A child sleep environment regulation method fusing user physiological parameters and environmental temperature

Abstract

The application discloses a child sleep environment regulation method fusing a user physiological parameter and an environmental temperature, and steps of the method comprise the following: when the child is less than or equal to 3 years old, determining corresponding state parameters sp according to basic information of the mother; when the child is more than 3 years old, determining corresponding state parameters sp according to basic information of the child; when the age interval of the child is 0-3 years old, determining target environmental parameters of an air conditioner according to the state parameters sp, the target environmental parameters comprising temperature, humidity and wind speed; when the child is more than 3 years old, determining target environmental parameters of the air conditioner according to the state parameters sp and age and gender in state information of the child; and controlling the air conditioner to operate according to the target environmental parameters of the air conditioner. The application solves the problem of single traditional air conditioner mode and neglect of individual differences by inputting basic information of the child and dynamically generating state parameters in combination with time states, and significantly improves environmental comfort.

Description

Technical Field

[0001] This invention relates to the field of multi-source data processing, specifically a method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature. Background Technology

[0002] With the deep integration of the Internet of Things, artificial intelligence, and smart home technologies, air conditioning systems are gradually evolving from simple environmental control devices into intelligent terminals with environmental perception and autonomous decision-making capabilities. Traditional air conditioners mostly rely on general environmental parameters (such as temperature, humidity, and human activity detection) for rule-based control, lacking refined and personalized environmental support for special groups (especially children). Children's thermoregulation abilities are not yet fully developed, they are highly sensitive to environmental disturbances, and their physiological state, metabolic level, and comfort preferences exhibit significant dynamic changes with age, gender, and physical condition. Existing air conditioning control strategies struggle to build personalized models based on user growth trajectories, failing to achieve continuous perception and adaptive adjustment of children's states.

[0003] While some air conditioners currently offer a child mode, these modes are essentially based on fixed thresholds or preset rules, lacking in-depth modeling of individual user differences and real-time status feedback mechanisms. Prolonged operation under non-personalized environmental parameters can not only affect children's sleep quality but may also have potential adverse effects on their physiological development.

[0004] Therefore, there is an urgent need to construct an intelligent air conditioning control method that integrates multimodal information perception, dynamic user profile modeling, and adaptive decision-making mechanisms, so as to achieve precise regulation and continuous optimization of children's sleep environment with the help of new-generation information technology. Summary of the Invention

[0005] The purpose of this invention is to provide a method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature, comprising the following steps:

[0006] Step 1: Read the entered user information and the user information entry time;

[0007] When the child's age range is [0 years old, 3 years old], the user information is the mother's basic information;

[0008] When a child is older than 3 years old, the user information refers to the child's basic information.

[0009] If the difference between the time when the child's basic information was entered and the current time is greater than a preset time threshold, the controller storing the multilayer perceptron neural network model will be used to update the child's basic information so that the updated child's basic information overwrites the read child's basic information.

[0010] If the difference between the entry time of the child's basic information and the current time is less than or equal to a preset time threshold, keep the read basic information of the child unchanged;

[0011] Step 2: start the mother-infant mode of the air conditioner within a preset time period, or start the mother-infant mode of the air conditioner under manual operation by a user;

[0012] Step 3: when the age of the child is in the interval [0 years old, 3 years old], determine the corresponding state parameter sp according to the basic information of the mother;

[0013] when the age of the child is more than 3 years old, determine the corresponding state parameter sp according to the basic information of the child;

[0014] Step 4: when the age of the child is in the interval [0 years old, 3 years old], determine the target environmental parameters of the air conditioner according to the state parameter sp, wherein the target environmental parameters include temperature, humidity and wind speed;

[0015] when the age of the child is more than 3 years old, determine the target environmental parameters of the air conditioner according to the state parameter sp, as well as the age and gender in the basic information of the child;

[0016] Step 5: control the operation of the air conditioner according to the target environmental parameters of the air conditioner;

[0017] during the operation of the air conditioner, if the user manually adjusts air conditioner parameters, perform feedback compensation on the target environmental parameters of the air conditioner according to the air conditioner parameters manually adjusted by the user, and update the target environmental parameters of the air conditioner;

[0018] during the operation of the air conditioner, if the outdoor temperature is not within a preset temperature range, compensate the target environmental parameters of the air conditioner according to the current date and time, and update the target environmental parameters of the air conditioner.

[0019] further, the basic information of the mother includes height, weight and BMI;

[0020] the basic information of the child includes gender, age, height, weight and BMI; ; , are weight and height, wherein the unit of weight is Kg, and the unit of height is m;

[0021] the age is obtained by: subtracting the date of birth from the current time, and rounding down to obtain the age.

[0022] further, when the child is male and the age is in the interval [0 years old, 3 years old], if the mother's BMI ≤ 18.5, the state parameter sp = 10%; if the mother's BMI satisfies 18.5 < BMI ≤ 25, the state parameter sp = 50%; if the mother's BMI > 25, the state parameter sp = 85%;

[0023] When the child is female and the age range is [0 years old, 3 years old], if the mother's BMI ≤ 18.5, the state parameter sp=10%; if the mother's BMI satisfies 18.5 < BMI ≤ 25, the state parameter sp=50%; if the mother's BMI > 25, the state parameter sp=85%.

[0024] Further, when the child is male and the age range is (3 years old, 6 years old], if the child's BMI ≤ 15.05, the state parameter sp=10%; if 15.05 < child's BMI ≤ 15.26, the state parameter sp=25%; if 15.26 < child's BMI ≤ 15.56, the state parameter sp=50%; if 15.56 < child's BMI ≤ 16.32, the state parameter sp=75%; if the child's BMI > 16.32, the state parameter sp=90%;

[0025] When the child is male and the age range is (6 years old, 10 years old], if the child's BMI ≤ 15.05, the state parameter sp=10%; if 15.05 < child's BMI ≤ 16.01, the state parameter sp=25%; if 16.01 < child's BMI ≤ 17.64, the state parameter sp=50%; if 17.64 < child's BMI ≤ 19.97, the state parameter sp=75%; if the child's BMI > 19.97, the state parameter sp=90%;

[0026] When the child is male and the age range is (10 years old, 12 years old], if the child's BMI ≤ 16.21, the state parameter sp=10%; if 16.21 < child's BMI ≤ 17.68, the state parameter sp=25%; if 17.68 < child's BMI ≤ 19.67, the state parameter sp=50%; if 19.67 < child's BMI ≤ 21.54, the state parameter sp=75%; if the child's BMI > 21.54, the state parameter sp=90%.

