowing out of the node.

[0033] Taking a one-dimensional thermal resistance network as an example, let the temperature of node i be T. i The temperatures of its left and right adjacent nodes are T respectively. i-1 and Ti+1 Then there is Based on this, the heat balance equations for all nodes can be listed, forming a set of equations. Finally, the set of equations is solved. Numerical methods such as Gaussian elimination and iterative methods can be used to solve the equations to obtain the temperature value of each node. Through the above thermal resistance network model and calculation process, the temperature of each node in the thermal resistance network can be calculated relatively accurately based on the thermal conductivity and thickness of the sheet metal layer, thereby understanding the temperature distribution within the sheet metal layer.

[0034] In one embodiment, in the modeling of the air and felt layers, the air gap and the felt are regarded as a composite insulation layer, and equivalent thermal resistance and heat capacity are introduced to construct the model. The model structure comprises three parts: thermal resistance, thermal capacity, and heat flow path. Regarding thermal resistance, the air gap and felt each have their own thermal resistance, which are connected in series. The equivalent thermal resistance of the composite insulation layer is equal to the sum of their thermal resistances. Air has low thermal conductivity and high thermal resistance, effectively hindering heat transfer. Felt, as a porous material, also has a certain insulation capacity and thermal resistance value. Regarding thermal capacity, both air and felt have their own thermal capacities, which can be equivalent to the thermal capacity of the composite insulation layer. Thermal capacity reflects the material's ability to store heat. Air thermal capacity is related to mass and specific heat capacity, while felt thermal capacity is related to material properties, mass, and other factors. The equivalent thermal capacity affects the rate of temperature change during heat transfer. Assuming heat is transferred from one side to the other through the composite insulation layer, the heat flow first passes through the air gap and then through the felt, or vice versa. When establishing the model, the direction of heat flow and boundary conditions are determined based on actual conditions, such as the given temperatures on both sides of the insulation layer or the heat flux density on one side. Then, the temperature distribution and heat flow within the insulation layer are calculated by solving the heat conduction equation.

[0035] Taking one-dimensional heat conduction as an example, the composite insulation layer is divided into multiple micro-segments, each with a corresponding equivalent thermal resistance and heat capacity, similar to an RC network in a circuit. A heat balance equation is established to describe the heat transfer and storage process in each micro-segment, thereby analyzing the thermal performance of the entire composite insulation layer under different thermal boundary conditions. Specifically, in the thermal resistance network, the air gap thermal resistance R... air Depends on its thickness L air air thermal conductivity λ air And the heat transfer area A can be expressed as Felt thermal resistance R felt Similarly, based on the felt thickness L felt Thermal conductivity λ felt The area A determines the result, i.e. The equivalent thermal resistance R of the composite insulation layer eq =R air +R felt In a heat capacity system, the heat capacity of air, C air air quality m air and specific heat capacity c air Related to, i.e., Cair =m air c air The heat capacity of felt C felt =m felt c felt The equivalent heat capacity C of the composite insulation layer eq =C air +C felt During heat transfer, it is assumed that heat flows from the high-temperature side T. h Heat is transferred to the low-temperature side T through the composite insulation layer. c The heat flux density q obeys Fourier's law, that is... The calculation process includes temperature change calculation and heat flow calculation. When calculating temperature change, according to the principle of heat balance, for a small time interval Δt, the heat flowing into the composite insulation layer equals the increase in the internal energy of the insulation layer, i.e., qAΔt=C eq ΔT, where ΔT is the change in temperature of the insulation layer, can be obtained by substituting the expression for heat flux density q. The temperature distribution of the insulation layer at different times can be obtained through iterative calculation. When calculating heat flow, the equivalent thermal resistance and the temperature difference between the two sides are known. The heat flow through the composite insulation layer can be calculated. In practical applications, numerical calculations or analytical solutions are performed based on specific boundary and initial conditions, combined with the above formulas, to analyze the insulation performance and temperature change patterns of the composite insulation layer. Real-world situations may be more complex, requiring adjustments to the model to account for factors such as convective and radiative heat transfer.

