A compression cooling garment and adaptive control method thereof

Abstract

This invention discloses a compression-type cooling suit and its adaptive control method. The method includes: training a pre-constructed artificial neural network model of human body load based on a database of the relationship between physiological state parameters, basal metabolic parameters, environmental parameters, and human body load, to obtain a trained artificial neural network model of human body load; acquiring real-time physiological state parameters, basal metabolic parameters, and environmental parameters, and inputting them into the trained artificial neural network model of human body load to obtain real-time human body load; inputting the real-time human body load into a pre-constructed compressor speed calculation model of the compression-type cooling suit to calculate the compressor speed for adaptive control, and generating instructions to perform variable frequency control on the compressor. This invention can achieve dynamic and accurate perception of the real heat demand of the human body, and achieve precise matching between cooling output and human body load demand, thereby improving thermal comfort, optimizing system energy efficiency, and enhancing high-temperature reliability.

Description

Technical Field

[0001] This invention belongs to the technical field of wearable cooling devices, specifically relating to a compression cooling suit and its adaptive control method. Background Technology

[0002] In outdoor operations involving high temperatures and humidity, such as power line inspections, fire rescues, and work in enclosed spaces (e.g., inside tank compartments), ambient temperatures can reach over 40°C. Workers operating under these temperatures are highly susceptible to heatstroke, heat fatigue, and even heat exhaustion. Current common on-site cooling methods, such as using ice packs or ordinary fan-type cooling suits, often suffer from problems such as short cooling duration, inaccurate temperature control, and ineffective humidity regulation.

[0003] Currently, the mainstream types of cooling clothing on the market mainly include phase change material (PCM) type, gas or liquid circulation type, and compressor refrigeration type. PCM cooling clothing relies on the phase change of the material to absorb heat, but its effective cooling time is limited and its temperature control accuracy is low. Gas or liquid circulation cooling clothing is mostly passive cooling, making it difficult to dynamically adjust according to the actual changes in the body's workload. While compressor refrigeration cooling clothing has high cooling efficiency, its control strategy usually relies solely on ambient temperature or a fixed program, lacking the ability to respond to the real-time thermal state of the human body. For example, traditional compression cooling clothing often adjusts the compressor by setting a preset temperature threshold, which can easily lead to over-cooling or under-cooling, and it also fails to fully consider individual differences in age, weight, gender, and dynamic work intensity.

[0004] Human body load is a complex parameter that dynamically changes with metabolic rate, ambient temperature and humidity, and physiological states such as heart rate and skin temperature. However, existing cooling clothing control systems use simple methods to adjust cooling capacity, which generally result in a mismatch between cooling capacity adjustment and the individual's cooling needs, leading to low overall energy efficiency and poor comfort.

[0005] Therefore, there is an urgent need for an adaptive control method that can integrate multi-source information, predict human body load in real time, and adjust compressor speed precisely accordingly to achieve efficient and personalized body temperature management, thereby improving the safety and work efficiency of workers in high-temperature and high-humidity environments. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a compression cooling suit and its adaptive control method. By using multi-parameter measurement and human body load artificial neural network model, the invention achieves dynamic and accurate perception of the real heat demand of the human body. By using the compressor speed calculation model of the compression cooling suit, the invention achieves adaptive control of the compressor speed and ultimately achieves accurate matching between cooling capacity output and human body load demand, thereby improving thermal comfort, optimizing system energy efficiency and enhancing high-temperature reliability.

[0007] This invention provides the following technical solution:

[0008] Firstly, an adaptive control method for a compression-type cooling suit is provided, comprising:

[0009] Based on a database of the relationship between physiological state parameters, basal metabolic parameters, environmental parameters and human body load, a pre-constructed artificial neural network model of human body load is trained to obtain a trained artificial neural network model of human body load.

[0010] Real-time physiological state parameters, basal metabolic parameters, and environmental parameters are acquired and input into a trained human load artificial neural network model to obtain real-time human load.

[0011] The real-time human load is input into a pre-built compressor speed calculation model for the compression cooling suit. The compressor speed for adaptive control is calculated, and instructions are generated to perform variable frequency control on the compressor.

[0012] Furthermore, the physiological parameters include heart rate and skin temperature; the basal metabolic parameters include height, weight, age, and sex; and the environmental parameters include ambient temperature and ambient humidity.

