Smart home control system and working method thereof

Abstract

The invention relates to the technical field of Internet of Things and smart home, in particular to a smart home control system and a working method thereof, and the method comprises the following steps: S1, collecting environment information in real time through sensor equipment installed in a home environment, and carrying out the primary processing of the environment information, including temperature, humidity and illuminance; s2, collecting user behavior information in real time, uploading the user behavior information to a server, and analyzing user activities according to the user behavior data; s3, based on the data collected in the S1 and the S2, generating an optimal equipment control strategy by using a predefined adaptive algorithm; s4, selecting an optimal equipment control strategy according to the user activity, and sending a control scheme to each household equipment through the Internet of Things gateway to realize intelligent adjustment of the household equipment; and S5, collecting user use feedback, and optimizing equipment control parameters according to feedback data.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things and smart home, and in particular to a smart home control system and a working method thereof. Background Art

[0002] In recent years, with the rapid development of Internet of Things technology, smart home systems have gradually become an important part of modern life. The core goal of smart home is to achieve automatic adjustment and efficient management of home environment through the interconnection of devices. However, traditional home systems often rely on manual operation or simple preset programs, and cannot respond to user needs and dynamic changes in the environment in a timely manner. This not only reduces the user's life experience, but also may cause energy waste.

[0003] Therefore, a smart home control system and a working method thereof are proposed. Summary of the invention

[0004] The present invention aims to solve the problem that most traditional home systems currently rely on manual operations or preset programs by users and are unable to perform intelligent adjustments based on real-time environment and user activities. In addition, the devices in many systems operate independently of each other, lack coordination and linkage, and cannot optimize device operation according to overall needs. In existing systems, user feedback information is often not effectively used for system strategy optimization, resulting in a poor user experience.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] A smart home control working method, characterized in that it includes the following steps:

[0007] S1: Through the sensor devices installed in the home environment, environmental information is collected in real time and processed preliminarily, including temperature, humidity and light intensity;

[0008] S2: collects user behavior information in real time and uploads it to the server, and is used to analyze user activities based on user behavior data;

[0009] S3: Based on the data collected in S1 and S2, the optimal equipment control strategy is generated using a predefined adaptive algorithm;

[0010] S4: Select the optimal device control strategy based on user activities, and send the control plan to each home device through the IoT gateway to achieve intelligent adjustment of home devices;

[0011] S5: Collect user feedback and optimize device control parameters based on the feedback data.

[0012] Optionally, the S1 specifically includes:

[0013] S11: Zigbee modules, power detectors, high-definition cameras, temperature and humidity sensors, and light sensors are configured in the home environment. All sensors are connected to the central control system through wireless communication protocols.

[0014] S12: Collecting environmental parameter data. The temperature sensor uses thermocouple or thermistor technology to measure environmental temperature data according to the resistance change caused by temperature change. The humidity sensor uses capacitive technology to detect the water vapor content in the air by measuring the change in dielectric constant to obtain humidity data. The light intensity sensor uses a photodiode to convert the received light intensity into an electrical signal output to measure the light intensity data.

[0015] S14: The Zigbee module is used to periodically send the collected data to the central control system, and integrate, filter and denoise the data.

[0016] Optionally, the S2 specifically includes:

[0017] S21: Install high-definition cameras in the home environment and connect power detectors to home devices. All devices are connected to the central control system through wireless communication protocols;

[0018] S22: Collecting user activity parameter data, the power detector is used to record the number and time of appliance usage; the high-definition camera is used to record user activity images;

[0019] S23: The central control system extracts the user activity time according to the user activity screen collected in S22 according to the timestamp alignment algorithm;

[0020] S24: The central control system calculates the user activity coefficient based on the number and time of appliance usage collected in S22, the user activity time extracted in S23, and the preset standard activity time.