[0027] Further, when the child is female and the age range is (3 years old, 6 years old], if the child's BMI ≤ 14.58, the state parameter sp=10%; if 14.58 < child's BMI ≤ 14.72, the state parameter sp=25%; if 14.72 < child's BMI ≤ 15.06, the state parameter sp=50%; if 15.06 < child's BMI ≤ 15.78, the state parameter sp=75%; if the child's BMI > 15.78, the state parameter sp=90%;

[0028] When the child is female and the age range is (6 years, 10 years), if the child's BMI ≤ 14.62, the state parameter sp = 10%; if 14.62 < BMI ≤ 15.24, the state parameter sp = 25%; if 15.24 < BMI ≤ 16.52, the state parameter sp = 50%; if 16.52 < BMI ≤ 18.02, the state parameter sp = 75%; if BMI > 18.02, the state parameter sp = 90%.

[0029] When the child is female and the age range is (10 years, 12 years), if the child's BMI ≤ 15.84, the state parameter sp = 10%; if 15.84 < child's BMI ≤ 17.10, the state parameter sp = 25%; if 17.10 < child's BMI ≤ 18.60, the state parameter sp = 50%; if 18.60 < child's BMI ≤ 20.33, the state parameter sp = 75%; if the child's BMI > 20.33, the state parameter sp = 90%.

[0030] Furthermore, the steps for updating the child's basic information using a controller that stores a multilayer perceptron neural network model include:

[0031] Step S1 inputs the recorded basic information of the child into the multilayer perceptron neural network model to predict the child's average annual height growth and average annual weight growth.

[0032] Step S2 calculates the child's current height and weight based on the child's basic information A, including height and weight, average annual height increase, average annual weight increase, and the difference between the time when basic information A was obtained and the current time.

[0033]

[0034] In the formula, , The child's height and weight at the current time; , For the child's basic information A, the child's height and weight; , These represent the average annual increase in height and the average annual increase in weight. To obtain the time difference between the current time and the time of the child's basic information A, In years;

[0035] Step S3 calculates the child's Body Mass Index (BMI) based on the child's current height and weight;

[0036] Step S4: Divide the current time by the birth date and month, and round down to obtain the child's age;

[0037] Step S5 updates the child's basic information, including age, height, weight, and BMI, based on steps S2-S4.

[0038] Furthermore, the outdoor temperature is monitored by a temperature sensor installed on the outdoor unit of the air conditioner.

[0039] Furthermore, when the child is between 0 and 3 years old, the target environmental parameters for the air conditioner are determined by the first air conditioner parameter prediction model;

[0040] When children are older than 3 years old, the target environmental parameters for air conditioning are determined by a second air conditioning parameter prediction model.

[0041] Both the first and second air conditioner parameter prediction models employ deep learning neural networks, including an input layer, a hidden layer, and an output layer. The deep learning neural network uses the Adam optimizer for parameter training. During training, the mean absolute error is used as the loss function, and the training objective is to minimize the loss function or reach the maximum number of iterations.

[0042] The first air conditioner parameter prediction model was trained using the mother's historical air conditioner parameter training dataset.

[0043] The second air conditioner parameter prediction model was trained using a historical training dataset of children's air conditioner parameters.

[0044] The historical training dataset of mother's air conditioner parameters was constructed as follows:

[0045] A1 determines the corresponding state parameter sp based on the mother's basic information and records the target environmental parameters of the mother in a comfortable state through a questionnaire survey; the target environmental parameters include temperature, humidity and wind speed.

[0046] Using the state parameter sp and the target environment parameter as the input and output of the first air conditioning parameter prediction model, a set of historical training data of the mother's air conditioning parameters is constructed.

[0047] A2 Repeat step A1 to obtain multiple sets of historical training data of mother's air conditioner parameters, thus forming the historical training dataset of mother's air conditioner parameters.

[0048] The historical training dataset for children's air conditioner parameters is constructed as follows:

[0049] B1 determines the corresponding state parameters sp based on the child's basic information and records the target environmental parameters of the mother in a comfortable state through a questionnaire survey; the target environmental parameters include temperature, humidity and wind speed.

[0050] Using the state parameter sp and the child's state information as inputs to the second air conditioning parameter prediction model, and the target environmental parameters as outputs of the second air conditioning parameter prediction model, a set of historical training data on children's air conditioning parameters is constructed.

[0051] B2 Repeat step B1 to obtain multiple sets of historical training data on children's air conditioner parameters, thereby constructing a historical training dataset of children's air conditioner parameters.

[0052] Furthermore, the steps for providing feedback compensation to the target environmental parameters of the air conditioner based on the user's manually adjusted air conditioner parameters include:

[0053] Step 1: Record the difference between the air conditioning parameters manually adjusted by the user and the target environmental parameters, and denote it as the adjustment amount;

[0054] Step 2 determines whether the adjustment amount is greater than the corresponding preset value. If so, the real-time compensation amount is set to the preset value; otherwise, the real-time compensation amount is set to the adjustment amount. The preset values ​​for temperature, humidity, and wind speed are 0.5℃, 5%, and 0.3m / s, respectively.

[0055] Step 3 determines whether the cumulative compensation amount within a preset period is greater than the cumulative preset value. If so, the cumulative compensation amount is set to equal the cumulative preset value; otherwise, the cumulative compensation amount is kept unchanged. The cumulative preset values ​​for temperature, humidity, and wind speed are 2℃, 20%, and 1m / s, respectively.

[0056] Step 4 uses the sum of the cumulative compensation amount and the target environmental parameters as the updated target environmental parameters.

[0057] Furthermore, the steps for compensating the target environmental parameters of the air conditioner based on the current date and time include:

[0058] Step 1 determines the season based on outdoor temperature. When the outdoor temperature is greater than 28 degrees Celsius... When the outdoor temperature is less than 18 degrees Celsius, proceed to step 2. Then proceed to step 3;

[0059] Step 2: Adjust seasonal temperature ; Temperature is one of the target environmental parameters;

[0060] If it is currently daytime, then adjust the day / night compensation temperature. If the current time is night, adjust the day-night compensation temperature. ;

[0061] The weighted sum of diurnal and seasonal compensation temperatures is used as the compensation temperature.

[0062] The sum of the compensated temperature and the target environmental parameters is used as the updated target environmental parameters;

[0063] Step 3: Adjust the seasonal temperature. ;

[0064] If it is currently daytime, then adjust the day / night compensation temperature. If the current time is night, adjust the day-night compensation temperature. ;

[0065] The sum of the compensated temperature and the target environmental parameters is used as the updated target environmental parameters.