[0013] Furthermore, the method for constructing the database of relationships between physiological state parameters, basal metabolic parameters, environmental parameters, and human body load includes:

[0014] Acquire existing physiological state parameters, basal metabolic parameters, environmental parameters, and human load data;

[0015] Dimensionless physiological state parameters, basal metabolic parameters, and environmental parameters are obtained by performing dimensionless processing on the physiological state parameters, basal metabolic parameters, and environmental parameters.

[0016] By matching and combining dimensionless physiological state parameters, basal metabolic parameters, environmental parameters and human load data, a database of the relationship between physiological state parameters, basal metabolic parameters, environmental parameters and human load is obtained.

[0017] Furthermore, the physiological state parameters are dimensionless using the following formula:

[0018] ;

[0019] ;

[0020] in, For dimensionless heart rate, Let τ be the heart rate at time τ. Resting heart rate Dimensionless skin temperature Let τ be the skin temperature at time τ. Core temperature;

[0021] The basal metabolic parameters were dimensionlessly processed using the following formula:

[0022] ;

[0023] ;

[0024] ;

[0025] in, H is a dimensionless height. p H0 represents the height of the person, and H0 represents the standard height. For dimensionless body weight, m p m0 represents the average person's weight, and m0 represents the ideal weight. Y is a dimensionless age. p Y0 represents the standard age of the personnel.

[0026] The environmental parameters are dimensionless using the following formula:

[0027] ;

[0028] ;

[0029] in, The ambient temperature is a dimensionless quantity. Let τ be the ambient temperature. Standard ambient temperature, The humidity is a dimensionless environmental value. Let τ be the relative humidity at time τ.

[0030] Furthermore, the method for constructing and training the artificial neural network model for human body load includes:

[0031] The database of relationships between physiological state parameters, basal metabolic parameters, environmental parameters and human body load is divided into a training set and a reference set;

[0032] A multilayer perceptron model was constructed as the basic network structure. The number of layers and the number of neurons in each layer were selected to obtain an artificial neural network model for human load.

[0033] The human load artificial neural network model was trained using a training set and validated using a reference set. The number of layers and neurons in each layer of the network were adjusted until the model met the accuracy requirements, resulting in a well-trained human load artificial neural network model.

[0034] Furthermore, the multilayer perceptron model includes an input layer, a hidden layer, and an output layer. The input layer includes 8 parameters, the hidden layer has 1 to 3 layers, and the output layer includes 1 parameter.

[0035] Furthermore, the method for constructing the compressor speed calculation model for the compression-type cooling suit includes:

[0036] A semi-empirical formula is established to establish the relationship between compressor speed, chilled water tank temperature, ambient temperature, and cooling capacity of a compression refrigeration unit:

[0037] (1);

[0038] In the formula, Q1 represents the cooling capacity of the compressor refrigeration unit, M represents the mass flow rate of the refrigerant, q represents the cooling capacity per unit mass flow rate of the refrigerant, f(N) represents a function of the compressor speed N, and q(T) represents the cooling capacity per unit mass flow rate of the refrigerant. e , T w () indicates ambient temperature T e and chilled water tank temperature T w The functions; f(N) and q(T) e , T w ) is the fitting function, obtained by fitting data after performance experiments on the compression cooling suit;

[0039] Establish a set of equations relating the cooling capacity of a compression refrigeration system to the heat transfer between the chilled water piping and the human body:

[0040] (2);

[0041] (3);

[0042] (4);

[0043] (5);

[0044] (6);

[0045] In the formula, c represents the specific heat capacity of chilled water, M1 represents the chilled water flow rate, and T... b Indicates the chilled water return temperature. Q1 represents the average temperature of the chilled water piping during the heat exchange process; Q2 represents the heat transfer between the chilled water piping and the environment; Q3 represents the heat transfer between the chilled water piping and the human body; U2 represents the heat transfer per unit temperature difference between the chilled water piping and the environment; U3 represents the heat transfer per unit temperature difference between the chilled water piping and the human body; T represents the heat transfer per unit temperature difference between the chilled water piping and the human body. s Indicates skin temperature;

[0046] By combining equations (1) to (6), we obtain the functional relationship between compressor speed N and the heat transfer Q3 between the chilled water pipeline and the human body:

[0047] (7).