[0021] Optionally, the S24 specifically includes:

[0022] S241: The formula for calculating the user activity coefficient according to the number and time of appliance use, user activity time, and preset standard activity data is:

[0023]

[0024] Among them, A c is the user activity coefficient, N is the number of times the appliance is used, T e T is the total time the appliance is used, in min. sta is the user activity time, in units of min, T s is the standard activity time, its unit is min, α and β are weight coefficients.

[0025] Optionally, the S3 specifically includes:

[0026] S31: pre-processing the environmental parameter data and user activity parameter data collected in S1 and S2, including data cleaning, denoising and standardization;

[0027] S32: extracting feature parameters from the preprocessed data using a feature extraction algorithm, including ambient temperature, humidity, light intensity, user activity time, and the number and time of appliance usage;

[0028] S33: Based on the extracted feature parameters, an adaptive control model is trained using historical data and a supervised learning algorithm to predict the optimal control strategy based on the input environment and user activity data;

[0029] S34: Input the environmental parameter data and user activity parameter data collected in real time into the trained adaptive control model. The adaptive control model calculates and generates the optimal joint control strategy based on the real-time input data. The joint control strategy includes the operating status and adjustment parameters of each smart device.

[0030] Optionally, the S32 specifically includes:

[0031] S321: Temperature feature extraction: For the preprocessed temperature data, a weighted moving average algorithm is used to calculate the average temperature value over a period of time. The formula is:

[0032]

[0033] Among them, T i is the temperature data at the i-th moment, in degrees Celsius, W i represents the weight of the i-th time point, n represents the size of the sliding window, is the weighted moving average temperature of the jth segment, in degrees Celsius;

[0034] S322: Humidity feature extraction: for the pre-processed humidity data, the exponential moving average algorithm is used to calculate the average humidity value over a period of time. The formula is:

[0035]

[0036] in, is the average humidity value at the tth time point, and its unit is %, H t represents the humidity data at the tth time point, its unit is %, and α is the smoothing coefficient;

[0037] S323: Light feature extraction: for the pre-processed light data, the moving median algorithm is used to calculate the fluctuation of light intensity. The formula is:

[0038]

[0039] Among them, L j represents the illumination data at the jth time point, in units of lux, and n is the sliding window size. is the moving median illuminance of the jth segment, and its unit is lux.

[0040] Optionally, the S33 specifically includes:

[0041] S331: extracting preprocessed feature parameters from the historical data set to form a training data set and a verification data set, the feature parameters including temperature, humidity, illuminance, user activity time, number of times and time of appliance use;

[0042] S332: Feature vector construction, combining the feature parameter values ​​at each time point into a feature vector: S331: Extracting preprocessed feature parameters from the historical data set to form a training data set and a validation data set, the feature parameters include temperature, humidity, illuminance, user activity time, number of times and time of appliance use;

[0043] S332: Feature vector construction, combining the feature parameter values ​​at each time point into a feature vector:

[0044]

[0045] S333: Define the target variable according to the historical control strategy, which represents the optimal control strategy corresponding to each feature vector: Y i =OptimalControl(X i );

[0046] S334: Select the multi-layer perceptron algorithm in the supervised learning algorithm for model training. The model structure of the multi-layer perceptron is:

[0047]

[0048] in, represents the control strategy predicted by the model, M represents the multilayer perceptron model, and θ represents the model parameters;

[0049] S335: Use mean square error as the loss function to evaluate the error between the control strategy predicted by the model and the target variable. The formula is:

[0050]

[0051] in, represents the control strategy predicted by the model, Y i represents the optimal control strategy corresponding to each eigenvector, and n represents the number of samples;

[0052] S336: Use the back propagation algorithm and gradient descent algorithm to minimize the loss function and update the model parameters. The formula is:

[0053]

[0054] Where θ is the model parameter, η is the learning rate, is the gradient of the loss function with respect to the model parameters θ;

[0055] S337: Use the validation data set to evaluate the performance of the trained model and optimize the model by adjusting the hyperparameters until the validation error of the model is minimized. The formula is:

[0056]

[0057] Among them, m is the number of samples in the validation set, is the predicted value of the validation set, Y j val is the true value of the validation set.

