Indoor temperature and humidity and light control method and system

Abstract

The application discloses an indoor temperature and humidity and light control method and system, belongs to the communication field, and through real-time collection of indoor temperature and humidity and light intensity, is respectively handled through a curtain opening degree model and an air conditioner temperature regulation model, obtains specific curtain opening degree values and air conditioner temperature values, and through humidity data, obtains specific humidity values; through the specific curtain opening degree values, the air conditioner temperature values and the humidity values, accurate regulation and control of the air conditioner and the curtain are realized, and the problem that the prior art lacks a unified automatic indoor temperature and humidity and light control method is solved.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, and specifically relates to a method and system for controlling indoor temperature, humidity and light. Background Technology

[0002] Ordinary air conditioners typically offer temperature setting and dehumidification functions, operating with the set temperature as the target. Curtains can be manually adjusted or their opening degree set. However, manually adjusting the air conditioner temperature or curtains requires the user to constantly adjust the settings based on changes in the indoor environment. Humans are relatively insensitive to temperature and light, and may not even be able to directly perceive humidity levels. This results in fluctuating temperatures and lighting conditions, leading to inconsistent indoor comfort and a lack of a unified, automatic method for controlling indoor temperature, humidity, and light. Summary of the Invention

[0003] In view of the above-mentioned shortcomings in the prior art, the present invention provides an indoor temperature, humidity and light control method and system that solves the problem of the lack of a unified automatic indoor temperature, humidity and light control method in the prior art.

[0004] To achieve the above-mentioned objectives, the technical solution adopted by this invention is: a method for controlling indoor temperature, humidity, and light, comprising the following steps:

[0005] S1. Collect and preprocess indoor temperature, humidity and light intensity to obtain preprocessed temperature data, humidity data and light intensity data;

[0006] S2. Process the light intensity data using the curtain opening model to obtain the curtain opening value;

[0007] S3. Use an air conditioning temperature control model to process the temperature data and obtain the air conditioning temperature value;

[0008] S4. Obtain the humidity value based on the humidity data;

[0009] S5. Adjust the air conditioner and curtains according to the air conditioner temperature value, curtain opening value, and humidity value respectively.

[0010] The beneficial effects of the method of the present invention are as follows: by collecting indoor temperature, humidity and light intensity in real time, and processing them through a curtain opening model and an air conditioning temperature control model respectively, specific curtain opening values ​​and air conditioning temperature values ​​are obtained. Then, through humidity data, specific humidity values ​​are obtained. By using specific curtain opening values, air conditioning temperature values ​​and humidity values, precise control of air conditioning and curtains can be achieved, which solves the problem that there is a lack of a unified automatic indoor temperature, humidity and light control method in the existing technology.

[0011] Further, step S1 includes the following sub-steps:

[0012] S11. Collect indoor temperature, humidity and light intensity data to obtain initial temperature data, initial humidity data and initial light intensity data;

[0013] S12. Apply amplitude limiting filtering to remove outliers from the initial temperature data, initial humidity data and initial light intensity data to obtain the first filtered temperature data, the first filtered humidity data and the first filtered light intensity data.

[0014] S13. The temperature data, humidity data, and light intensity data filtered in the first filtering are filtered again using the moving average filter to obtain the preprocessed temperature data, humidity data, and light intensity data.

[0015] The beneficial effects of the above-mentioned further scheme are as follows: the first filtering removes out-of-range values, and the second filtering makes the overall data conform to the actual temperature, humidity and light intensity trends, eliminating the influence of individual outliers.

[0016] Furthermore, the filtering formula in step S13 is:

[0017]

[0018] Where χ represents the filtered temperature, humidity, or light intensity data, x new For the latest data in the moving average filter's moving block, x med For the median data in the moving average filter sliding block, x max For the largest first-filtered temperature data, first-filtered humidity data, or first-filtered light intensity data, x min For the minimum first-filter temperature data, first-filter humidity data, or first-filter light intensity data, x i For the i-th first filtered temperature data, first filtered humidity data, or first filtered light intensity data in the sliding block of the moving average filter, N is the length of the sliding block, and χ′ is the first filtered temperature data, first filtered humidity data, or first filtered light intensity data after the previous filtering.