[0048] Furthermore, the step of inputting real-time human load into a pre-built compressor speed calculation model for a compression-type cooling suit to calculate the compressor speed for adaptive control includes:

[0049] Let the heat transfer between the chilled water pipe and the human body, Q3, be equal to the real-time human body load. The compressor speed N is calculated by the functional relationship between the compressor speed N and the heat transfer between the chilled water pipe and the human body, Q3.

[0050] Secondly, a compression cooling suit is provided, including a cooling suit, a compression refrigeration unit, sensors, and a controller;

[0051] The cooling suit has chilled water pipes coiled inside, and the outlet end of the chilled water pipes is connected to a chilled water tank.

[0052] The compression refrigeration unit includes a compressor, a condenser, a throttle valve, and an evaporator connected in sequence, wherein the evaporator is used to cool the liquid in the chilled water tank;

[0053] The sensor is used to collect heart rate, skin temperature, ambient temperature, ambient humidity, and chilled water tank temperature in real time;

[0054] The controller is connected to the compression refrigeration unit and the sensor respectively, and is used to execute the adaptive control method for the compression cooling suit as described in any one of the first aspects.

[0055] Furthermore, the chilled water tank is connected to the evaporator via a circulating water pump, and the evaporator is connected to the inlet end of the chilled water pipeline via an inlet pipe.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] This invention obtains real-time human body load through multi-parameter measurement and calculation using an artificial neural network model of human body load, achieving dynamic and accurate perception of the human body's actual thermal needs. Furthermore, by calculating the compressor speed using a calculation model of real-time human body load and compressor speed of a compression cooling suit, adaptive control of the compressor speed is achieved, ultimately realizing precise matching between cooling capacity output and human body load needs. This improves thermal comfort, optimizes system energy efficiency, and enhances high-temperature reliability, helping to improve the safety and work efficiency of personnel working in high-temperature and high-humidity environments. Attached Figure Description

[0058] Figure 1This is a flowchart illustrating the adaptive control method for the compression cooling suit in an embodiment of the present invention.

[0059] Figure 2 This is a schematic diagram of the process of constructing and training an artificial neural network model for human load in an embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the structure of the artificial neural network model for human body load in an embodiment of the present invention;

[0061] Figure 4 This is a schematic diagram of the process for real-time calculation and control of compressor speed in an embodiment of the present invention;

[0062] Figure 5 This is a schematic diagram of the structure of the compression cooling suit in an embodiment of the present invention;

[0063] The following are marked in the diagram: 1. Cooling suit; 11. Neck; 12. Back; 13. Left chest; 14. Right chest; 21. Compressor; 22. Condenser; 23. Throttler; 24. Evaporator; 25. Fan; 31. Chilled water piping; 32. Chilled water tank; 33. Circulating water pump. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0065] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0066] Example 1

[0067] like Figure 1 As shown in the figure, this embodiment provides an adaptive control method for a compression cooling suit, the steps of which are as follows:

[0068] Step 1: Construct a database of the relationships between physiological state parameters, basal metabolic parameters, environmental parameters, and human body load.

[0069] Physiological parameters include heart rate (HR) and skin temperature (T). s The basal metabolic parameters include height (H), weight (m), age (Y), and sex (G); the environmental parameters include ambient temperature (T). e and ambient humidity .

[0070] Step 1.1: Obtain existing physiological state parameters, basal metabolic parameters, environmental parameters, and human body load data.

[0071] Specifically, existing data can come from online databases or laboratory / enthalpy difference chamber measurement data.

[0072] Step 1.2: Perform dimensionless processing on the physiological state parameters, basal metabolic parameters, and environmental parameters to obtain dimensionless physiological state parameters, basal metabolic parameters, and environmental parameters.

[0073] The physiological state parameters are dimensionless using the following formula:

[0074] ;

[0075] ;

[0076] in, The heart rate is dimensionless. Let τ be the heart rate in bpm. This is the resting heart rate, a constant, typically taken as 60 bpm; Dimensionless skin temperature Let τ be the skin temperature at time τ, in K. The core temperature is , which is a constant, typically taken as 309.65 K.

