Intelligent household terminal equipment linkage control method based on Internet of Things technology

Abstract

The invention discloses a smart home terminal equipment linkage control method based on the Internet of Things technology, and relates to the technical field of smart home, and the method comprises the steps: constructing a smart home control architecture; collecting home environment information in real time through a sensor arranged on the intelligent home terminal equipment; data transmission of the home environment information is carried out through the Internet of Things communication technology; the control terminal performs data preprocessing on the historical home environment information; processing the to-be-learned data by using a random forest classification algorithm; constructing a long-short-term memory neural network model, and predicting family environment information at a future moment; performing adaptive control on the smart home terminal equipment by using a fuzzy logic control method; and based on the real-time home environment information, the user side performs linkage control on the intelligent home terminal equipment. The control terminal and the user terminal carry out linkage control on all the intelligent home terminal devices, and the stability of the system and the user experience are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to a linkage control method for smart home terminal equipment based on the Internet of Things technology. Background Art

[0002] With the rapid development of economy and the continuous improvement of people's living standards, people have higher and higher requirements for the quality of home life. Traditional home life can no longer meet people's growing needs, and the emergence of smart home just solves this problem. The rapid development of Internet of Things technology provides strong technical support for the realization of smart home. Internet of Things technology realizes the connection of things through wireless communication technology, sensor technology, data processing and analysis technology, etc., and lays the foundation for the linkage control between smart home terminal devices.

[0003] The problems currently faced by the linkage control method of smart home terminal devices based on the Internet of Things technology involve multiple aspects such as device compatibility, system stability, and user experience. In order to promote the development and popularization of the smart home industry, it is necessary to continuously strengthen technological innovation and standardization work to improve the security, stability and user experience of the system. At the same time, it is also necessary to strengthen industry supervision and self-discipline to ensure the healthy and orderly development of the smart home market. Summary of the invention

[0004] In order to solve the above technical problems, a linkage control method for smart home terminal devices based on Internet of Things technology is provided. This technical solution solves the problems raised in the above background technology.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] The method for linkage control of smart home terminal devices based on Internet of Things technology includes:

[0007] Build smart home control architecture;

[0008] Collect home environment information in real time through sensors equipped on smart home terminal devices;

[0009] The home environment information is transmitted to the home gateway through the Internet of Things communication technology, and the home gateway summarizes the home environment information and transmits it to the control terminal;

[0010] The control terminal performs data preprocessing on the historical home environment information to obtain the data to be learned;

[0011] The random forest classification algorithm is used to process the learning data, and the random forest classification model is obtained to recognize different states of the home environment and different behavior patterns of users in real time;

[0012] Construct a long short-term memory neural network model based on gated recurrent units to predict home environment information at future moments;

[0013] Based on the current home environment status, user behavior patterns and future home environment information, the fuzzy logic control method is used to adaptively control the smart home terminal devices;

[0014] Based on real-time home environment information, the user side performs linkage control on smart home terminal devices.

[0015] Preferably, the construction of the smart home control architecture specifically includes:

[0016] Set up device nodes on each smart home terminal device;

[0017] All device nodes are combined to form a terminal node;

[0018] Set the data transmission center as the home gateway;

[0019] Combining home gateways with terminal nodes to form the underlying structure of the smart home control architecture;

[0020] Set up a control terminal to process data transmitted by the home gateway and centrally control the terminal nodes;

[0021] Set up the user end to manually control the smart home terminal devices;

[0022] The control terminal and the user end are combined to form the upper structure of the smart home control architecture.

[0023] Preferably, the real-time collection of home environment information by sensors equipped by smart home terminal devices specifically includes:

[0024] Collect ambient temperature change data through temperature sensors;

[0025] Collect environmental humidity change data through humidity sensors;

[0026] Collect environmental volume change data through volume sensors;

[0027] Collect light intensity change data through light sensors;

[0028] Collect user behavior trajectory data through human body sensors.