[0019] The beneficial effects of the above-mentioned further scheme are as follows: the difference between the latest data and the median data is used as the weight, and based on the median data in the sliding block, the influence of all data in the sliding block and the previous filtering χ′ is considered to comprehensively filter the latest input data, so that the filtered data maintains the overall trend of the real data and achieves the effect of filtering out outliers.

[0020] Furthermore, in step S2, the curtain opening model adopts a BP neural network that has been trained.

[0021] Furthermore, the formula for updating the weights of the BP neural network during training is as follows:

[0022]

[0023]

[0024]

[0025] Among them, w t+1 Let w be the weights of the BP neural network trained in the (t+1)th iteration. t Let η′ be the weights of the BP neural network trained for the t-th time. t Let η′ be the first step length of the BP neural network trained for the t-th time. t-1 Let η″ be the first step length of the BP neural network trained for the (t-1)th time. t Let y be the second step size of the BP neural network trained for the tth time, L be the loss function, and y be the second step size of the BP neural network trained for the tth time. t Let y be the output of the BP neural network trained for the tth time. * Let α be the target opening value of the curtain during the training process, th be the adjustment factor, and th be the error threshold.

[0026] The beneficial effect of the above further scheme is that, using the error threshold th as the boundary for constant training, in the initial stage y t -y * The difference is large, therefore, the step size is updated using a first-step long formula, increasing the step size with each training iteration to cause the weights to decrease rapidly. Simultaneously, an adjustment factor is set to regulate the degree of weight decrease; in the later stages of training, y... t -y * The difference is small; therefore, the second step size formula is used to update the step size, reducing the step size with each training iteration. Simultaneously, through... The step size is adjusted to be smaller when the difference is small and larger when the difference is large, so as to gradually approach the target value.

[0027] Further, step S3 specifically involves inputting the difference between the temperature data output by the air conditioning temperature control model and the temperature data collected in the current room into the air conditioning temperature control model to obtain the air conditioning temperature value used to control the air conditioning temperature.

[0028] The beneficial effect of the above-mentioned further solution is as follows: There is a temperature difference between the temperature data output by the air conditioning temperature control model and the current indoor temperature. Therefore, if the temperature data output by the air conditioning temperature control model is used to control the temperature data output by the air conditioning temperature control model, it will be difficult to keep the indoor temperature constant and it will often be in a state of fluctuation. Therefore, the difference between the temperature data output by the air conditioning temperature control model and the current indoor temperature data should be used as the input of the air conditioning temperature control model, so as to control the temperature data output by the air conditioning temperature control model and keep the indoor temperature constant.

[0029] Furthermore, the air conditioning temperature control model in step S3 is as follows:

[0030]

[0031] Among them, y j The air conditioning temperature value used to regulate the air conditioning temperature. Let be the difference between the temperature data output by the air conditioning temperature control model in the j-th iteration and the indoor temperature data collected in the j-th iteration. J represents the difference between the temperature data output by the air conditioning temperature control model in the (j-1)th iteration and the temperature data collected indoors in the (j-1)th iteration. J is the number of differences, θ1 is the first weighting parameter, θ2 is the second weighting parameter, and θ3 is the third weighting parameter.

[0032] The beneficial effect of the above-mentioned further scheme is that the difference between the first weight parameter θ1 and the temperature data output by the air conditioning temperature control model in the jth time and the temperature data collected indoors in the jth time is... Multiplication is used to measure differences. With air conditioning temperature value y j In the proportional case, multiple historical differences are summed. And multiply by the second weighting parameter θ2 to make the air conditioning temperature value y j Stable, through This helps to reflect the trend of the difference and accelerate the response speed.