[0077] The basal metabolic parameters were dimensionlessly processed using the following formula:

[0078] ;

[0079] ;

[0080] ;

[0081] ;

[0082] in, Height without dimensions; H p H0 represents the height of the person, in meters; H0 is the standard height, generally taken as 1.7 meters. m is a dimensionless body weight. p The value is the person's weight, in kg; m0 is the standard weight, generally taken as 65 kg. Age is dimensionless; Y p Y0 represents the age of the personnel, in years; Y0 is the standard age, generally taken as 20 years old. Gender is dimensionless.

[0083] The environmental parameters are dimensionless using the following formula:

[0084] ;

[0085] ;

[0086] in, The ambient temperature is a dimensionless quantity. Let τ be the ambient temperature at time τ, in K. The standard ambient temperature is a constant, typically taken as 308.15K. Dimensionless ambient humidity; Let τ be the relative humidity at time τ, and % represents the percentage of water vapor pressure in the air compared to the saturated water vapor pressure at the same temperature.

[0087] Step 1.3: Match and combine dimensionless physiological state parameters, basal metabolic parameters, environmental parameters and human body load data to obtain a database of the relationship between physiological state parameters, basal metabolic parameters, environmental parameters and human body load.

[0088] Step 2: Based on a database of relationships between physiological state parameters, basal metabolic parameters, environmental parameters, and human body load, train the pre-constructed artificial neural network model of human body load to obtain a trained artificial neural network model of human body load, such as... Figure 2 As shown.

[0089] Step 2.1: Divide the database of relationships between physiological state parameters, basal metabolic parameters, environmental parameters and human body load into training set and reference set.

[0090] The database can be divided into training and reference sets using a sliding window method or cross-validation. Specifically, the database is divided into an 80% training set and a 20% reference set, and both the training and reference sets need to cover diverse operating conditions, such as different physiological parameters, basal metabolic parameters, and environmental parameters.

[0091] Step 2.2: Construct a multilayer perceptron model as the basic network structure, select the number of layers and the number of neurons in each layer to obtain the human load artificial neural network model.

[0092] The Multilayer Perceptron (MLP) model can describe the complex nonlinear relationship between human body load and various influencing factors. The MLP model consists of an input layer, hidden layers, and an output layer.

[0093] In this embodiment, the input layer has 8 parameters, which are the 8 dimensionless parameters from step 1.2. The hidden layer design follows the principle of hierarchical feature extraction, with 1 to 3 hidden layers. It learns the nonlinear interaction relationship between parameters, and the ReLU activation function is used to alleviate gradient vanishing and improve training efficiency. The output layer has 1 parameter, which is the human body load.

[0094] Step 2.3: Train the human load artificial neural network model using the training set and validate the model using the reference set. Adjust the number of layers and the number of neurons in each layer of the network until the model meets the accuracy requirements, and obtain the trained human load artificial neural network model.

[0095] Configure at least three different artificial neural network models and train them using the training set data, employing an early stopping strategy to avoid overfitting. Validate the prediction error of personnel load using the reference set data. Model performance is measured using the root mean square error (RMSE) and coefficient of determination (R²). 2 Quantify and set an error threshold, such as setting RMSE below 5%, R 2 A value greater than 0.9 is used to ensure prediction accuracy. The number of hidden layers, the number of hidden layers in the three models, and the activation function are adjusted to ensure their RMSE and R-value are within acceptable limits. 2 All errors were less than the error threshold. Comparing the error magnitudes of the three models, the artificial neural network model with the smallest error was selected as the artificial neural network model for human load assessment.

[0096] Figure 3 This is an artificial neural network model structure, employing a three-layer architecture. The input layer, located on the left, has eight nodes, including dimensionless parameters such as heart rate, skin temperature, height, weight, age, gender, ambient temperature, and ambient humidity. The hidden layer, a single layer located in the middle, has 7-16 nodes and receives data from the input layer. Through activation function calculations, it automatically learns the complex relationship between the input parameters and the human body's workload. The output layer, located on the right, has only one node representing the human body's workload and selects an appropriate activation function, such as linear activation.

[0097] Step 3: Obtain real-time physiological state parameters, basal metabolic parameters, and environmental parameters, and input them into the trained human load artificial neural network model to obtain the real-time human load Q. body .