[0029] Preferably, the transmitting of the home environment information to the home gateway through the Internet of Things communication technology, and the home gateway summarizing and transmitting the home environment information to the control terminal specifically includes:

[0030] Determine whether the smart home terminal device node is less than the node preset threshold. If so, no output is made. If not, the home environment information is transmitted to the home gateway using ZigBee wireless communication technology.

[0031] Determine whether the distance between the smart home terminal device and the home gateway is less than a preset distance threshold. If so, use Bluetooth wireless communication technology to transmit the home environment information to the home gateway. If not, use WiFi wireless communication technology to transmit the home environment information to the home gateway.

[0032] Use WiFi wireless communication technology to aggregate and transmit home environment information to the control terminal.

[0033] Preferably, the control terminal performs data preprocessing on the home environment information to obtain the data to be learned, specifically including:

[0034] Remove duplicate data and extreme values ​​caused by failures;

[0035] The family environment information is normalized using a normalization formula to obtain normalized data;

[0036] Obtain the scaling function, wavelet mother function and adjustment factor;

[0037] Substitute the normalized data into the scaling function;

[0038] Substitute the normalized data into the wavelet mother function;

[0039] Based on the scale function, the wavelet mother function and the adjustment factor, a wavelet transform coefficient formula is constructed;

[0040] Determine whether the wavelet transform coefficient of the normalized data is higher than the coefficient setting threshold. If so, output the normalized data corresponding to the home environment information as the data to be learned; if not, output it as noise data;

[0041] The normalization formula is:

[0042]

[0043] In the formula, x i，j ' is the jth normalized data of the i-th group, x i，j is the jth family environment information of the i-th group, x i，min , x i，min are the maximum and minimum values ​​of the family environment information of group i, respectively;

[0044] The scaling function is:

[0045]

[0046] Where ψ(x') is the scaling function, j is the imaginary unit, and x' is the normalized data;

[0047] The wavelet mother function is:

[0048]

[0049] Where φ(x') is the wavelet mother function;

[0050] The wavelet transform coefficient formula is:

[0051]

[0052] Where WT(x') is the wavelet transform coefficient and τ is the adjustment factor.

[0053] Preferably, the process of processing the learning data using the random forest classification algorithm to identify different states of the home environment and different behavior patterns of the user specifically includes:

[0054] S101: defining different states of a home environment and different behavior patterns of a user, wherein the different states of the home environment include weekdays, holidays, daytime, nighttime, sunny days, cloudy days, rainy days, summer, winter, and spring and autumn, and the different behavior patterns of a user include sleeping, exercising, cooking, leisure, and working;

[0055] S102: setting the to-be-learned data under different states of the home environment and different behavior modes of the user as a model training data set;

[0056] S103: Randomly select a subset of data to be learned that has not been selected from the model training data set as the root node of the decision tree;

[0057] S104: establishing a first leaf node at the root node according to the change in ambient temperature as a splitting criterion;

[0058] S105: establishing a second leaf node at the first leaf node according to the change of environmental humidity as a splitting criterion;

[0059] S106: establishing a third leaf node at the second leaf node according to the change in the ambient volume as a splitting criterion;

[0060] S107: establishing a fourth leaf node at the third leaf node according to the change in light intensity as a splitting criterion;

[0061] S108: establishing a fifth leaf node at the fourth leaf node according to the user behavior trajectory as a splitting criterion;

[0062] S109: Summarize the root node and all its leaf nodes and output them as a decision tree;

[0063] S110: Determine whether corresponding decision trees are established for all the data subsets to be learned in the model training data set. If so, summarize and output all decision trees as a random forest classification model. If not, return to step S103;

[0064] S111: Input the real-time home environment information into the random forest classification model to obtain the current home environment status and user behavior pattern.