[0033] Furthermore, the GA genetic algorithm is used to train the first weight parameter θ1, the second weight parameter θ2, and the third weight parameter θ3.

[0034] The beneficial effect of the above-mentioned further scheme is that the first weight parameter θ1, the second weight parameter θ2 and the third weight parameter θ3 are quickly iterated by the GA genetic algorithm.

[0035] Furthermore, the training process of the GA genetic algorithm includes the following steps:

[0036] S31. Construct a weight parameter vector from the first weight parameter θ1, the second weight parameter θ2, and the third weight parameter θ3;

[0037] S32. Randomly generate initial values ​​for the elements in the weight parameter vector to obtain the k-th iteration vector, with k initially set to 1.

[0038] S33. Calculate the fitness of the k-th iteration vector based on the element values ​​in the k-th iteration vector;

[0039] S34. If the fitness of the k-th iteration vector is less than the threshold, then training ends. The element value in the current k-th iteration vector is the weight parameter. If it is greater than the threshold, then jump to step S35.

[0040] S35. Calculate the crossover probability and mutation probability based on the fitness of the vector in the k-th iteration.

[0041] S36. Perform crossover and mutation operations on the elements in the k-th iteration vector according to the crossover probability and mutation probability to obtain the next iteration vector, increment k by 1, and jump to step S33.

[0042] The beneficial effects of the above-mentioned further scheme are: the iteration is measured by fitness, thereby stopping or continuing the iteration; the element values ​​are continuously changed by crossover probability and mutation probability until the fitness is satisfied.

[0043] An indoor temperature, humidity and light control system includes: a light sensor, a temperature and humidity sensor, a wireless transmission module, a controller, an air conditioner and curtains;

[0044] The light sensor is used to collect initial light intensity data; the temperature and humidity sensor is used to collect initial temperature data and initial humidity data; the wireless transmission module is used to transmit the initial temperature data, initial humidity data, and initial light intensity data to the controller; the controller is used to execute the above-mentioned indoor temperature, humidity, and light control methods to obtain the curtain opening value, air conditioner temperature value, and humidity value, and adjust the variable frequency air compressor of the air conditioner according to the air conditioner temperature value, adjust the dehumidification unit of the air conditioner according to the humidity value, and regulate the opening and closing degree of the curtains according to the curtain opening value.

[0045] The beneficial effects of the system of the present invention are as follows: light intensity data, temperature data and humidity data are collected by light sensor and temperature and humidity sensor, and the above-mentioned indoor temperature, humidity and light control method is executed by controller. By obtaining the curtain opening value, air conditioning temperature value and humidity value, the corresponding device is driven to achieve unified control of indoor temperature, humidity and light, so that the indoor people are always in a suitable range of temperature, humidity and light. Attached Figure Description

[0046] Figure 1 A flowchart of a method for controlling indoor temperature, humidity and light;

[0047] Figure 2This is a system block diagram of an indoor temperature, humidity, and light control system. Detailed Implementation

[0048] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0049] Example 1:

[0050] like Figure 1 As shown, a method for controlling indoor temperature, humidity, and light includes the following steps:

[0051] S1. Collect and preprocess indoor temperature, humidity and light intensity to obtain preprocessed temperature data, humidity data and light intensity data;

[0052] S2. Process the light intensity data using the curtain opening model to obtain the curtain opening value;

[0053] S3. Use an air conditioning temperature control model to process the temperature data and obtain the air conditioning temperature value;

[0054] S4. Obtain the humidity value based on the humidity data;

[0055] S5. Adjust the air conditioner and curtains according to the air conditioner temperature value, curtain opening value, and humidity value respectively.