[0098] Specifically, basal metabolic parameters (height, weight, age, gender) of staff are collected through questionnaires, and physiological parameters (heart rate, skin temperature) and environmental parameters (ambient temperature and humidity) are measured through sensors.

[0099] Step 4: Input the real-time human load into the pre-built compressor speed calculation model of the compression-type cooling suit, calculate the compressor speed for adaptive control, and generate instructions to perform variable frequency control on the compressor, such as... Figure 4 As shown.

[0100] The compression cooling suit in this embodiment includes a cooling suit and a compression refrigeration unit. Chilled water pipes are coiled inside the cooling suit. The water in the chilled water pipes is supplied by a chilled water tank and cooled by the compression refrigeration unit.

[0101] Step 4.1: Construct a compressor speed calculation model for the compression-type cooling suit.

[0102] Step 4.1.1: Establish a semi-empirical formula for the relationship between compressor speed, chilled water tank temperature, ambient temperature, and cooling capacity of the compression refrigeration unit:

[0103] (1);

[0104] In the formula, Q1 represents the cooling capacity of the compressor refrigeration unit, W; M represents the mass flow rate of the refrigerant, kg / s; q represents the cooling capacity per unit mass flow rate of the refrigerant, J / kg; f(N) represents a function of the compressor speed N, q(T) e , T w () indicates ambient temperature T e and chilled water tank temperature T w The functions; f(N) and q(T) e , T w The fitting function is obtained by fitting data from performance experiments on compression cooling suits. It can be fitted using linear, polynomial, or nonlinear fitting formulas. The fitting R0 value is given. 2 >0.9.

[0105] Step 4.1.2: Establish a set of equations relating the cooling capacity of the compression refrigeration system to the heat transfer between the chilled water pipes and the human body:

[0106] During steady-state cooling, the cooling capacity of the chilled water is equal to the cooling capacity of the compressor chiller, satisfying the following formula:

[0107] (2);

[0108] In the formula, c represents the specific heat capacity of chilled water, J / (kg·K); M1 represents the chilled water flow rate, kg / s; T b Indicates the chilled water return temperature, in K;

[0109] The average temperature of the chilled water piping during the heat exchange process is calculated using the following formula:

[0110] (3);

[0111] In the formula, This represents the average temperature, in K, of the chilled water piping during the heat exchange process.

[0112] The heat transfer between the chilled water piping and the environment, and the heat transfer between the chilled water piping and the human body, are calculated using the following formulas:

[0113] (4);

[0114] (5);

[0115] (6);

[0116] In the formula, Q2 represents the heat transfer between the chilled water pipes and the environment, Q3 represents the heat transfer between the chilled water pipes and the human body, U2 represents the heat transfer per unit temperature difference when the chilled water pipes transfer heat to the environment (W / K); U3 represents the heat transfer per unit temperature difference when the chilled water pipes transfer heat to the human body (W / K); T s Indicates skin temperature.

[0117] In the aforementioned formulas (1) to (6), T s T w T e The parameters are measured at the measuring points; U2, U3, c, and M1 are known constants obtained through prior experiments; T b Q1, Q2, Q3 N are the unknowns in the six formulas.

[0118] Step 4.1.3: Solve equations (1) to (6) simultaneously to obtain the functional relationship between compressor speed N and the heat transfer Q3 between the chilled water pipeline and the human body:

[0119] ;

[0120] Right now: (7).

[0121] Step 4.2: Make the heat transfer between the chilled water pipes and the human body Q3 equal to the real-time human body load Q. body The compressor speed N is calculated by the functional relationship between the compressor speed N and the heat transfer Q3 between the chilled water pipeline and the human body.

[0122] Step 4.3: Output a control command to adjust the compressor speed to N, and realize adaptive control of the compression cooling suit by responding to the control command through electronic control.

[0123] Example 2

[0124] like Figure 5 As shown, this embodiment provides a compression cooling suit, including a cooling suit 1, a compression refrigeration unit, a sensor, and a controller.

[0125] The compression refrigeration unit includes a compressor 21, a condenser 22, a throttle 23 and an evaporator 24 connected in sequence, with refrigerant circulating within it, and the condenser 22 dissipating heat through a fan 25.