[0066] This invention solves the problems of traditional air conditioning modes being singular and ignoring individual differences by recording children's basic information (such as age, gender, BMI, etc.), thus significantly improving environmental comfort.

[0067] This invention distinguishes historical data based on the information entry time and generates status information based on different data sources, ensuring that the assessment results are consistent with the child's real-time physiological state and avoiding parameter lag caused by growth changes.

[0068] This invention uses a comprehensive assessment of parameters such as body mass index (BMI) and gender to automatically set air conditioning parameters such as temperature, humidity, and wind speed that meet the health needs of children. This ensures children's sleep quality and developmental environment while avoiding excessive energy consumption, achieving efficient and energy-saving operation. Attached Figure Description

[0069] Figure 1 This is a flowchart of the method.

[0070] Figure 2 For feedback control process;

[0071] Figure 3 For outdoor temperature compensation process; Detailed Implementation

[0072] The present invention will be further described below with reference to embodiments, but it should not be construed that the scope of the present invention is limited to the following embodiments. Various substitutions and modifications made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention should be included within the scope of protection of the present invention.

[0073] Example 1:

[0074] See Figures 1-3 A method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature includes the following steps:

[0075] Step 1: Read the entered user information and the user information entry time;

[0076] When the child's age range is [0 years old, 3 years old], the user information is the mother's basic information;

[0077] When a child is older than 3 years old, the user information refers to the child's basic information.

[0078] If the difference between the time when the child's basic information was entered and the current time is greater than a preset time threshold, the controller storing the multilayer perceptron neural network model will be used to update the child's basic information so that the updated child's basic information overwrites the read child's basic information.

[0079] If the difference between the time when the child's basic information was entered and the current time is less than or equal to the preset time threshold, the child's basic information will remain unchanged.

[0080] Step 2: Activate the air conditioner's mother-and-baby mode within a preset time period, or activate it manually. The preset time period is selected from the user's sleep time. When the child's age is [0 years, 3 years], select the mother's sleep time period, which defaults to 9 PM to 8 AM; the user can adjust this. When the child is over 3 years old, select the child's sleep time period, which defaults to 9 PM to 8 AM; the user can adjust this.

[0081] Step 3: When the child's age range is [0 years, 3 years], determine the corresponding state parameter sp based on the mother's basic information;

[0082] When a child is older than 3 years old, the corresponding state parameter sp is determined based on the child's basic information;

[0083] Step 4: When the child's age range is [0 years, 3 years], determine the target environmental parameters of the air conditioner based on the state parameter sp, including temperature, humidity and wind speed;

[0084] When a child is older than 3 years old, the target environmental parameters for the air conditioner are determined based on the state parameter sp and the child's age and gender in the basic information.

[0085] Step 5: Control the air conditioner's operation based on the target environmental parameters.

[0086] During the operation of the air conditioner, if the user manually adjusts the air conditioner parameters, the target environmental parameters of the air conditioner will be updated based on the manually adjusted air conditioner parameters.

[0087] If the outdoor temperature is outside the preset temperature range during air conditioner operation, the target environmental parameters of the air conditioner will be compensated and updated according to the current date and time.

[0088] Example 2:

[0089] A method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as in Example 1, further including the mother's basic information including height, weight, and BMI;

[0090] Basic information about children includes gender, age, height, weight, and BMI; ; , are body weight and height, wherein the unit of body weight is Kg, and the unit of height is m;

[0091] The age is obtained by: calculating the difference between the current time and the birth date, and performing floor rounding to obtain the age.

[0092] Example 3:

[0093] A method for regulating and controlling children's sleeping environment integrating user physiological parameters and ambient temperature, the technical content is the same as any one of Examples 1-2, further, when the child is male and the age range is [0 years old, 3 years old], if the mother's BMI ≤ 18.5, the state parameter sp=10%; if the mother's BMI satisfies 18.5 < BMI ≤ 25, the state parameter sp=50%; if the mother's BMI > 25, the state parameter sp=85%;

[0094] When the child is female and the age range is [0 years old, 3 years old], if the mother's BMI ≤ 18.5, the state parameter sp=10%; if the mother's BMI satisfies 18.5 < BMI ≤ 25, the state parameter sp=50%; if the mother's BMI > 25, the state parameter sp=85%.

[0095] Example 4:

[0096] A method for regulating and controlling children's sleeping environment integrating user physiological parameters and ambient temperature, the technical content is the same as any one of Examples 1-3, further, when the child is male and the age range is (3 years old, 6 years old], if the child's BMI ≤ 15.05, the state parameter sp=10%; if 15.05 < the child's BMI ≤ 15.26, the state parameter sp=25%; if 15.26 < the child's BMI ≤ 15.56, the state parameter sp=50%; if 15.56 < the child's BMI ≤ 16.32, the state parameter sp=75%; if the child's BMI > 16.32, the state parameter sp=90%;

[0097] When the child is male and the age range is (6 years old, 10 years old], if the child's BMI ≤ 15.05, the state parameter sp=10%; if 15.05 < the child's BMI ≤ 16.01, the state parameter sp=25%; if 16.01 < the child's BMI ≤ 17.64, the state parameter sp=50%; if 17.64 < the child's BMI ≤ 19.97, the state parameter sp=75%; if the child's BMI > 19.97, the state parameter sp=90%;

[0098] When the child is male and the age range is (10 years, 12 years), if the child's BMI ≤ 16.21, the state parameter sp = 10%; if 16.21 < child's BMI ≤ 17.68, the state parameter sp = 25%; if 17.68 < child's BMI ≤ 19.67, the state parameter sp = 50%; if 19.67 < child's BMI ≤ 21.54, the state parameter sp = 75%; if the child's BMI > 21.54, the state parameter sp = 90%.

[0099] Example 5:

[0100] A method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-4, further comprising the following steps: when the child is female and the age range is (3 years, 6 years), if the child's BMI ≤ 14.58, then the state parameter sp = 10%; if 14.58 < child's BMI ≤ 14.72, then the state parameter sp = 25%; if 14.72 < child's BMI ≤ 15.06, then the state parameter sp = 50%; if 15.06 < child's BMI ≤ 15.78, then the state parameter sp = 75%; if the child's BMI > 15.78, then the state parameter sp = 90%.

[0101] When the child is female and the age range is (6 years, 10 years), if the child's BMI ≤ 14.62, the state parameter sp = 10%; if 14.62 < BMI ≤ 15.24, the state parameter sp = 25%; if 15.24 < BMI ≤ 16.52, the state parameter sp = 50%; if 16.52 < BMI ≤ 18.02, the state parameter sp = 75%; if BMI > 18.02, the state parameter sp = 90%.