[0056] The beneficial effects of Embodiment 1 of the present invention are as follows: by collecting indoor temperature, humidity and light intensity in real time, and processing them through a curtain opening model and an air conditioning temperature control model respectively, specific curtain opening values ​​and air conditioning temperature values ​​are obtained. Then, through humidity data, specific humidity values ​​are obtained. By using specific curtain opening values, air conditioning temperature values ​​and humidity values, precise control of air conditioning and curtains is achieved, which solves the problem that there is a lack of a unified automatic indoor temperature, humidity and light control method in the prior art.

[0057] In this embodiment, the humidity data from the sensor can be referenced to obtain a specific humidity value. After obtaining the specific humidity value, it is determined whether the humidity value is greater than a threshold. If so, the dehumidification unit of the air conditioner is turned on to dehumidify.

[0058] Example 2:

[0059] Step S1 in Embodiment 1 includes the following sub-steps:

[0060] S11. Collect indoor temperature, humidity and light intensity data to obtain initial temperature data, initial humidity data and initial light intensity data;

[0061] S12. Apply amplitude limiting filtering to remove outliers from the initial temperature data, initial humidity data and initial light intensity data to obtain the first filtered temperature data, the first filtered humidity data and the first filtered light intensity data.

[0062] S13. The temperature data, humidity data, and light intensity data filtered in the first filtering are filtered again using the moving average filter to obtain the preprocessed temperature data, humidity data, and light intensity data.

[0063] The beneficial effects of Example 2 are as follows: the first filtering removes out-of-range values, and the second filtering makes the overall data conform to the actual temperature, humidity and light intensity trends, eliminating the influence of individual outliers.

[0064] Example 3:

[0065] The filtering formula in step S13 of Example 2 is as follows:

[0066]

[0067] Where χ represents the filtered temperature, humidity, or light intensity data, x new For the latest data in the moving average filter's moving block, x med For the median data in the moving average filter sliding block, x max For the largest first-filtered temperature data, first-filtered humidity data, or first-filtered light intensity data, x min For the minimum first-filter temperature data, first-filter humidity data, or first-filter light intensity data, x i For the i-th first filtered temperature data, first filtered humidity data, or first filtered light intensity data in the sliding block of the moving average filter, N is the length of the sliding block, and χ′ is the first filtered temperature data, first filtered humidity data, or first filtered light intensity data after the previous filtering.

[0068] In this embodiment, the filtering formula can only filter data of the same type. When filtering temperature data, the filtering formula is used to filter the temperature data corresponding to the meaning of all characters in the formula. When filtering humidity data, the filtering formula is used to filter the humidity data corresponding to the meaning of all characters in the formula. When filtering light intensity data, the filtering formula is used to filter the light intensity data corresponding to the meaning of all characters in the formula.

[0069] The beneficial effect of Example 3 is that the difference between the latest data and the median data is used as the weight, and based on the median data in the sliding block, the influence of all data in the sliding block and the previous filtered χ′ is considered to comprehensively filter the latest input data, so that the filtered data maintains the overall trend of the real data and achieves the effect of filtering out outliers.

[0070] Example 4:

[0071] In Example 1, the curtain opening model is implemented using a trained BP neural network.

[0072] The formula for updating the weights of the BP neural network during training is as follows:

[0073]

[0074]

[0075]

[0076] Among them, w t+1 Let w be the weights of the BP neural network trained in the (t+1)th iteration. t Let η′ be the weights of the BP neural network trained for the t-th time. t Let η′ be the first step length of the BP neural network trained for the t-th time. t-1 Let η″ be the first step length of the BP neural network trained for the (t-1)th time. t Let y be the second step size of the BP neural network trained for the tth time, L be the loss function, and y be the second step size of the BP neural network trained for the tth time. t Let y be the output of the BP neural network trained for the tth time. * Let α be the target opening value of the curtain during the training process, th be the adjustment factor, and th be the error threshold.