[0058] Optionally, the S4 specifically includes:

[0059] S41: Analyze the optimal control strategy generated by S3, extract the specific control parameters and status information of each smart device, and form a control vector:

[0060] C i =[C1, C2, ..., C m ];

[0061] Among them C i represents the control parameters of the i-th device;

[0062] S42: Calculate the matching degree between the user activity coefficient generated by S2 and the optimal control strategy. The formula is:

[0063]

[0064] Where S is the matching degree between the user activity coefficient and the optimal control strategy, C i represents the control parameters of the ith device, A c is the user activity coefficient, W i is the weight coefficient of the i-th device; W j is the weight coefficient of the i-1th device; Y i is the optimal control strategy corresponding to each eigenvector;

[0065] S43: Select the optimal control strategy with the highest matching degree, control and adjust the device according to the Internet of Things, and select the strategy method of the scene mode with the highest matching degree as follows:

[0066] S=[S1,S2,...,S n ];

[0067]

[0068] Where S is the matching degree between the user activity coefficient and the optimal control strategy, n is the number of candidate control strategies; Y i * It is the optimal control strategy with the highest matching degree between the user activity coefficient and the optimal control strategy.

[0069] Optionally, the S5 specifically includes:

[0070] S51: configuring mobile terminal communication, collecting user feedback information; and classifying the user feedback information into positive feedback and negative feedback;

[0071] S52: Optimize the device control parameters according to the user feedback type. The formula is:

[0072]

[0073] in is the optimized control parameter of the ith device, C i The current control parameters of the i-th device; is the adjustment factor for positive feedback, is the adjustment coefficient of negative feedback, and k is the feedback weight coefficient.

[0074] The present invention also provides a smart home control system for implementing the smart home control method as described above, which is characterized by comprising:

[0075] The environmental monitoring module is used to collect and initially process environmental information in real time, including temperature, humidity and light intensity, through sensor devices installed in the home environment;

[0076] User data collection and analysis module, used to collect user behavior information in real time and upload it to the server; and used to analyze user activities based on user behavior data;

[0077] A data processing module is used to generate an optimal device control strategy using a predefined adaptive algorithm after receiving temperature data, humidity data, illumination data and user behavior data;

[0078] The device control module is used to select the optimal device control strategy based on user activities; and is used to send the control plan to each home device through the IoT gateway to achieve intelligent adjustment of home devices;

[0079] The feedback optimization module is used to collect user feedback and optimize device control parameters based on the feedback data.

[0080] Beneficial effects of the present invention:

[0081] 1. The system can perceive the changes in the home environment in real time, provide reliable basic data for the subsequent generation of control strategies, and reduce the interference of noise and abnormal data by preliminary data processing, thereby improving the accuracy of the control strategy.

[0082] 2. The system can judge the user's current needs based on the user's activity frequency and habits, and improve the device's responsiveness. The collection and analysis of user behavior data enables the system to personalize home appliances.

[0083] 3. The adaptive algorithm can continuously adjust the control strategy according to real-time data, so that the system can provide the best equipment configuration under different environments and user conditions, improve the energy efficiency of the equipment and the comfort of the user, and reduce the manual operation of the user.

[0084] 4. The system continuously learns from user feedback, continuously optimizes control parameters, and provides more personalized services, which can not only improve user satisfaction, but also avoid energy waste caused by unreasonable equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the technical solution of the real-time example of the present invention, the following briefly introduces the drawings of the real-time example. The drawings described below only relate to some real-time examples of the present invention, but are not intended to limit the present invention.

[0086] In the attached picture:

[0087] Figure 1 Schematic diagram of a smart home control system and its working method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0088] The following will clearly and completely describe the concept of the present invention and the technical effects produced in combination with real-time examples to fully understand the purpose, characteristics and effects of the present invention.