[0102] When the child is female and the age range is (10 years, 12 years), if the child's BMI ≤ 15.84, the state parameter sp = 10%; if 15.84 < child's BMI ≤ 17.10, the state parameter sp = 25%; if 17.10 < child's BMI ≤ 18.60, the state parameter sp = 50%; if 18.60 < child's BMI ≤ 20.33, the state parameter sp = 75%; if the child's BMI > 20.33, the state parameter sp = 90%.

[0103] Example 6:

[0104] A method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-5, further comprising the step of updating the child's basic information using a controller storing a multilayer perceptron neural network model, including:

[0105] Step S1 inputs the recorded basic information of the child into the multilayer perceptron neural network model to predict the child's average annual height growth and average annual weight growth.

[0106] Step S2 calculates the child's current height and weight based on the child's basic information A, including height and weight, average annual height increase, average annual weight increase, and the difference between the time when basic information A was obtained and the current time.

[0107]

[0108] In the formula, , The child's height and weight at the current time; , For the child's basic information A, the child's height and weight; , These represent the average annual increase in height and the average annual increase in weight. To obtain the time difference between the current time and the time of the child's basic information A, In years;

[0109] Step S3 calculates the child's Body Mass Index (BMI) based on the child's current height and weight;

[0110] Step S4: Divide the current time by the birth date and month, and round down to obtain the child's age;

[0111] Step S5 updates the child's basic information, including age, height, weight, and BMI, based on steps S2-S4.

[0112] Example 7:

[0113] A method for regulating a child's sleep environment that integrates user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-6, further wherein the outdoor temperature is monitored by a temperature sensor installed on the outdoor unit of the air conditioner.

[0114] Example 8:

[0115] A method for regulating a child's sleep environment that integrates user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-7, further wherein when the child is [0 years old, 3 years old], the target environmental parameters of the air conditioner are determined by a first air conditioner parameter prediction model;

[0116] When children are older than 3 years old, the target environmental parameters for air conditioning are determined by a second air conditioning parameter prediction model.

[0117] Both the first and second air conditioner parameter prediction models employ deep learning neural networks, including an input layer, a hidden layer, and an output layer. The deep learning neural network uses the Adam optimizer for parameter training. During training, the mean absolute error is used as the loss function, and the training objective is to minimize the loss function or reach the maximum number of iterations.

[0118] The first air conditioner parameter prediction model was trained using the mother's historical air conditioner parameter training dataset.

[0119] The second air conditioner parameter prediction model was trained using a historical training dataset of children's air conditioner parameters.

[0120] The historical training dataset of mother's air conditioner parameters was constructed as follows:

[0121] A1 determines the corresponding state parameter sp based on the mother's basic information and records the target environmental parameters of the mother in a comfortable state through a questionnaire survey; the target environmental parameters include temperature, humidity and wind speed.

[0122] Using the state parameter sp and the target environment parameter as the input and output of the first air conditioning parameter prediction model, a set of historical training data of the mother's air conditioning parameters is constructed.

[0123] A2 Repeat step A1 to obtain multiple sets of historical training data of mother's air conditioner parameters, thus forming the historical training dataset of mother's air conditioner parameters.

[0124] The historical training dataset for children's air conditioner parameters is constructed as follows:

[0125] B1 determines the corresponding state parameters sp based on the child's basic information and records the target environmental parameters of the mother in a comfortable state through a questionnaire survey; the target environmental parameters include temperature, humidity and wind speed.

[0126] Using the state parameter sp and the child's state information as inputs to the second air conditioning parameter prediction model, and the target environmental parameters as outputs of the second air conditioning parameter prediction model, a set of historical training data on children's air conditioning parameters is constructed.

[0127] B2 Repeat step B1 to obtain multiple sets of historical training data on children's air conditioner parameters, thereby constructing a historical training dataset of children's air conditioner parameters.

[0128] Example 9:

[0129] A method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-8, further comprising the step of providing feedback compensation to the target environmental parameters of the air conditioner based on the air conditioner parameters manually adjusted by the user, including:

[0130] Step 1: Record the difference between the air conditioning parameters manually adjusted by the user and the target environmental parameters, and denote it as the adjustment amount;

[0131] Step 2 determines whether the adjustment amount is greater than the corresponding preset value. If so, the real-time compensation amount is set to the preset value; otherwise, the real-time compensation amount is set to the adjustment amount. The preset values ​​for temperature, humidity, and wind speed are 0.5℃, 5%, and 0.3m / s, respectively.

[0132] Step 3 determines whether the cumulative compensation amount within a preset period is greater than the cumulative preset value. If so, the cumulative compensation amount is set to equal the cumulative preset value; otherwise, the cumulative compensation amount is kept unchanged. The cumulative preset values ​​for temperature, humidity, and wind speed are 2℃, 20%, and 1m / s, respectively.

[0133] Step 4 uses the sum of the cumulative compensation amount and the target environmental parameters as the updated target environmental parameters.

[0134] Example 10:

[0135] A method for regulating a child's sleep environment that integrates user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-9, further comprising the step of compensating the target environmental parameters of the air conditioner according to the current date and time, including:

[0136] Step 1 determines the season based on outdoor temperature. When the outdoor temperature is greater than 28 degrees Celsius... When the outdoor temperature is less than 18 degrees Celsius, proceed to step 2. Then proceed to step 3;

[0137] Step 2: Adjust seasonal temperature ; Temperature is one of the target environmental parameters;

[0138] If it is currently daytime, then adjust the day / night compensation temperature. If the current time is night, adjust the day-night compensation temperature. ;

[0139] The weighted sum of diurnal and seasonal compensation temperatures is used as the compensation temperature.

[0140] The sum of the compensated temperature and the target environmental parameters is used as the updated target environmental parameters;

[0141] Step 3: Adjust the seasonal temperature. ;

[0142] If it is currently daytime, then adjust the day / night compensation temperature. If the current time is night, adjust the day-night compensation temperature. ;

[0143] The sum of the compensated temperature and the target environmental parameters is used as the updated target environmental parameters.

[0144] Example 11:

[0145] A method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-10, wherein the multilayer perceptron neural network model is a feedforward fully connected neural network, including 1 input layer, 2 hidden layers and 1 output layer.