[0077] The beneficial effect of Example 4 is that, using the error threshold th as the boundary for constant training, in the initial stage y t -y * The difference is large, therefore, the step size is updated using a first-step long formula, increasing the step size with each training iteration to cause the weights to decrease rapidly. Simultaneously, an adjustment factor is set to regulate the degree of weight decrease; in the later stages of training, y... t -y * The difference is small; therefore, the second step size formula is used to update the step size, reducing the step size with each training iteration. Simultaneously, through... The step size is adjusted to be smaller when the difference is small and larger when the difference is large, so as to gradually approach the target value.

[0078] Example 5:

[0079] Specifically, step S3 in Example 1 involves inputting the difference between the temperature data output by the air conditioning temperature control model and the temperature data collected in the room into the air conditioning temperature control model to obtain the air conditioning temperature value used to control the air conditioning temperature.

[0080] The beneficial effect of Example 5 is as follows: There is a temperature difference between the temperature data output by the air conditioning temperature control model and the current indoor temperature. Therefore, if the temperature data output by the air conditioning temperature control model is used to control the temperature data output by the air conditioning temperature control model, it will be difficult to keep the indoor temperature constant and it will often fluctuate. Therefore, the difference between the temperature data output by the air conditioning temperature control model and the current indoor temperature data should be used as the input of the air conditioning temperature control model, so as to control the temperature data output by the air conditioning temperature control model and keep the indoor temperature constant.

[0081] Example 6:

[0082] The air conditioning temperature control model in step S3 of Example 1 is as follows:

[0083]

[0084] Among them, y j The air conditioning temperature value used to regulate the air conditioning temperature. Let be the difference between the temperature data output by the air conditioning temperature control model in the j-th iteration and the indoor temperature data collected in the j-th iteration. J represents the difference between the temperature data output by the air conditioning temperature control model in the (j-1)th iteration and the temperature data collected indoors in the (j-1)th iteration. J is the number of differences, θ1 is the first weighting parameter, θ2 is the second weighting parameter, and θ3 is the third weighting parameter.

[0085] The beneficial effect of Example 6 is that the difference between the first weight parameter θ1 and the temperature data output by the air conditioning temperature control model in the jth instance and the temperature data collected indoors in the jth instance is... Multiplication is used to measure differences. With air conditioning temperature value y j In the proportional case, multiple historical differences are summed. And multiply by the second weighting parameter θ2 to make the air conditioning temperature value y j Stable, through This helps to reflect the trend of the difference and accelerate the response speed.

[0086] Example 7:

[0087] For Example 6, the GA genetic algorithm is used to train the first weight parameter θ1, the second weight parameter θ2, and the third weight parameter θ3.

[0088] The beneficial effect of Example 7 is that the first weight parameter θ1, the second weight parameter θ2, and the third weight parameter θ3 are obtained quickly through the GA genetic algorithm.

[0089] Example 8:

[0090] For Example 7, the training process of the GA genetic algorithm includes the following steps:

[0091] S31. Construct a weight parameter vector from the first weight parameter θ1, the second weight parameter θ2, and the third weight parameter θ3;

[0092] S32. Randomly generate initial values ​​for the elements in the weight parameter vector to obtain the k-th iteration vector, with k initially set to 1.

[0093] S33. Calculate the fitness of the k-th iteration vector based on the element values ​​in the k-th iteration vector;

[0094] S34. If the fitness of the k-th iteration vector is less than the threshold, then training ends. The element value in the current k-th iteration vector is the weight parameter. If it is greater than the threshold, then jump to step S35.

[0095] S35. Calculate the crossover probability and mutation probability based on the fitness of the vector in the k-th iteration.

[0096] S36. Perform crossover and mutation operations on the elements in the k-th iteration vector according to the crossover probability and mutation probability to obtain the next iteration vector, increment k by 1, and jump to step S33.

[0097] The beneficial effect of Example 8 is that the iteration is measured by fitness, thereby stopping or continuing the iteration, and the element values ​​are continuously changed by crossover probability and mutation probability until the fitness is met.