[0089] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0090] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0091] See also Figure 1 ,

[0092] A smart home control working method comprises the following steps:

[0093] S1: Through the sensor devices installed in the home environment, environmental information is collected in real time and processed preliminarily, including temperature, humidity and light intensity;

[0094] S2: collects user behavior information in real time and uploads it to the server, and is used to analyze user activities based on user behavior data;

[0095] S3: Based on the data collected in S1 and S2, the optimal equipment control strategy is generated using a predefined adaptive algorithm;

[0096] S4: Select the optimal device control strategy based on user activities, and send the control plan to each home device through the IoT gateway to achieve intelligent adjustment of home devices;

[0097] S5: Collect user feedback and optimize device control parameters based on the feedback data.

[0098] The S1 specifically includes:

[0099] S11: Zigbee modules, power detectors, high-definition cameras, temperature and humidity sensors, and light sensors are configured in the home environment. All sensors are connected to the central control system through wireless communication protocols.

[0100] S12: Collecting environmental parameter data. The temperature sensor uses thermocouple or thermistor technology to measure environmental temperature data according to the resistance change caused by temperature change. The humidity sensor uses capacitive technology to detect the water vapor content in the air by measuring the change in dielectric constant to obtain humidity data. The light intensity sensor uses a photodiode to convert the received light intensity into an electrical signal output to measure the light intensity data.

[0101] S14: The Zigbee module is used to periodically send the collected data to the central control system, and integrate, filter and denoise the data;

[0102] Through the combination of the above steps, environmental parameters such as temperature, humidity and light can be accurately sensed, providing a reliable data basis for the control strategy. Through the combination of multiple sensors, comprehensive home environment monitoring can be achieved, which can realize wireless transmission of sensor data, reduce wiring costs and complexity, and improve the flexibility of system deployment.

[0103] S2 specifically includes:

[0104] S21: Install high-definition cameras in the home environment and connect power detectors to home devices. All devices are connected to the central control system through wireless communication protocols;

[0105] S22: Collecting user activity parameter data, the power detector is used to record the number and time of appliance usage; the high-definition camera is used to record user activity images;

[0106] S23: The central control system extracts the user activity time according to the user activity screen collected in S22 according to the timestamp alignment algorithm;

[0107] S24: The central control system calculates the user activity coefficient according to the number and time of appliance usage collected in S22 and the user activity time extracted in S23 and the preset standard activity time;

[0108] By combining the above steps, the user's activity status and the use of electrical appliances can be monitored in real time, providing data support for device control strategies. The use of wireless communication protocols improves the flexibility of the equipment and the stability of the system. By extracting the correspondence between the user's activity time and the length of time the appliance is used, the system can understand the user's needs and habits more accurately. The user activity coefficient provides a quantitative assessment of the user's activity level, helping the system adjust the device status according to the user's needs.

[0109] S24 specifically includes:

[0110] S241: The formula for calculating the user activity coefficient according to the number and time of appliance use, user activity time, and preset standard activity data is:

[0111]

[0112] Among them, A c is the user activity coefficient, N is the number of times the appliance is used, T e T is the total time the appliance is used, in min. sta is the user activity time, in units of min, T s is the standard activity time, its unit is min, α and β are weight coefficients, 0<α<1, 0<β<1, α+β=1.

[0113] S3 specifically includes:

[0114] S31: pre-processing the environmental parameter data and user activity parameter data collected in S1 and S2, including data cleaning, denoising and standardization;

[0115] S32: extracting feature parameters from the preprocessed data using a feature extraction algorithm, including ambient temperature, humidity, light intensity, user activity time, and the number and time of appliance usage;

[0116] S33: Based on the extracted feature parameters, an adaptive control model is trained using historical data and a supervised learning algorithm to predict the optimal control strategy based on the input environment and user activity data;