[0126] The cooling suit 1 includes a neck area 11, a back area 12, a left chest area 13, and a right chest area 14. Each part has a chilled water pipe 31 coiled around its inner side. The outlet end of the chilled water pipe 31 is connected to a chilled water tank 32. The chilled water tank 32 is connected to the inlet end of an evaporator 24 via a circulating water pump 33. The outlet end of the evaporator 24 is connected to the inlet end of the chilled water pipe 31 via a water inlet pipe. During operation, water from the chilled water tank 32 is pumped into the evaporator 24 by the circulating water pump 33 for cooling, and then flows into the chilled water pipe 31, forming a loop. As the chilled water flows through the chilled water pipe 31, it carries away the body's heat.

[0127] The sensor is used to collect real-time data on the heart rate, skin temperature, ambient temperature, ambient humidity, and chilled water tank temperature of the person wearing the cooling suit.

[0128] The controller is connected to the compression refrigeration unit, the sensor, and the circulating water pump, respectively, and is used to execute the adaptive control method for the compression cooling suit described in Example 1.

[0129] Example 3

[0130] This embodiment provides a compression-type cooling suit as described in Embodiment 2, wherein the rated power of the compression refrigeration unit is 90W, the rated cooling capacity is 180W, the rated speed of the compressor is 3000rpm, and the rated operating conditions are an ambient temperature of 40℃ and a chilled water tank temperature of 25℃, and it can cool in an ambient temperature of 35~45℃.

[0131] This embodiment uses the method in Embodiment 1 to adaptively control the compression cooling suit, as detailed below.

[0132] Three schemes are designed to construct artificial neural network models according to the method in step 2, namely:

[0133] Option 1: The input layer has 8 parameters; the output layer has 1 parameter and uses the Linear activation function; the hidden layer has 1 layer with 8 nodes and uses the ReLU activation function. This option has the same number of nodes as the input layer, is computationally efficient, and uses ReLU to alleviate gradient vanishing, making it suitable for rapid validation.

[0134] Option 2: The input layer has a total of 8 parameters; the output layer has 1 parameter, with the activation function Linear; the hidden layer is a single layer with 16 nodes, with the activation function Tanh. This option expands the feature dimension, and Tanh provides smooth non-linearity.

[0135] Option 3: The input layer has a total of 8 parameters; the output layer has 1 parameter, with the activation function Linear; the hidden layers have 2 layers, with the first layer having 12 nodes and the second layer having 4 nodes, both with the activation function ReLU. This option is a classic "funnel" structure, progressively compressing and refining information, with a stable structure and good generalization ability.

[0136] The three schemes were trained using the training set and verified using the reference set. Scheme 3 was found to have the smallest calculation error, so it was selected as the artificial neural network model for human load in subsequent calculations.

[0137] The compressor speed calculation model for the compression-type cooling suit is constructed according to the method in step 4.1, as follows:

[0138] First, establish f(N) and q(T). e , T w The fitting function for ) is, in this embodiment, f(N) = N. Then, according to formula (1), the fitting equation is obtained, and the fitting R is obtained. 2 The value is 0.96, as detailed below:

[0139] .

[0140] The chilled water flow rate M1 can be determined using methods such as volumetric method, gravimetric method, volumetric flow meter, and mass flow meter. In this embodiment, the chilled water tank temperature T is measured during a preliminary experiment. w =15.6℃, chilled water return temperature T b =20.1℃, chilled water flow rate M1=720ml / min=0.012kg / s, chilled water specific heat capacity c is 4180J / kg·℃.

[0141] The cooling capacity of the compressor chiller is calculated as Q1 = 225.72W according to formula (2).

[0142] The average temperature of the chilled water pipeline during the heat exchange process is calculated according to formula (3). =17.85℃.

[0143] The ambient temperature T was measured in the experiment. e The temperature was 41.0℃, and the skin temperature T on the body surface was... s The temperature was 34.47℃, the average temperature of the outer surface of the clothing was 26.2℃, and the area of ​​the clothing was 0.72m². 2 The heat transfer coefficient between the outer surface of the clothing and the environment is 8.7 W / (m²). 2 (℃), the heat transfer between the clothing and the environment is equal to the heat transfer from the chilled water pipe to the environment, therefore Q2 = 0.72 × 8.7 × (41.0 - 26.2) = 92.7 W.