[0098] Example 9:

[0099] Regarding Embodiment 8 above, the formulas for the crossover probability and mutation probability in step S35 are as follows:

[0100]

[0101]

[0102] Where ζ1 is the crossover probability, f max f is the maximum fitness value min To minimize fitness, ζ2 is the average fitness, f1 is the larger fitness among the vector elements in the k-th iteration, K is the set total number of iterations, ζ2 is the mutation probability, and f2 is the fitness of the mutated element.

[0103] The beneficial effects of Example 9 are as follows: the designed crossover probability ζ1 can accelerate the parameter evolution speed in the early stage, while the crossover probability ζ1 is smaller in the later stage, which makes it converge smoothly. The mutation probability ζ2 is larger in the early stage, which can ensure sufficient mutation space to make the data mutate. The mutation probability ζ2 is smaller in the later stage of iteration, so that the excellent individuals are not mutated. At the same time, the mutation probability is corrected by the fluctuation range of the fitness f2 of the mutated elements, which ensures the stability of mutation.

[0104] Example 10:

[0105] In accordance with embodiments 1 to 9 above, an indoor temperature, humidity and light control system is proposed, comprising: a light sensor, a temperature and humidity sensor, a wireless transmission module, a controller, an air conditioner and curtains;

[0106] The light sensor is used to collect initial light intensity data; the temperature and humidity sensor is used to collect initial temperature data and initial humidity data; the wireless transmission module is used to transmit the initial temperature data, initial humidity data, and initial light intensity data to the controller; the controller is used to execute the indoor temperature, humidity, and light control method as described in any of the embodiments 1 to 9, to obtain the curtain opening value, the air conditioner temperature value, and the humidity value, and to adjust the variable frequency air compressor of the air conditioner according to the air conditioner temperature value, adjust the dehumidification unit of the air conditioner according to the humidity value, and control the opening and closing degree of the curtains according to the curtain opening value.

[0107] The beneficial effects of Example 10 are as follows: by collecting light intensity data, temperature data and humidity data through light sensor and temperature and humidity sensor, the controller executes the above-mentioned indoor temperature, humidity and light control method. By obtaining the curtain opening value, air conditioning temperature value and humidity value, the corresponding equipment is driven to achieve unified control of indoor temperature, humidity and light, so that the indoor people are always in a suitable range of temperature, humidity and light.

[0108] In this embodiment, the wireless transmission module is of the LoRa type. An infrared human body sensor is also installed indoors to detect whether people have entered the room, facilitating the activation of the entire system. Simultaneously, multiple terminals can be set up: a video equipment control terminal, an image control terminal, a conference screen control terminal, a lighting control terminal, a curtain control terminal, and an air conditioning control terminal. A control tablet computer can also be installed on one side of the controller for operating indoor equipment, facilitating integrated management of indoor devices, such as… Figure 2 As shown.

Claims

1. A method for controlling indoor temperature, humidity, and light, characterized in that, Includes the following steps: S1. Collect and preprocess indoor temperature, humidity and light intensity to obtain preprocessed temperature data, humidity data and light intensity data; S2. Process the light intensity data using the curtain opening model to obtain the curtain opening value; The curtain opening model in S2 uses a trained BP neural network. The formula for updating the weights of the BP neural network during training is as follows: in, For the first The weights of the BP neural network trained in this cycle. For the first The weights of the BP neural network trained in this cycle. For the first The first step of the BP neural network trained in this cycle is long. For the first The first step of the BP neural network trained in this cycle is long. For the first The second step length of the BP neural network trained in this iteration. For loss function, For the first The output of the BP neural network after training. The target opening value of the curtains during the training process. As a regulating factor, This is the error threshold; S3. Use an air conditioning temperature control model to process the temperature data and obtain the air conditioning temperature value; S4. Obtain the humidity value based on the humidity data; S5. Adjust the air conditioner and curtains according to the air conditioner temperature value, curtain opening value, and humidity value respectively.