[0117] S34: inputting the real-time collected environmental parameter data and user activity parameter data into the trained adaptive control model, and the adaptive control model calculates and generates an optimal joint control strategy based on the real-time input data, and the joint control strategy includes the operating status and adjustment parameters of each smart device;

[0118] By combining the above steps, errors and outliers can be reduced, the reliability of model predictions can be improved, and the extracted feature parameters provide a complete data basis for the intelligent control model. The system can accurately understand the user's life patterns and equipment usage, and provide support for the formulation of personalized control strategies. At the same time, the system continuously performs incremental learning through new data, so that the model has dynamic optimization capabilities. The system automatically adjusts the device status according to real-time data to achieve intelligent operation. Personalized linkage control enables linkage operation between devices, improving user convenience and equipment energy efficiency.

[0119] S32 specifically includes:

[0120] S321: Temperature feature extraction: For the preprocessed temperature data, a weighted moving average algorithm is used to calculate the average temperature value over a period of time. The formula is:

[0121]

[0122] Among them, T i is the temperature data at the i-th moment, in degrees Celsius, W i Represents the weight of the i-th time point, 0<W i <1, n represents the size of the sliding window, is the weighted moving average temperature of the jth segment, in degrees Celsius;

[0123] S322: Humidity feature extraction: for the pre-processed humidity data, the exponential moving average algorithm is used to calculate the average humidity value over a period of time. The formula is:

[0124]

[0125] in, is the average humidity value at the tth time point, and its unit is %, H t It represents the humidity data at the t-th time point, its unit is %, α is the smoothing coefficient, and the value range of α is 0<α<1;

[0126] S323: Light feature extraction: for the pre-processed light data, the moving median algorithm is used to calculate the fluctuation of light intensity. The formula is:

[0127]

[0128] Among them, L j represents the illumination data at the jth time point, in units of lux, and n is the sliding window size. is the moving median illuminance of the jth segment, and its unit is lux.

[0129] S33 specifically includes:

[0130] S331: extracting preprocessed feature parameters from the historical data set to form a training data set and a verification data set, the feature parameters including temperature, humidity, illuminance, user activity time, number of times and time of appliance use;

[0131] S332: Feature vector construction, combining the feature parameter values ​​at each time point into a feature vector: S331: Extracting preprocessed feature parameters from the historical data set to form a training data set and a validation data set, the feature parameters include temperature, humidity, illuminance, user activity time, number of times and time of appliance use;

[0132] S332: Feature vector construction, combining the feature parameter values ​​at each time point into a feature vector:

[0133]

[0134] S333: Define the target variable according to the historical control strategy, which represents the optimal control strategy corresponding to each feature vector: Y i =OptimalControl(X i );

[0135] S334: Select the multi-layer perceptron algorithm in the supervised learning algorithm for model training. The model structure of the multi-layer perceptron is:

[0136]

[0137] in, represents the control strategy predicted by the model, M represents the multilayer perceptron model, and θ represents the model parameters;

[0138] S335: Use mean square error as the loss function to evaluate the error between the control strategy predicted by the model and the target variable. The formula is:

[0139]

[0140] in, represents the control strategy predicted by the model, Y i represents the optimal control strategy corresponding to each eigenvector, and n represents the number of samples;

[0141] S336: Use the back propagation algorithm and gradient descent algorithm to minimize the loss function and update the model parameters. The formula is:

[0142]

[0143] Where θ is the model parameter, η is the learning rate, is the gradient of the loss function with respect to the model parameters θ;

[0144] S337: Use the validation data set to evaluate the performance of the trained model and optimize the model by adjusting the hyperparameters until the validation error of the model is minimized. The formula is:

[0145]

[0146] Among them, m is the number of samples in the validation set, is the predicted value of the validation set, Y j val is the true value of the validation set.

[0147] These steps ensure that the system can generate efficient and accurate device control strategies based on historical and real-time data to achieve intelligent home management and personalized control.