[0036] In one embodiment, interior trim modeling requires establishing dynamic thermal response equations and interior-driver contact heat transfer equations based on the heat capacity and surface radiation characteristics of the interior materials. The dynamic thermal response equations are based on the law of conservation of energy, meaning the heat accumulation of the interior materials equals the absorbed heat minus the dissipated heat. This process considers the material's heat capacity, reflecting its ability to absorb or release heat, and the surface radiation characteristics, which determine the rate at which the material exchanges heat with the surrounding environment through radiation. The expression is as follows: In the formula, m is the mass of the interior material, C is the specific heat capacity of the material, T is the temperature of the material, t is the time, and q is the mass of the interior material. in The heat flux density that enters the material through conduction and other means, q out It is the heat flux density that leaves the material through conduction and other means. is the radiative heat flux density, ∈ is the emissivity of the material, σ is the Stefan-Boltzmann constant, A is the surface area of ​​the material, and T sur The temperature is the ambient temperature. The equation takes into account parameters such as the material's mass, specific heat capacity, emissivity, surface area, ambient temperature, and heat flux density, and outputs the temperature of the interior material as a function of time.

[0037] The heat transfer equation between the interior and driver contact is based on Fourier's law, which states that heat is transferred from a high-temperature object to a low-temperature object at a rate proportional to the temperature gradient. Considering the heat transfer when the driver is in contact with the interior, the contact area, contact thermal resistance, and the temperature difference between the two are key factors. Its expression is q. c =hA(T d -T i ), where q c Where h is the contact heat flux density, A{c} is the contact heat transfer coefficient, and T is the contact area. d It is the driver's skin temperature, T i This refers to the interior material temperature. The equation takes the contact heat transfer coefficient, contact area, driver's skin temperature, and interior material temperature as inputs, and outputs the heat flux density between the driver and the interior. This heat flux density can be used to further analyze the driver's perceived thermal comfort and other factors.

[0038] In one embodiment, the driver model is a driver thermophysiological model designed to simulate the heat generation, transfer, and exchange processes in different parts of the driver's body, focusing on key nodes such as the head, chest, and feet. Specifically, the model designates the driver's head as one node, considering heat generation from brain metabolism, which is approximately 20W. This node primarily exchanges heat with the roof via radiation, transferring heat to the surrounding environment. The chest is designated as another node, generating approximately 40W of heat through cardiopulmonary activity. This node, in contact with the seat, transfers heat to the seat via conduction, simulating heat exchange between the chest and the seat. The feet are also designated as a node, quantifying heat generation at approximately 15W. This node exchanges heat with the floor via conduction and also has a convective coupling effect with the air conditioning vents, simulating heat transfer between the feet and the surrounding environment. By simulating the heat generation at these three nodes and their heat exchange with the surrounding environment, we can comprehensively describe the driver's thermophysiological state in the in-vehicle environment, providing theoretical basis and model support for further research on driver thermal comfort and in-vehicle environment control.

[0039] S102. Obtain the internal and external environmental parameters of the cab, train and integrate the pre-set neural network model according to the internal and external environmental parameters and the multiple hierarchical thermal resistance network models to obtain the thermal resistance network model, and predict the heat flow of the cab through the thermal resistance network model.