[0065] Preferably, the construction of a long short-term memory neural network model based on a gated recurrent unit to predict home environment information at a future moment specifically includes:

[0066] According to the gated recurrent unit formula, the input gate, forget gate and output gate functions are calculated respectively;

[0067] Construct a long short-term memory neural network model based on input gate, forget gate and output gate functions;

[0068] Input the real-time home environment information into the long short-term memory neural network model to obtain the home environment information at the future moment;

[0069] The gated recurrent unit formula is:

[0070]

[0071] In the formula, α t , β t , γ t are the input gate, forget gate and output gate functions respectively, W α , W β , W γ are the parameter matrices of the input gate, forget gate, and output gate from time t-1 to time t, U α , U β , U γ are the parameter matrices from the input layer to the current layer of the input gate, forget gate, and output gate, respectively, x' t-1 , x' t are the data to be learned at time t-1 and time t respectively, and tanh(y) is the hyperbolic tangent function;

[0072] The hyperbolic tangent function is:

[0073]

[0074] Where y is the input data of the hyperbolic tangent function, and tanh(y) is the output data of the hyperbolic tangent function.

[0075] Preferably, the adaptive control of the smart home terminal device using the fuzzy logic control method based on the current home environment state, user behavior pattern and future home environment information specifically includes:

[0076] Determine that the input variables of the fuzzy logic control system are ambient temperature, ambient humidity, ambient volume, light intensity and user behavior trajectory, and the output variables are the operating parameters of each smart home terminal device;

[0077] The input variables are converted into fuzzy quantities using normal membership functions;

[0078] Obtain at least one fuzzy rule, determine the relationship between the fuzzy quantity of the input variable and the fuzzy quantity of the output variable, and obtain the fuzzy quantity of the output variable;

[0079] Based on the centroid method formula, the fuzzy quantity of the output variable is defuzzified using the centroid method to obtain the exact value of the output variable;

[0080] Output the precise value of the output variable as a control instruction to each smart home terminal device;

[0081] The normal type membership function is:

[0082]

[0083] In the formula, μ is the fuzzy value of the input variable, z is the input variable, a is the preset fuzzy threshold, and σ is an adjustable parameter;

[0084] The centroid method formula is:

[0085]

[0086] Where v' is the exact value of the output variable, v is the fuzzy value of the output variable, and μ(v) is the fuzzy value of the corresponding input variable.

[0087] Preferably, the user end performs linkage control on the smart home terminal device based on the real-time home environment information specifically includes:

[0088] The user terminal presets at least one scene mode and sets the operating parameters of each device in each scene mode, wherein the scene modes include leaving home mode, returning home mode, sleeping mode and dining mode;

[0089] The user presets the automatic control conditions of the equipment, and the system automatically triggers the linkage control of the equipment according to the preset conditions;

[0090] Determine whether the ambient temperature is higher than the preset temperature and the human body sensor detects that the user is in the room. If so, the air conditioner is automatically turned on. If not, no output is made.

[0091] Determine whether the ambient humidity is higher than the preset humidity. If so, put away the clothes drying rack, close the window and turn on the dryer. If not, do not output anything.

[0092] The user end controls the smart furniture terminal device through preset voice commands;

[0093] The user can remotely view the status of home appliances in real time and perform linkage control via smartphones and tablets.

[0094] Compared with the prior art, the present invention has the following beneficial effects:

[0095] The Internet of Things technology enables smart home devices to be intelligently adjusted according to environmental changes and user needs. The smart home system can also provide personalized health management, entertainment, and family atmosphere adjustment services based on the health status, preferences, and living habits of family members, significantly improving the convenience, comfort, safety, energy conservation, and environmental protection of home life. Through the control terminal and the user end, each smart home terminal device can be linked and controlled to improve the system stability and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 This is a flow chart of the linkage control method of smart home terminal devices based on Internet of Things technology of the present invention;

[0097] Figure 2 A flow chart of the method for constructing a smart home control architecture of the present invention;

[0098] Figure 3 This is a flow chart of the method for collecting home environment information in real time according to the present invention;

[0099] Figure 4 This is a flow chart of the information transmission method using the Internet of Things communication technology of the present invention;

[0100] Figure 5 A flow chart of a method for preprocessing data of historical home environment information by a control terminal of the present invention;

[0101] Figure 6 This is a flow chart of a method for processing learning data using a random forest classification algorithm according to the present invention;