[0148] S4 specifically includes:

[0149] S41: Analyze the optimal control strategy generated by S3, extract the specific control parameters and status information of each smart device, and form a control vector:

[0150] C i =[C1, C2, ..., C m ];

[0151] Among them C i represents the control parameters of the i-th device;

[0152] S42: Calculate the matching degree between the user activity coefficient generated by S2 and the optimal control strategy. The formula is:

[0153]

[0154] Where S is the matching degree between the user activity coefficient and the optimal control strategy, C i represents the control parameters of the ith device, A c is the user activity coefficient, W i is the weight coefficient of the ith device, W j is the weight coefficient of the i-1th device, 0<W i <1,0<W j <1, W i +W j =1;Y i is the optimal control strategy corresponding to each eigenvector;

[0155] S43: Select the optimal control strategy with the highest matching degree, control and adjust the equipment according to the Internet of Things, ensure that the system adjusts the control strategy in real time according to user activities and device status, and provide users with a more intelligent and convenient home experience.

[0156] S43 specifically includes:

[0157] S431: The strategy for selecting the scene mode with the highest matching degree is:

[0158] S=[S1,S2,…,S n ];

[0159]

[0160] Where S is the matching degree between the user activity coefficient and the optimal control strategy, n is the number of candidate control strategies; Y i * It is the optimal control strategy with the highest matching degree between the user activity coefficient and the optimal control strategy.

[0161] S5 specifically includes:

[0162] S51: configuring mobile terminal communication, collecting user feedback information; and classifying the user feedback information into positive feedback and negative feedback;

[0163] S52: Optimize the device control parameters according to the user feedback type. The formula is:

[0164]

[0165] in is the optimized control parameter of the ith device, C i The current control parameters of the i-th device; is the adjustment factor for positive feedback, is the adjustment coefficient of negative feedback, k is the feedback weight coefficient, and the threshold range of k is 0<k<1. Through real-time user feedback information and classification, it provides a basis for subsequent control parameter optimization. Positive feedback enhances equipment performance, and negative feedback corrects unreasonable control strategies to achieve closed-loop optimization. The system can distinguish different types of feedback and formulate corresponding optimization strategies.

[0166] The present invention also provides a smart home control system, comprising:

[0167] The environmental monitoring module is used to collect and initially process environmental information in real time, including temperature, humidity and light intensity, through sensor devices installed in the home environment;

[0168] User data collection and analysis module, used to collect user behavior information in real time and upload it to the server; and used to analyze user activities based on user behavior data;

[0169] A data processing module is used to generate an optimal device control strategy using a predefined adaptive algorithm after receiving temperature data, humidity data, illumination data and user behavior data;

[0170] The device control module is used to select the optimal device control strategy based on user activities; and is used to send the control plan to each home device through the IoT gateway to achieve intelligent adjustment of home devices;

[0171] The feedback optimization module is used to collect user feedback and optimize device control parameters based on the feedback data.

[0172] In summary, this smart home system has built a complete closed-loop system from data perception, strategy generation, real-time control to feedback optimization through sensor networks, IoT control and machine learning models. Its intelligent management not only improves the quality of life and convenience of users, but also effectively reduces energy consumption and realizes efficient operation and intelligent adjustment of home appliances. By continuously collecting user feedback and performing dynamic optimization, the system has the ability to self-learn and adapt, ensuring that the device control strategy is always consistent with user needs and meets the needs of future smart life.

[0173] The above embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, other embodiments obtained by technical users in this field without paying creative work all fall within the scope of protection of the present invention.

Claims

1. A smart home control working method, characterized in that: The following steps are involved: S1: Through the sensor devices installed in the home environment, environmental information is collected in real time and processed preliminarily, including temperature, humidity and light intensity; S2: collects user behavior information in real time and uploads it to the server, and is used to analyze user activities based on user behavior data; S3: Based on the data collected in S1 and S2, the optimal equipment control strategy is generated using a predefined adaptive algorithm; S4: Select the optimal device control strategy based on user activities, and send the control plan to each home device through the IoT gateway to achieve intelligent adjustment of home devices; S5: Collect user feedback and optimize device control parameters based on the feedback data.