[0144] Substituting Q2=92.7W into formula (4), we get U2=4.0 W / K;

[0145] Substituting Q2=92.7W into formula (6), we get Q3=133.02W;

[0146] Substituting Q3=133.02W into formula (5), we get U3=8.0W / K.

[0147] The functional relationship between N and Q3 can be obtained from formula (7) and the aforementioned preliminary experimental results:

[0148] ;

[0149] In actual use, following steps 3 and 4, the compressor speed is calculated and adjusted every minute, as shown in operating conditions 1 and 2 below.

[0150] Operating Condition 1: Real-time physiological state parameters, basal metabolic parameters, and environmental parameters were collected and processed to be dimensionless. The results are shown in Table 1 below.

[0151] Table 1. Physiological state parameters, basal metabolic parameters, and environmental parameters

[0152]

[0153] The dimensionless real-time physiological state parameters, basal metabolic parameters, and environmental parameters are input into the aforementioned trained artificial neural network model of human load to obtain the real-time human load Q. body =81W.

[0154] Substituting Q3=81W into the functional relationship between N and Q3 above, the compressor speed is calculated to be 1389 rpm, which is 53.7% energy saving compared to the compressor running at full power.

[0155] Condition 2: Real-time physiological state parameters, basal metabolic parameters, and environmental parameters were collected and processed to be dimensionless. The results are shown in Table 2 below.

[0156] Table 2. Physiological state parameters, basal metabolic parameters, and environmental parameters.

[0157]

[0158] The dimensionless real-time physiological state parameters, basal metabolic parameters, and environmental parameters are input into the aforementioned trained artificial neural network model of human load to obtain the real-time human load Q. body =128W.

[0159] Substituting Q3=128W into the functional relationship between N and Q3 above, the compressor speed is calculated to be 2861 rpm, which is an energy saving rate of 4.6% compared to the compressor running at full power.

[0160] If effective cooling is required at higher temperatures (>45°C), a compression refrigeration cycle needs to be designed to match the cooling characteristics at that temperature. This implementation example represents only one case of a compression-type cooling garment and its adaptive control method, not all embodiments. The method is equally applicable to the parameters of all compression-type cooling garments based on the same principle.

[0161] 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.

[0162] 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.

[0163] 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.

[0164] 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.

[0165] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An adaptive control method for a compression-type cooling suit, characterized in that, include: Based on a database of the relationship between physiological state parameters, basal metabolic parameters, environmental parameters and human body load, a pre-constructed artificial neural network model of human body load is trained to obtain a trained artificial neural network model of human body load. Real-time physiological state parameters, basal metabolic parameters, and environmental parameters are acquired and input into a trained human load artificial neural network model to obtain real-time human load. The real-time human load is input into a pre-built compressor speed calculation model for the compression cooling suit. The compressor speed for adaptive control is calculated, and instructions are generated to perform variable frequency control on the compressor.

2. The adaptive control method for the compression cooling suit according to claim 1, characterized in that, The physiological parameters include heart rate and skin temperature; the basal metabolic parameters include height, weight, age, and sex; and the environmental parameters include ambient temperature and ambient humidity.

3. The adaptive control method for the compression cooling suit according to claim 1, characterized in that, The method for constructing the database of relationships between physiological state parameters, basal metabolic parameters, environmental parameters, and human body load includes: Acquire existing physiological state parameters, basal metabolic parameters, environmental parameters, and human load data; Dimensionless physiological state parameters, basal metabolic parameters, and environmental parameters are obtained by performing dimensionless processing on the physiological state parameters, basal metabolic parameters, and environmental parameters. By matching and combining dimensionless physiological state parameters, basal metabolic parameters, environmental parameters and human load data, a database of the relationship between physiological state parameters, basal metabolic parameters, environmental parameters and human load is obtained.

4. The adaptive control method for the compression cooling suit according to claim 3, characterized in that, The physiological state parameters are dimensionless using the following formula: ； ； in, For dimensionless heart rate, Let τ be the heart rate at time τ. Resting heart rate Dimensionless skin temperature Let τ be the skin temperature at time τ. Core temperature; The basal metabolic parameters were dimensionlessly processed using the following formula: ； ； ； in, H is a dimensionless height. p H0 represents the height of the person, and H0 represents the standard height. For dimensionless body weight, m p m0 represents the average person's weight, and m0 represents the ideal weight. Y is a dimensionless age. p Y0 represents the standard age of the personnel. The environmental parameters are dimensionless using the following formula: ； ； in, The ambient temperature is a dimensionless quantity. Let τ be the ambient temperature. Standard ambient temperature, The humidity is a dimensionless environmental value. Let τ be the relative humidity at time τ.