[0040] In one embodiment, neural network-assisted modeling plays a crucial role in the field of cockpit thermal management. The model's external influence input parameters encompass temperature gradients at each layer, ambient temperature and humidity, solar radiation intensity, vehicle speed affecting the convection coefficient, temperature difference between the inside and outside of the glass, sheet metal temperature gradient, air layer heat flux density, interior surface temperature, and air conditioning operating status. The model parameter output correction term is the residual between the neural network's predicted physical model and the actual temperature, and based on this, the thermal resistance network parameters, such as the equivalent thermal conductivity, are dynamically adjusted. Considering the electric vehicle cockpit scenario under high summer temperatures, the external influence input parameters are first defined. In a summer midday test scenario, environmental parameters are collected in real time, such as an ambient temperature of 40℃, relative humidity of 50%, and solar radiation intensity of 900W / m². 2 The system collects vehicle operating parameters, such as vehicle speed 80km / h, air conditioning in cooling mode with a set temperature of 22℃ and fan speed at level 3; and structural parameters, such as outer glass temperature 65℃, inner glass temperature 30℃, and a temperature difference of 35℃ between the inside and outside of the glass; roof sheet metal temperature 60℃, interior trim temperature 32℃, and a sheet metal-interior trim temperature gradient of 28℃; and the heat flux density of the air layer near the air conditioning vents is -50W / m³. 2 The value near the car window is 15W / m. 2The seat surface temperature is 35℃, and the dashboard surface temperature is 38℃. Next, a preliminary physical model calculation is performed based on a thermal resistance network. Using a traditional thermal resistance network model, the cab is simplified into a node system consisting of glass, sheet metal, interior trim, and air connected in series / parallel. Heat transfer is calculated based on Fourier's law of thermal conduction and Newton's law of cooling. Based on the above input parameters, the model initially predicts an average temperature of 28℃ inside the cab. Then, a neural network prediction residual is calculated. Historical operating condition data, including input parameters and actual temperatures under different seasons, vehicle speeds, and air conditioning conditions, are collected in advance to construct a dataset. The neural network is then trained using methods such as LSTM or a multilayer perceptron. The 12 input parameters of the current operating condition are input into the trained neural network in real time. The model outputs a prediction residual of -3℃, meaning the physical model's prediction is 3℃ higher than the actual temperature. Afterward, the thermal resistance network parameters are dynamically adjusted based on the residual predicted by the neural network. The temperature difference (-3℃) was reduced by lowering the equivalent thermal conductivity of the glass and sheet metal, such as adjusting the equivalent thermal conductivity of the glass from 1.2 W / (m·K) to 1.0 W / (m·K) to reduce heat transfer into the cabin. The convective heat transfer coefficient between the air conditioning vents and the air was fine-tuned to enhance the cooling effect. The emissivity of the interior surfaces was reduced to reduce the temperature rise caused by solar radiation. After the adjustments, the thermal resistance network model was recalculated, and the corrected temperature was 25℃. The model was then validated and iterated. The average temperature inside the cabin was measured to be 25.5℃ using the in-vehicle temperature sensor. The error of the corrected model was reduced to 0.5℃, which significantly improved the accuracy compared to the initial model. At the same time, the input parameters of the current operating condition, the prediction residual, the corrected parameters, and the actual temperature were fed back to the neural network to continuously update the model and improve the long-term prediction accuracy. This model dynamically corrects the thermal resistance network parameters through a neural network, enabling it to adapt to complex and ever-changing driving environments such as sudden drops in solar radiation caused by heavy rain and changes in wind speed during high-speed driving. Compared to traditional static models, it can predict temperature more accurately and assist the air conditioning system in intelligently adjusting the cooling capacity to achieve a balance between energy saving and comfort.

[0041] In one embodiment, to comprehensively and accurately collect thermal management data related to the light truck cab, numerous high-precision sensors are deployed at key locations both inside and outside the cab. Specifically, temperature sensors with an accuracy of ±0.1℃ are installed on the inner and outer surfaces of the glass, the inner and outer surfaces of the sheet metal, different cross-sections of the air layer, key points inside the felt layer, interior surfaces, dashboard air vents, and the driver's head, chest, and feet. Humidity sensors are placed inside the cab to monitor changes in internal humidity in real time. Flow velocity sensors are installed at the air conditioning vents and air inlets to accurately capture airflow speed. In addition, light intensity sensors and vehicle speed sensors are also provided to comprehensively collect environmental and operational data related to thermal management. All sensors are connected to the onboard computer via a data acquisition module, with the data acquisition frequency set between 2Hz and 5Hz to ensure real-time and accurate capture of dynamic changes in the thermal environment. The collected data covers a wide range, including information from different seasons, different regional climates, various driving conditions, and different load conditions. Driving conditions include urban congestion, highway cruising, slow-moving rural roads, and idling. Load conditions include unloaded, half-loaded, and fully loaded. Simultaneously, diverse CAN bus integrated data is generated through simulation to obtain relevant data on air conditioning, batteries, etc. On the software side, the embedded AI chip can run the thermal management model in real time and output optimal control commands through a PID controller or reinforcement learning agent.

[0042] In one embodiment, preprocessing is required for the massive amounts of collected data. First, data cleaning is performed, using statistical methods to identify and remove outlier data points. For example, for temperature data, if the temperature at a certain moment deviates from the average temperature of adjacent moments in the same area by more than three standard deviations, it is considered outlier and removed. For flow rate data, negative values ​​or values ​​exceeding the sensor's range by several times are also considered outliers and removed. Second, missing data is handled using appropriate imputation methods. For continuous data, such as a small number of missing values ​​in a temperature sequence, linear interpolation is used to fill in the missing values ​​based on the trend of adjacent data points. For discrete data, such as occasionally missing records of vehicle load conditions, the mode is selected based on the frequency of occurrence of that condition in historical data. Finally, key features are carefully selected and extracted from the preprocessed data and used as input to the neural network. These features include ambient temperature, ambient humidity, solar radiation intensity, vehicle speed, temperature difference between the inside and outside of the glass, sheet metal temperature gradient, air layer heat flux density, and interior surface temperature. These comprehensively reflect the internal and external factors affecting the thermal environment of the cab and the thermal state of various components.