2. A smart home control working method according to claim 1, characterized in that: The S1 specifically includes: S11: Zigbee modules, power detectors, high-definition cameras, temperature and humidity sensors, and light sensors are configured in the home environment. All sensors are connected to the central control system through wireless communication protocols. S12: Collecting environmental parameter data. The temperature sensor uses thermocouple or thermistor technology to measure environmental temperature data according to the resistance change caused by temperature change. The humidity sensor uses capacitive technology to detect the water vapor content in the air by measuring the change in dielectric constant to obtain humidity data. The light intensity sensor uses a photodiode to convert the received light intensity into an electrical signal output to measure the light intensity data. S14: The Zigbee module is used to periodically send the collected data to the central control system, and integrate, filter and denoise the data.

3. The smart home control method according to claim 1, characterized in that: The S2 specifically includes: S21: Install high-definition cameras in the home environment and connect power detectors to home devices. All devices are connected to the central control system through wireless communication protocols; S22: Collecting user activity parameter data, the power detector is used to record the number and time of appliance usage; the high-definition camera is used to record user activity images; S23: The central control system extracts the user activity time according to the user activity screen collected in S22 according to the timestamp alignment algorithm; S24: The central control system calculates the user activity coefficient based on the number and time of appliance usage collected in S22, the user activity time extracted in S23, and the preset standard activity time.

4. A smart home control method according to claim 3, characterized in that: The S24 specifically includes: S241: The formula for calculating the user activity coefficient according to the number and time of appliance use, user activity time, and preset standard activity data is: Among them, A c is the user activity coefficient, N is the number of times the appliance is used, T e T is the total time the appliance is used, in min. sta is the user activity time, in units of min, T s is the standard activity time, its unit is min, α and β are weight coefficients.

5. The smart home control method according to claim 1, characterized in that: The S3 specifically includes: S31: pre-processing the environmental parameter data and user activity parameter data collected in S1 and S2, including data cleaning, denoising and standardization; S32: extracting feature parameters from the preprocessed data using a feature extraction algorithm, including ambient temperature, humidity, light intensity, user activity time, and the number and time of appliance usage; S33: Based on the extracted feature parameters, an adaptive control model is trained using historical data and a supervised learning algorithm to predict the optimal control strategy based on the input environment and user activity data; S34: Input the environmental parameter data and user activity parameter data collected in real time into the trained adaptive control model. The adaptive control model calculates and generates the optimal joint control strategy based on the real-time input data. The joint control strategy includes the operating status and adjustment parameters of each smart device.

6. A smart home control method according to claim 5, characterized in that: The S32 specifically includes: S321: Temperature feature extraction: For the preprocessed temperature data, a weighted moving average algorithm is used to calculate the average temperature value over a period of time. The formula is: Among them, T i is the temperature data at the i-th moment, in degrees Celsius, W i represents the weight of the i-th time point, n represents the size of the sliding window, is the weighted moving average temperature of the jth segment, in degrees Celsius; S322: Humidity feature extraction: for the pre-processed humidity data, the exponential moving average algorithm is used to calculate the average humidity value over a period of time. The formula is: in, is the average humidity value at the tth time point, and its unit is %, H t represents the humidity data at the tth time point, its unit is %, and α is the smoothing coefficient; S323: Light feature extraction: for the pre-processed light data, the moving median algorithm is used to calculate the fluctuation of light intensity. The formula is: Among them, L j represents the illumination data at the jth time point, its unit is lux, n is the sliding window size, is the moving median illuminance of the jth segment, and its unit is lux.