[0146] The input vector X includes the following features:

[0147] 1) Child's gender (using 0 / 1 encoding, e.g., 1 for male and 0 for female);

[0148] 2) Child's age (unit: years);

[0149] 3) Body Mass Index (BMI) for children;

[0150] 4) The status parameter sp corresponding to the BMI in the entered basic information of the child; the method for determining the status parameter sp is the same as in Examples 4-5;

[0151] The structure of the multilayer perceptron neural network is as follows:

[0152] The input layer contains 4 neurons;

[0153] The first hidden layer contains 16 neurons and uses the ReLU activation function;

[0154] The second hidden layer contains 16 neurons and uses the ReLU activation function;

[0155] Output layer: Contains 2 neurons, without using an activation function, and outputs the average annual height increase G_H (cm / year) and the average annual weight increase G_W (kg / year), respectively.

[0156] This multilayer perceptron model is trained using supervised learning. The training data comes from the historical growth data of multiple children. For each child, their height and weight are recorded at different time points. Based on the changes in height and weight within adjacent time periods, the true average annual height growth G_H_true and average annual weight growth G_W_true are calculated, and {G_H_true, G_W_true} are used as the training label Y_train.

[0157] During the training phase, the input vector X_train and the corresponding label Y_train are batch-input into the multilayer perceptron neural network model. The Adam optimizer is used, and the mean absolute error (MAE) is used as the loss function to iteratively update the network weights and biases until the loss converges on the validation set, thus obtaining the final prediction model.

[0158] During air conditioner operation, the user status information assessment module takes the child's basic information (including gender, age, and BMI) and status parameters SP as input vector X, and inputs it into a pre-trained multilayer perceptron model to obtain the predicted average annual height growth G_H and average annual weight growth G_W. Then, based on the above G_H, G_W and time difference ΔT, the child's basic information is updated according to the aforementioned formula.

[0159] Example 12:

[0160] A method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-11. Further, the first and second air conditioning parameter prediction models are both air conditioning parameter prediction models, which are feedforward fully connected deep learning neural networks, including 1 input layer, 2 hidden layers and 1 output layer.

[0161] The input vector X_env includes the following features:

[0162] 1) Child's gender (0 / 1 encoding);

[0163] 2) Child's age (unit: years);

[0164] 3) Body Mass Index (BMI) for children;

[0165] 4) State parameters SP (numerical features, which can be one of 10, 25, 50, 75, or 90, and are input into the model after normalization).

[0166] Therefore, the input vector X_env must include at least the four features mentioned above.

[0167] The structure of the air conditioning parameter prediction model is as follows:

[0168] The input layer consists of 4 neurons.

[0169] The first hidden layer contains 16 neurons and uses the ReLU activation function;

[0170] The second hidden layer contains 16 neurons and uses the ReLU activation function;

[0171] Output layer: Contains 3 neurons, without using an activation function, and outputs the target temperature T_target, target relative humidity RH_target, and target wind speed V_target respectively.

[0172] The air conditioner parameter prediction model is trained using supervised learning. The training data comes from the historical usage data of multiple users in children's mode. For each historical record, the child's information at that time (gender, age, BMI, SP, etc.) and the air conditioner operating parameters (temperature, humidity, fan speed) that the user ultimately adopted or maintained stably within a certain period of time are collected. The stable operating parameters are used as the supervision label Y_env = {T_target_true, RH_target_true, V_target_true}.

[0173] During the training phase, (X_env, Y_env) are input into the air conditioning parameter prediction model in batches. The Adam optimizer is used, and the network parameters are iteratively updated with the mean absolute error (MAE) as the loss function until the loss converges on the validation set, thus obtaining the air conditioning parameter prediction model for online prediction.

[0174] During the operation of the air conditioner, when it is necessary to set the target environmental parameters of the air conditioner according to the child's status, the air conditioner parameter setting module takes the current child's gender, age, BMI and status parameter SP as the input vector X_env, and inputs it into the pre-trained air conditioner parameter prediction model to obtain the target temperature T_target, target relative humidity RH_target and target wind speed V_target. The air conditioner control module controls the operation of the air conditioner accordingly.

[0175] Example 13:

[0176] A method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-12, further wherein the processing method of the first and second air conditioning parameter prediction models is as follows:

[0177] The input vector X first undergoes a linear transformation of the first layer of the neural network (defined by weights W1 and bias b1), and then is processed by a non-linear activation function f (using ReLU, whose mathematical formula is: f(x)=max(0,x)) to obtain the preliminary feature representation A1.

[0178] Deep processing:

[0179] This feature indicates that A1 is then fed into the next layer, repeating the same process to extract higher-level, more abstract features.

[0180] The output prediction Y of the air conditioning parameter prediction model is as follows:

[0181] After processing by all hidden layers, the final features are fed into the output layer. The output layer performs a linear transformation to directly calculate the final prediction result Y—namely, temperature, humidity, and wind speed.

[0182] Example 14:

[0183] A method for regulating a child's sleep environment that integrates user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-13, further comprising the following steps: when the detected outdoor temperature is greater than or equal to 28°C or less than or equal to 18°C:

[0184] When the child is between 0-3 years old, if the mother's state parameter sp=10%, then the temperature Ta=25.5°C, humidity RH=50%, and wind speed Va=0.15m / s; if the mother's state parameter sp=50%, then the temperature Ta=25°C, humidity RH=50%, and wind speed Va=0.2m / s; if the mother's state parameter sp=85%, then the temperature Ta=24.5°C, humidity RH=50%, and wind speed Va=0.25m / s. When the child is male and between 3-6 years old, if the state parameter sp=10%, then the temperature Ta=26.5°C, humidity RH=50%, and wind speed Va=0.2m / s; if the state parameter sp=25%, then Ta=26.3°C, RH=50%, and Va=0.2m / s; if the state parameter sp=50%, then Ta=26°C, RH=50%. Va = 0.25 m / s. If the state parameters sp = 75%, Ta = 25.8°C, RH = 50%, Va = 0.25 m / s. If the state parameters sp = 90%, then Ta = 25.5°C, RH = 50%, Va = 0.3 m / s.

[0185] When the child is male and aged 7-10 years, if the state parameter sp=10%, then the temperature Ta=26°C, humidity RH=50%, and wind speed Va=0.2m / s; if the state parameter sp=25%, then Ta=25.8°C, RH=50%, and Va=0.2m / s; if the state parameter sp=50%, then Ta=25.5°C, RH=50%, and Va=0.25m / s; if the state parameter sp=75%, then Ta=25.3°C, RH=50%, and Va=0.25m / s; if the state parameter sp=90%, then Ta=25°C, RH=50%, and Va=0.3m / s.