5. The adaptive control method for the compression cooling suit according to claim 1, characterized in that, The construction and training methods of the artificial neural network model for human body load include: The database of relationships between physiological state parameters, basal metabolic parameters, environmental parameters and human body load is divided into a training set and a reference set; A multilayer perceptron model was constructed as the basic network structure. The number of layers and the number of neurons in each layer were selected to obtain an artificial neural network model for human load. The human load artificial neural network model was trained using a training set and validated using a reference set. The number of layers and neurons in each layer of the network were adjusted until the model met the accuracy requirements, resulting in a well-trained human load artificial neural network model.

6. The adaptive control method for the compression cooling suit according to claim 5, characterized in that, The multilayer perceptron model includes an input layer, a hidden layer, and an output layer. The input layer includes 8 parameters, the hidden layer has 1 to 3 layers, and the output layer includes 1 parameter.

7. The adaptive control method for the compression cooling suit according to claim 1, characterized in that, The method for constructing the compressor speed calculation model for the compression-type cooling suit includes: A semi-empirical formula is established to establish the relationship between compressor speed, chilled water tank temperature, ambient temperature, and the cooling capacity of a compression refrigeration unit: （1）； In the formula, Q1 represents the cooling capacity of the compressor refrigeration unit, M represents the mass flow rate of the refrigerant, q represents the cooling capacity per unit mass flow rate of the refrigerant, f(N) represents a function of the compressor speed N, and q(T) represents the cooling capacity per unit mass flow rate of the refrigerant. e , T w () indicates ambient temperature T e and chilled water tank temperature T w The functions; f(N) and q(T) e , T w ) is the fitting function, obtained by fitting data after performance experiments on the compression cooling suit; Establish a set of equations relating the cooling capacity of a compression refrigeration system to the heat transfer between the chilled water piping and the human body: （2）； （3）； （4）； （5）； （6）； In the formula, c represents the specific heat capacity of chilled water, M1 represents the chilled water flow rate, and T... b Indicates the chilled water return temperature. Q1 represents the average temperature of the chilled water piping during the heat exchange process; Q2 represents the heat transfer between the chilled water piping and the environment; Q3 represents the heat transfer between the chilled water piping and the human body; U2 represents the heat transfer per unit temperature difference between the chilled water piping and the environment; U3 represents the heat transfer per unit temperature difference between the chilled water piping and the human body; T represents the heat transfer per unit temperature difference between the chilled water piping and the human body. s Indicates skin temperature; By combining equations (1) to (6), we obtain the functional relationship between compressor speed N and the heat transfer Q3 between the chilled water pipeline and the human body: （7）。 8. The adaptive control method for the compression cooling suit according to claim 1, characterized in that, The process of inputting real-time human load into a pre-built compressor speed calculation model for a compression-type cooling suit to calculate the compressor speed for adaptive control includes: Let the heat transfer between the chilled water pipe and the human body, Q3, be equal to the real-time human body load. The compressor speed N is calculated by the functional relationship between the compressor speed N and the heat transfer between the chilled water pipe and the human body, Q3.

9. A compression cooling suit, characterized in that, Includes cooling suits, compressor refrigeration units, sensors, and controllers; The cooling suit has chilled water pipes coiled inside, and the outlet end of the chilled water pipes is connected to a chilled water tank. The compression refrigeration unit includes a compressor, a condenser, a throttle valve, and an evaporator connected in sequence, wherein the evaporator is used to cool the liquid in the chilled water tank; The sensor is used to collect heart rate, skin temperature, ambient temperature, ambient humidity, and chilled water tank temperature in real time; The controller is connected to the compression refrigeration unit and the sensor respectively, and is used to execute the adaptive control method of the compression cooling suit according to any one of claims 1 to 8.

10. The compression cooling suit according to claim 9, characterized in that, The chilled water tank is connected to the evaporator via a circulating water pump, and the evaporator is connected to the inlet end of the chilled water pipeline via an inlet pipe.

Images