7. A smart home control method according to claim 6, characterized in that: The S33 specifically includes: S331: extracting preprocessed feature parameters from the historical data set to form a training data set and a verification data set, the feature parameters including temperature, humidity, illuminance, user activity time, number of times and time of appliance use; S332: Feature vector construction, combining the feature parameter values ​​at each time point into a feature vector: S331: Extracting preprocessed feature parameters from the historical data set to form a training data set and a validation data set, the feature parameters include temperature, humidity, illuminance, user activity time, number of times and time of appliance use; S332: Feature vector construction, combining the feature parameter values ​​at each time point into a feature vector: X i =[T i ,H i ,L i ,A i ,N i ,T ei ]; S333: Define the target variable according to the historical control strategy, which represents the optimal control strategy corresponding to each feature vector: Y i =OptimalControl(X i ); S334: Select the multi-layer perceptron algorithm in the supervised learning algorithm for model training. The model structure of the multi-layer perceptron is: in, represents the control strategy predicted by the model, M represents the multilayer perceptron model, and θ represents the model parameters; S335: Use mean square error as the loss function to evaluate the error between the control strategy predicted by the model and the target variable. The formula is: in, represents the control strategy predicted by the model, Y i represents the optimal control strategy corresponding to each eigenvector, and n represents the number of samples; S336: Use the back propagation algorithm and gradient descent algorithm to minimize the loss function and update the model parameters. The formula is: Where θ is the model parameter, η is the learning rate, is the gradient of the loss function with respect to the model parameters θ; S337: Use the validation data set to evaluate the performance of the trained model and optimize the model by adjusting the hyperparameters until the validation error of the model is minimized. The formula is: Among them, m is the number of samples in the validation set, is the predicted value of the validation set, Y j val is the true value of the validation set.

8. The smart home control method according to claim 1, characterized in that: The S4 specifically includes: S41: Analyze the optimal control strategy generated by S3, extract the specific control parameters and status information of each smart device, and form a control vector: C i =[C1,C2,…,C m ]; Among them C i represents the control parameters of the i-th device; S42: Calculate the matching degree between the user activity coefficient generated by S2 and the optimal control strategy. The formula is: Where S is the matching degree between the user activity coefficient and the optimal control strategy, C i represents the control parameters of the ith device, A c is the user activity coefficient, W i is the weight coefficient of the i-th device; W j is the weight coefficient of the i-1th device; Y i is the optimal control strategy corresponding to each eigenvector; S43: Select the optimal control strategy with the highest matching degree, control and adjust the device according to the Internet of Things, and select the strategy method of the scene mode with the highest matching degree as follows: S=[S1,S2,…,S n ]; Where S is the matching degree between the user activity coefficient and the optimal control strategy, n is the number of candidate control strategies; Y i * It is the optimal control strategy with the highest matching degree between the user activity coefficient and the optimal control strategy.

9. The smart home control method according to claim 1, characterized in that: The S5 specifically includes: S51: configuring mobile terminal communication, collecting user feedback information; and classifying the user feedback information into positive feedback and negative feedback; S52: Optimize the device control parameters according to the user feedback type. The formula is: in is the optimized control parameter of the ith device, C i The current control parameters of the i-th device; is the adjustment factor for positive feedback, is the adjustment coefficient of negative feedback, and k is the feedback weight coefficient.

10. A smart home control system, used to implement the smart home control method according to any one of claims 1 to 9, characterized in that: include: The environmental monitoring module is used to collect and initially process environmental information in real time, including temperature, humidity and light intensity, through sensor devices installed in the home environment; User data collection and analysis module, used to collect user behavior information in real time and upload it to the server; and used to analyze user activities based on user behavior data; A data processing module is used to generate an optimal device control strategy using a predefined adaptive algorithm after receiving temperature data, humidity data, illumination data and user behavior data; The device control module is used to select the optimal device control strategy based on user activities; and is used to send the control plan to each home device through the IoT gateway to achieve intelligent adjustment of home devices; The feedback optimization module is used to collect user feedback and optimize device control parameters based on the feedback data.

Images