[0043] In one embodiment, considering the complexity of the thermal management system and the input-output relationship, a multilayer perceptron (MLP) is selected as the basic neural network architecture to construct a network structure including an input layer, multiple hidden layers, and an output layer. The number of nodes in the input layer corresponds to the number of key features extracted, ensuring that all influencing factors are fully received. The hidden layer has 2-4 layers, with the number of neurons in each layer optimized experimentally, initially set within the range of 30-100 neurons. ReLU is used as the activation function to introduce nonlinearity and enhance the model's ability to express complex heat transfer relationships. The output layer is designed to predict the temperature and humidity values ​​of multiple key areas inside the driver's cabin, such as the temperature of the driver's head, chest, and feet, as well as the average humidity inside the cabin, thereby intuitively assessing thermal comfort and air quality.

[0044] In one embodiment, the preprocessed dataset is divided into training, validation, and test sets at a ratio of 70%, 15%, and 15%, respectively. Next, the constructed neural network model is trained using the training set, employing stochastic gradient descent (SGD) and its variants, such as the Adam optimization algorithm, combined with mean squared error (MSE) as the loss function. The network weights are continuously adjusted using backpropagation to gradually reduce the model's loss value on the training set. Initially, a relatively large learning rate, such as 0.01, is set to accelerate convergence. As training progresses, when the validation set loss value no longer decreases after 3-5 consecutive iterations, a learning rate decay strategy is applied, reducing the learning rate by 50% to prevent the model from getting trapped in local optima and thus ensuring good generalization ability.

[0045] In one embodiment, after a validated model is deployed to the thermal management control system of a light truck, the model can receive data collected by sensors in real time as input. Its output, predicted cab temperature and humidity values, can guide the coordinated control of thermal management components such as the air conditioning system, ventilation system, and battery cooling system. For example, based on the predicted temperature rise trend, the air conditioning cooling power can be adjusted in advance; based on the humidity prediction value, the dehumidification function can be turned on or off in a timely manner, thereby achieving precise thermal comfort adjustment and effectively reducing energy consumption. Furthermore, as the light truck continues to be used, new data needs to be collected periodically, and the update cycle should be determined according to the vehicle's usage frequency and the degree of change in operating conditions, generally every 1-2 months. The deployed model is then updated and trained using the new data. When incorporating new data into the training set, it needs to be processed according to the previously described data preprocessing steps to ensure that the model always adapts to the actual operating conditions of the vehicle and maintains excellent predictive performance.

[0046] like Figure 2 As shown in the illustration, this application also provides a thermal management device for a light truck cab, comprising:

[0047] At least one processor; and,

[0048] A memory that is communicatively connected to at least one processor; wherein,

[0049] The memory stores instructions that can be executed by at least one processor to enable a thermal management device for a light truck cab to perform the method as described in any of the embodiments described above.

[0050] This application also provides a non-volatile computer storage medium storing computer-executable instructions, which are configured as described in any of the above embodiments.

[0051] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0052] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0053] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0054] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0055] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0056] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0057] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0058] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0059] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0060] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0061] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0062] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0063] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0064] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0065] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A thermal management method for a light truck cab, characterized in that, include: Acquire key component data and driver data for the cab. The key components include glass, outer sheet metal, air layer, felt layer, and interior materials. Determine a set of equations for the corresponding key components based on the key component data and driver data. Then, determine multiple layered thermal resistance network models based on the set of equations. The internal and external environmental parameters of the cab are obtained. Based on the internal and external environmental parameters and the multiple hierarchical thermal resistance network models, a pre-set neural network model is trained and integrated to obtain a thermal resistance network model. The heat flow of the cab is then predicted using the thermal resistance network model. Acquire key component data of the cab and driver data, specifically including: Collect data on the light transmittance, thermal conductivity, specific heat capacity, geometry, and area of ​​the glass components; Record the thickness, thermal conductivity, specific heat capacity, geometry, and coverage area of ​​the outer sheet metal layer; Measure the thickness of the air layer, air velocity, and heat exchange coefficient; Obtain data on the thermal insulation performance, moisture absorption characteristics, and heat capacity of the felt layer; Collect data on the surface emissivity, thermal resistance, and geometry of interior materials; Determine the thermophysiological model corresponding to the driver, and obtain node thermal data based on multiple nodes of the thermophysiological model.