[0186] When the child is male and in the age range of 11-12 years, if the state parameter sp=10%, then the temperature Ta=25.5°C, humidity RH=50%, and wind speed Va=0.2m / s; if the state parameter sp=25%, then Ta=25.3°C, RH=50%, and Va=0.2m / s; if the state parameter sp=50%, then Ta=25°C, RH=50%, and Va=0.25m / s; if the state parameter sp=75%, then Ta=24.8°C, RH=50%, and Va=0.25m / s; if the state parameter sp=90%, then Ta=24.5°C, RH=50%, and Va=0.3m / s.

[0187] Example 15:

[0188] A method for regulating a child's sleep environment that integrates user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-14, further comprising the following steps: when the detected outdoor temperature is greater than or equal to 28°C or less than or equal to 18°C:

[0189] When the child's age range is 0-3 years, if the mother's state parameter sp=10%, then the temperature Ta=25.5°C, humidity RH=50%, and wind speed Va=0.15m / s; if the mother's state parameter sp=50%, then the temperature Ta=25°C, humidity RH=50%, and wind speed Va=0.2m / s; if the mother's state parameter sp=85%, then the temperature Ta=24.5°C, humidity RH=50%, and wind speed Va=0.25m / s.

[0190] When the child is female and aged 3-6 years, if the state parameter sp=10%, then the temperature Ta=26.8°C, humidity RH=50%, and wind speed Va=0.15m / s; if the state parameter sp=25%, then Ta=26.5°C, RH=50%, and Va=0.15m / s; if the state parameter sp=50%, then Ta=26.3°C, RH=50%, and Va=0.2m / s; if the state parameter sp=75%, then Ta=26°C, RH=50%, and Va=0.25m / s; if the state parameter sp=90%, then Ta=25.8°C, RH=50%, and Va=0.25m / s.

[0191] When the child is female and aged 7-10 years, if the state parameter sp=10%, then the temperature Ta=26.5°C, humidity RH=50%, and wind speed Va=0.15m / s; if the state parameter sp=25%, then Ta=26.3°C, RH=50%, and Va=0.15m / s; if the state parameter sp=50%, then Ta=26°C, RH=50%, and Va=0.2m / s; if the state parameter sp=75%, then Ta=25.5°C, RH=50%, and Va=0.25m / s; if the state parameter sp=90%, then Ta=25.3°C, RH=50%, and Va=0.25m / s.

[0192] When the child is female and aged 11-12 years, if the state parameter sp = 10%, then the temperature Ta = 26.3°C, humidity RH = 50%, and wind speed Va = 0.15 m / s; if the state parameter sp = 25%, then Ta = 26°C, RH = 50%, and Va = 0.15 m / s; if the state parameter sp = 50%, then Ta = 25.5°C, RH = 50%, and Va = 0.2 m / s; if the state parameter sp = 75%, then Ta = 25°C, RH = 50%, and Va = 0.25 m / s; and if the state parameter sp = 90%, then Ta = 24.8°C, RH = 50%, and Va = 0.25 m / s. The above table lookup method can be used as a simplified implementation of the first and second air conditioning parameter prediction models. When a neural network is not used, the target temperature, humidity, and wind speed can be directly determined by looking up the table.

[0193] Example 16:

[0194] In the aforementioned embodiments, the input to the multilayer perceptron neural network model includes at least the child's gender, age, BMI, and state parameter SP. To further improve prediction accuracy, in this embodiment, features such as height, weight, and time difference ΔT can be added, and SP is one-hot encoded to form an extended input vector. A method for regulating the sleep environment of children by integrating user physiological parameters and ambient temperature, with the same technical content as any one of embodiments 1-15, further involves querying the WHO BMI percentile curve or table based on the child's age and gender to obtain the accurate BMI percentile. Then, this continuous percentile is mapped to the SP level:

[0195] Percentile ≤ 10% → SP level 10%;

[0196] 10% < percentile ≤ 25% → SP level 25%;

[0197] 25% < percentile ≤ 50% → SP level 50%;

[0198] 50% < percentile ≤ 75% → SP level 75%;

[0199] Percentile > 75% → SP level 90%.

[0200] The SP level is converted into five binary features: SP10%, SP25%, SP50%, SP75%, and SP90%. For example, when the SP level is 75%, it is encoded as [0,0,0,1,0].

[0201] Input feature vector (standardized: [gender=1, age=5, height=110, weight=20, BMI=16.53, ΔT=5, SP10%=0, SP25%=0, SP50%=0, SP75%=1, SP90%=0])

[0202] The model outputs the average annual height increase (e.g., 5 cm / year) and the average annual weight increase (e.g., 2 kg / year); predicting that after 5 years: height ≈ 110 + 5*5 = 135cm, weight ≈ 20 + 2*5 = 30kg

[0203] Calculate and predict BMI and SP classification.

[0204] Example 17:

[0205] An air conditioner that considers the comfort of children's sleep using the method described in any one of Examples 1-16, the air conditioner is used for temperature regulation during children's sleep, including a mother and baby mode activation module, a user information input module, a user status information evaluation module, a status parameter evaluation module, an air conditioner parameter setting module, an air conditioner parameter adjustment module, an air conditioner parameter feedback compensation module, an outdoor temperature compensation module, and an air conditioner control module.

[0206] The mother-and-baby mode activation module activates the air conditioner's mother-and-baby mode at a preset time, or activates the air conditioner's mother-and-baby mode under the user's manual operation.

[0207] When the air conditioner's mother and baby mode is activated, the user status information assessment module, status parameter assessment module, air conditioner parameter setting module, air conditioner parameter adjustment module, air conditioner parameter feedback compensation module, and outdoor temperature compensation module start operating.

[0208] When a child is 3 years of age or younger, the user information entry module enters the mother's basic information;

[0209] When a child is older than 3 years old, the user information entry module enters the child's basic information;

[0210] When a child is 3 years of age or younger, the user status information assessment module determines the corresponding status parameter sp based on the mother's basic information.

[0211] When a child is 3 years of age or older, the user status information assessment module determines the corresponding status parameter sp based on the child's basic information;

[0212] When the child's age range is 0-3 years, the air conditioning parameter setting module determines the target environmental parameters of the air conditioner based on the status parameter sp, including temperature, humidity and wind speed;

[0213] When a child is older than 3 years old, the air conditioning parameter setting module determines the target environmental parameters of the air conditioner based on the status parameter sp and the child's age and gender in the status information.

[0214] During the operation of the air conditioner, if the user manually adjusts the air conditioner parameters, the air conditioner parameter feedback compensation module will provide feedback compensation to the target environmental parameters of the air conditioner based on the manually adjusted air conditioner parameters and update the target environmental parameters of the air conditioner.