2. The indoor temperature, humidity, and light control method according to claim 1, characterized in that, S1 includes the following steps: S11. Collect indoor temperature, humidity and light intensity data to obtain initial temperature data, initial humidity data and initial light intensity data; S12. Apply amplitude limiting filtering to remove outliers from the initial temperature data, initial humidity data and initial light intensity data to obtain the first filtered temperature data, the first filtered humidity data and the first filtered light intensity data. S13. The temperature data, humidity data, and light intensity data filtered in the first filtering are filtered again using the moving average filter to obtain the preprocessed temperature data, humidity data, and light intensity data.

3. The indoor temperature, humidity, and light control method according to claim 2, characterized in that, The filtering formula in S13 is as follows: in, This refers to filtered temperature, humidity, or light intensity data. The latest data in the moving average filter sliding block, The median data in the moving average filter sliding block. For the highest first-filter temperature data, first-filter humidity data, or first-filter light intensity data, The minimum first-filter temperature data, first-filter humidity data, or first-filter light intensity data. The first sliding block in the moving average filter The first filtered temperature data, the first filtered humidity data, or the first filtered light intensity data. The length of the slider. This refers to the temperature data, humidity data, or light intensity data from the first filter after the previous filter.

4. The indoor temperature, humidity, and light control method according to claim 1, characterized in that, Specifically, S3 involves inputting the difference between the temperature data output by the air conditioning temperature control model and the temperature data collected indoors into the air conditioning temperature control model to obtain the air conditioning temperature value used to control the air conditioning temperature.

5. The indoor temperature, humidity, and light control method according to claim 1, characterized in that, The air conditioning temperature control model in S3 is as follows: in, The air conditioning temperature value used to regulate the air conditioning temperature. For the first The temperature data output by the air conditioning temperature control model of the first time is the same as that of the second time. The difference in temperature data collected in the second indoor unit For the first The temperature data output by the air conditioning temperature control model of the first time is the same as that of the second time. The difference in temperature data collected in the second indoor unit The number of differences. As the first weight parameter, This is the second weighting parameter. This is the third weighting parameter.

6. The indoor temperature, humidity, and light control method according to claim 5, characterized in that, The first weight parameter is selected using the GA genetic algorithm. Second weighting parameter and the third weight parameter Conduct training.

7. The indoor temperature, humidity, and light control method according to claim 6, characterized in that, The training process of the GA genetic algorithm includes the following steps: S31, Set the first weight parameter Second weighting parameter and the third weight parameter , construct as a weight parameter vector; S32. Randomly generate initial values ​​for the elements in the weight parameter vector to obtain the k-th iteration vector, with k initially set to 1. S33. Calculate the fitness of the k-th iteration vector based on the element values ​​in the k-th iteration vector; S34. If the fitness of the k-th iteration vector is less than the threshold, then training ends. The element value in the current k-th iteration vector is the weight parameter. If it is greater than the threshold, then jump to step S35. S35. Calculate the crossover probability and mutation probability based on the fitness of the vector in the k-th iteration. S36. Perform crossover and mutation operations on the elements in the k-th iteration vector according to the crossover probability and mutation probability to obtain the next iteration vector, increment k by 1, and jump to step S33.

8. An indoor temperature, humidity, and light control system, characterized in that, include: Light sensor, temperature and humidity sensor, wireless transmission module, controller, air conditioner and curtains; The light sensor is used to collect initial light intensity data; the temperature and humidity sensor is used to collect initial temperature data and initial humidity data; the wireless transmission module is used to transmit the initial temperature data, initial humidity data, and initial light intensity data to the controller; the controller is used to execute the indoor temperature, humidity, and light control method as described in any one of claims 1 to 7, to obtain the curtain opening value, the air conditioner temperature value, and the humidity value, and to adjust the variable frequency air compressor of the air conditioner according to the air conditioner temperature value, to adjust the dehumidification unit of the air conditioner according to the humidity value, and to control the opening and closing degree of the curtains according to the curtain opening value.

Images