2. The method according to claim 1, characterized in that, Based on the key component data and driver data, a set of equations is determined for the corresponding key components, specifically including: The unsteady-state heat conduction equation is determined based on the data of the glass component. The unsteady-state heat conduction equation includes light transmittance, solar radiation absorptivity, and convective heat transfer between the inner and outer surfaces. The thermal resistance network model is determined based on the data of the outer sheet metal, and the thermal balance equation is determined based on each node of the thermal resistance network model. The equivalent thermal resistance and heat capacity models are determined based on the data of the air layer and the felt layer, and the heat transfer equations are determined based on the equivalent thermal resistance and heat capacity models. The dynamic thermal response equation and the contact heat transfer equation are determined based on the data of the interior materials and the nodal heat data. The thermophysiological model, the unsteady-state heat conduction equation, the heat balance equation, the heat transfer equation, the dynamic thermal response equation, and the contact heat transfer equation are integrated to determine the set of equations.

3. The method according to claim 1, characterized in that, Obtain parameters of the internal and external environment of the cab, specifically including: Real-time monitoring of ambient temperature, relative humidity, and solar radiation intensity outside the cab; The vehicle's speed is recorded to determine the air convection coefficient based on the vehicle's speed. Temperature and humidity data from multiple locations inside the driver's cab are collected to obtain the operating status of the air conditioning system, which includes the set temperature, fan speed setting, and air outlet temperature.

4. The method according to claim 1, characterized in that, The pre-set neural network model is trained and integrated based on the internal and external environmental parameters and the multiple hierarchical thermal resistance network models, specifically including: The internal and external environmental parameters are preprocessed to determine key features, including ambient temperature, ambient humidity, solar radiation intensity, vehicle speed, temperature difference between inside and outside the glass, sheet metal temperature gradient, air layer heat flux density, and interior surface temperature. A pre-defined multilayer perceptron is determined, and the network structure of the multilayer perceptron is determined, wherein the network structure includes an input layer, multiple hidden layers, and an output layer; The number of input layer nodes is equal to the number of key features; The number of neurons in each layer of the hidden layer is set to a predetermined initial range, and ReLU is used as the activation function; The output layer is designed to display temperature and humidity values ​​for multiple areas inside the driver's cab.

5. The method according to claim 1, characterized in that, The prediction of heat flow in the cab using the thermal resistance network model specifically includes: The cab is divided into multiple node systems, which include thermal resistance series and thermal resistance parallel. Heat transfer is calculated using a pre-set set of formulas to obtain preliminary prediction results. Input the current operating condition parameters into the thermal resistance network model to output the prediction residual. Adjust the key parameters of the thermal resistance network model based on the prediction residual. The key parameters include the equivalent thermal conductivity, convective heat transfer coefficient, and radiative heat transfer parameter. The temperature prediction results were obtained by recalculating using the adjusted thermal resistance network model.

6. The method according to claim 1, characterized in that, After predicting the heat flow in the cab using the thermal resistance network model, the method further includes: The prediction results of the thermal resistance network model are verified, and the verified thermal resistance network model is deployed into the thermal management control system of the light truck to predict the temperature and humidity values ​​of the cab through the thermal management control system, so as to coordinate and control the thermal management components based on the temperature and humidity values. The thermal management components include the air conditioning subsystem, the ventilation subsystem, and the battery cooling subsystem.

7. The method according to claim 1, characterized in that, The method further includes: Regularly collect new data on light trucks during actual operation, including thermal management data under different seasons, regional climates, various driving conditions, and different load conditions; The thermal resistance network model deployed in the thermal management and control system is updated and trained based on the new data to adapt the thermal resistance network model to the actual operating conditions of the vehicle.

8. A thermal management device for a light truck cab, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor to enable the thermal management device for a light truck cab to perform the method as described in any one of claims 1-7.

9. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are configured to be the method as described in any one of claims 1-7.

Images