[0215] During the operation of the air conditioner, the outdoor temperature compensation module determines whether the outdoor temperature is within the preset range. If not, it compensates the target environmental parameters of the air conditioner according to the current date and time and updates the target environmental parameters of the air conditioner.

[0216] The air conditioning control module controls the operation of the air conditioner according to the target environmental parameters.

[0217] Example 18:

[0218] An air conditioner that takes into account the sleep comfort of children, with the same technical content as Embodiment 17, further wherein the air conditioner operation process is as follows:

[0219] STEP1. Info=[Name,Gender,Age,Height,Weight];

[0220] Name and Gender are constant values ​​by default. Age, Height, and Weight are time variables.

[0221] The initial input result is denoted as Aget0,Heightt0,Weightt0, and the input time is denoted as t0.

[0222] Here, Age is the input birth year and month, which is the difference between the current time and rounded down. For example, if the current time is July 2025 and the birth time is April 2021, the difference is 4 years and 3 months, which is rounded down to 4 years, i.e., Age0 = 4.

[0223] Height is measured in meters (m), and is rounded to two decimal places, for example, 1.23m.

[0224] Weight is measured in kg, with one decimal place, for example, 28.6 kg.

[0225] STEP2, Calculate the information correction value:

[0226] BMIt0 = Weightt0 / (Heightt0) 2 ;

[0227] Infoadt0=[Name,Gender,Aget0,BMIt0];

[0228] STEP 3: Confirm physical condition:

[0229] Methods for predicting children's condition (Tables 1 and 2: Determining SPt0, where SPt0 is one of five preset values ​​[SP10, SP25, SP50, SP75, SP90]).

[0230] Infot0=[Name,Gender,Aget0,SPt0];

[0231] STEP4, Recommended target environment parameters:

[0232] Look up Table 2 to determine the target parameters for temperature and humidity, which will serve as the execution target for the air conditioner.

[0233] STEP 5, Data Update:

[0234] The variable Aget1 = Aget0 + the change over time, rounded down. For example, if Aget0 is 4, after 2 years and 1 month it will be 6 years and 1 month, which is rounded down to 6, so Aget1 = 6.

[0235] Infot1=[Name,Gender,Aget1,SPt0];

[0236] STEP 6, Recommended target environment parameters:

[0237] Temperature, humidity, and wind speed are determined by looking up a target parameter mapping table (as shown in Tables 1 and 2) and used as the execution targets for the air conditioner. The update method can be MLP prediction or manual update by the user; the table lookup method can be independent of the MLP.

[0238] Table 1

[0239]

[0240] Table 2

[0241]

[0242] Alternatively, temperature, humidity, and wind speed can be determined using either the first or second air conditioning parameter prediction model as the execution targets of the air conditioner.

Claims

1. A method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature, characterized in that, comprising the following steps: Step 1: read the entered user information and the entry time of the user information; when the age of the child is in the interval [0 years old, 3 years old], the user information is basic information of the mother; when the age of the child is more than 3 years old, the user information is basic information of the child; if the difference between the entry time of the child's basic information and the current time is greater than a preset time threshold, a controller storing a multi-layer perceptron neural network model is used to update the child's basic information, and the read child's basic information is overwritten with the updated child's basic information; if the difference between the entry time of the child's basic information and the current time is less than or equal to the preset time threshold, the read child's basic information is maintained unchanged; Step 2: start the mother-baby mode of the air conditioner within a preset time period, or start the mother-baby mode of the air conditioner under manual operation by a user; Step 3: when the age of the child is in the interval [0 years old, 3 years old], determine the corresponding state parameter sp according to the basic information of the mother; when the age of the child is more than 3 years old, determine the corresponding state parameter sp according to the basic information of the child; Step 4: when the age of the child is in the interval [0 years old, 3 years old], determine the target environmental parameters of the air conditioner according to the state parameter sp, wherein the target environmental parameters comprise temperature, humidity and wind speed; when the age of the child is more than 3 years old, determine the target environmental parameters of the air conditioner according to the state parameter sp, and the age and gender in the basic information of the child; Step 5: control the operation of the air conditioner according to the target environmental parameters of the air conditioner; during the operation of the air conditioner, if a user manually adjusts air conditioner parameters, feedback compensation is performed on the target environmental parameters of the air conditioner according to the manually adjusted air conditioner parameters, and the target environmental parameters of the air conditioner are updated; during the operation of the air conditioner, if the outdoor temperature is not within a preset temperature interval, compensation is performed on the target environmental parameters of the air conditioner according to the current date and time, and the target environmental parameters of the air conditioner are updated; the basic information of the mother comprises height, weight and BMI; Basic information about children includes gender, age, height, weight, and BMI; ; , For weight and height, where weight is in kg and height is in m; the age is obtained by: calculating the difference between the current time and the birth date, and rounding down to obtain the age.

2. The method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, when the child is male and the age is in the interval [0 years old, 3 years old], if the mother's BMI ≤ 18.5, the state parameter sp = 10%; if the mother's BMI satisfies 18.5 < BMI ≤ 25, the state parameter sp = 50%; if the mother's BMI > 25, the state parameter sp = 85%; when the child is female and the age is in the interval [0 years old, 3 years old], if the mother's BMI ≤ 18.5, the state parameter sp = 10%; if the mother's BMI satisfies 18.5 < BMI ≤ 25, the state parameter sp = 50%; if the mother's BMI > 25, the state parameter sp = 85%.

3. The method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, when the child is male and the age is in the interval (3 years old, 6 years old], if the child's BMI ≤ 15.05, the state parameter sp = 10%; if 15.05 < the child's BMI ≤ 15.26, the state parameter sp = 25%; if 15.26 < the child's BMI ≤ 15.56, the state parameter sp = 50%; if 15.56 < the child's BMI ≤ 16.32, the state parameter sp = 75%; if the child's BMI > 16.32, the state parameter sp = 90%; When the child is male and the age range is (6 years, 10 years), if the child's BMI ≤ 15.05, the state parameter sp = 10%; if 15.05 < BMI ≤ 16.01, the state parameter sp = 25%; if 16.01 < BMI ≤ 17.64, the state parameter sp = 50%; if 17.64 < BMI ≤ 19.97, the state parameter sp = 75%; if BMI > 19.97, the state parameter sp = 90%. When the child is male and the age range is (10 years, 12 years), if the child's BMI ≤ 16.21, the state parameter sp = 10%; if 16.21 < child's BMI ≤ 17.68, the state parameter sp = 25%; if 17.68 < child's BMI ≤ 19.67, the state parameter sp = 50%; if 19.67 < child's BMI ≤ 21.54, the state parameter sp = 75%; if the child's BMI > 21.54, the state parameter sp = 90%.

4. The method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, When the child is female and the age range is (3 years, 6 years), if the child's BMI ≤ 14.58, the state parameter sp = 10%; if 14.58 < BMI ≤ 14.72, the state parameter sp = 25%; if 14.72 < BMI ≤ 15.06, the state parameter sp = 50%; if 15.06 < BMI ≤ 15.78, the state parameter sp = 75%; if BMI > 15.78, the state parameter sp = 90%. When the child is female and the age range is (6 years, 10 years), if the child's BMI ≤ 14.62, then the state parameter sp = 10%; if 14.62 < child's BMI ≤ 15.24, then the state parameter sp = 25%; if 15.24 < child's BMI ≤ 16.52, then the state parameter sp = 50%. If a child's BMI is 16.52 < 18.02, then the state parameter sp = 75%; if a child's BMI is > 18.02, then the state parameter sp = 90%. When the child is female and the age range is (10 years, 12 years), if the child's BMI ≤ 15.84, the state parameter sp = 10%; if 15.84 < child's BMI ≤ 17.10, the state parameter sp = 25%; if 17.10 < child's BMI ≤ 18.60, the state parameter sp = 50%; if 18.60 < child's BMI ≤ 20.33, the state parameter sp = 75%; if the child's BMI > 20.33, the state parameter sp = 90%.

5. The method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, The steps for updating a child's basic information using a controller that stores a multilayer perceptron neural network model include: Step S1 inputs the recorded basic information of the child into the multilayer perceptron neural network model to predict the child's average annual height growth and average annual weight growth. Step S2 calculates the child's current height and weight based on the child's basic information A, including height and weight, average annual height increase, average annual weight increase, and the difference between the time when basic information A was obtained and the current time. In the formula, , The child's height and weight at the current time; , For the child's basic information A, the child's height and weight; , These represent the average annual increase in height and the average annual increase in weight. To obtain the time difference between the current time and the time of the child's basic information A, In years; Step S3 calculates the child's Body Mass Index (BMI) based on the child's current height and weight; Step S4: Divide the current time by the birth date and month, and round down to obtain the child's age; Step S5 updates the child's basic information, including age, height, weight, and BMI, based on steps S2-S4.

6. The method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, The outdoor temperature is monitored by a temperature sensor installed on the outdoor unit of the air conditioner.

7. The method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, When the child is between 0 and 3 years old, the target environmental parameters of the air conditioner are determined by the first air conditioner parameter prediction model. When children are older than 3 years old, the target environmental parameters for air conditioning are determined by a second air conditioning parameter prediction model. Both the first and second air conditioner parameter prediction models employ deep learning neural networks, including an input layer, a hidden layer, and an output layer. The deep learning neural network uses the Adam optimizer for parameter training. During training, the mean absolute error is used as the loss function, and the training objective is to minimize the loss function or reach the maximum number of iterations. The first air conditioner parameter prediction model was trained using the mother's historical air conditioner parameter training dataset. The second air conditioner parameter prediction model was trained using a historical training dataset of children's air conditioner parameters. The historical training dataset of mother's air conditioner parameters was constructed as follows: A1 determines the corresponding state parameter sp based on the mother's basic information and records the target environmental parameters of the mother in a comfortable state through a questionnaire survey; the target environmental parameters include temperature, humidity and wind speed. Using the state parameter sp and the target environment parameter as the input and output of the first air conditioning parameter prediction model, a set of historical training data of the mother's air conditioning parameters is constructed. A2 Repeat step A1 to obtain multiple sets of historical training data on mother's air conditioner parameters, thereby constructing a historical training dataset of mother's air conditioner parameters. The historical training dataset for children's air conditioner parameters is constructed as follows: B1 determines the corresponding state parameters sp based on the child's basic information and records the target environmental parameters of the child in a comfortable state through a questionnaire survey; the target environmental parameters include temperature, humidity and wind speed; Using state parameters sp, child's gender, age, and BMI as inputs to the second air conditioning parameter prediction model, and target environmental parameters as outputs of the second air conditioning parameter prediction model, a set of historical training data on children's air conditioning parameters is constructed. B2 Repeat step B1 to obtain multiple sets of historical training data on children's air conditioner parameters, thereby constructing a historical training dataset of children's air conditioner parameters.

8. The method for regulating children's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, The steps for providing feedback compensation to the target environmental parameters of the air conditioner based on the user's manually adjusted air conditioner parameters include: Step 1: Record the difference between the air conditioning parameters manually adjusted by the user and the target environmental parameters, and denote it as the adjustment amount; Step 2 determines whether the adjustment amount is greater than the corresponding preset value. If so, the real-time compensation amount is set to the preset value; otherwise, the real-time compensation amount is set to the adjustment amount. The preset values ​​for temperature, humidity, and wind speed are 0.5℃, 5%, and 0.3m / s, respectively. Step 3 determines whether the cumulative compensation amount within a preset period is greater than the cumulative preset value. If so, the cumulative compensation amount is set to equal the cumulative preset value; otherwise, the cumulative compensation amount is kept unchanged. The cumulative preset values ​​for temperature, humidity, and wind speed are 2℃, 20%, and 1m / s, respectively. Step 4 uses the sum of the cumulative compensation amount and the target environmental parameters as the updated target environmental parameters.

9. A method for regulating a child's sleep environment by integrating user physiological parameters and ambient temperature according to claim 1, characterized in that, The steps for compensating for the target environmental parameters of the air conditioner based on the current date and time include: Step 1 determines the season based on outdoor temperature. When the outdoor temperature is greater than 28 degrees Celsius... When the outdoor temperature is less than 18 degrees Celsius, proceed to step 2. Then proceed to step 3; Step 2: Adjust seasonal temperature ; Temperature is one of the target environmental parameters; If it is currently daytime, then adjust the day / night compensation temperature. If the current time is night, adjust the day-night compensation temperature. ; The weighted sum of diurnal and seasonal compensation temperatures is used as the compensation temperature. The sum of the compensated temperature and the target environmental parameters is used as the updated target environmental parameters; Step 3: Adjust the seasonal temperature. ; If it is currently daytime, then adjust the day / night compensation temperature. If the current time is night, adjust the day-night compensation temperature. ; The sum of the compensated temperature and the target environmental parameters is used as the updated target environmental parameters.

Images