[0102] Figure 7 A flow chart of a method for predicting home environment information at a future moment of the present invention;

[0103] Figure 8 This is a flow chart of a method for adaptively controlling a smart home terminal device using a fuzzy logic control method according to the present invention;

[0104] Fig. 9 This is a flow chart of the method for the user end of the present invention to perform linkage control on smart home terminal devices. DETAILED DESCRIPTION

[0105] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0106] Reference Figure 1 As shown, the linkage control method of smart home terminal devices based on the Internet of Things technology includes:

[0107] Build smart home control architecture;

[0108] Collect home environment information in real time through sensors equipped on smart home terminal devices;

[0109] The home environment information is transmitted to the home gateway through the Internet of Things communication technology, and the home gateway summarizes the home environment information and transmits it to the control terminal;

[0110] The control terminal performs data preprocessing on the historical home environment information to obtain the data to be learned;

[0111] The random forest classification algorithm is used to process the learning data, and the random forest classification model is obtained to recognize different states of the home environment and different behavior patterns of users in real time;

[0112] Construct a long short-term memory neural network model based on gated recurrent units to predict home environment information at future moments;

[0113] Based on the current home environment status, user behavior patterns and future home environment information, the fuzzy logic control method is used to adaptively control the smart home terminal devices;

[0114] Based on real-time home environment information, the user side performs linkage control on smart home terminal devices.

[0115] Reference Figure 2 As shown in the figure, building a smart home control architecture specifically includes:

[0116] Set up device nodes on each smart home terminal device;

[0117] All device nodes are combined to form a terminal node;

[0118] Set the data transmission center as the home gateway;

[0119] Combining home gateways with terminal nodes to form the underlying structure of the smart home control architecture;

[0120] Set up a control terminal to process data transmitted by the home gateway and centrally control the terminal nodes;

[0121] Set up the user end to manually control the smart home terminal devices;

[0122] The control terminal and the user end are combined to form the upper structure of the smart home control architecture.

[0123] The architecture of a smart home control system usually includes an underlying structure and an upper structure. The underlying structure is composed of terminal nodes and home gateways. The home gateway serves as the communication hub of the equipment to realize external network communication. The upper structure mainly includes clients and control terminals. The client can view and obtain the system operation status in real time, and the control terminal can realize community energy consumption management and centralized control.

[0124] Reference Figure 3 As shown in the figure, the real-time collection of home environment information by sensors equipped with smart home terminal devices specifically includes:

[0125] Collect ambient temperature change data through temperature sensors;

[0126] Collect environmental humidity change data through humidity sensors;

[0127] Collect environmental volume change data through volume sensors;

[0128] Collect light intensity change data through light sensors;

[0129] Collect user behavior trajectory data through human body sensors.

[0130] Smart home terminal devices are usually equipped with various sensors, such as temperature sensors, humidity sensors, light sensors, human body sensors, etc., which are used to perceive home environment information in real time. These sensors transmit the collected data to the home gateway or cloud platform through the Internet of Things technology to provide data support for intelligent control.

[0131] Reference Figure 4 As shown, the home environment information is transmitted to the home gateway through the Internet of Things communication technology, and the home gateway aggregates and transmits the home environment information to the control terminal, specifically including:

[0132] Determine whether the smart home terminal device node is less than the node preset threshold. If so, no output is made. If not, the home environment information is transmitted to the home gateway using ZigBee wireless communication technology.

[0133] Determine whether the distance between the smart home terminal device and the home gateway is less than a preset distance threshold. If so, use Bluetooth wireless communication technology to transmit the home environment information to the home gateway. If not, use WiFi wireless communication technology to transmit the home environment information to the home gateway.

[0134] Use WiFi wireless communication technology to aggregate and transmit home environment information to the control terminal.

[0135] ZigBee is a low-power, low-rate, low-cost IoT communication protocol suitable for low-power, low-cost device connection and data transmission in smart home environments. Wi-Fi has high-speed, long-distance wireless communication capabilities and is suitable for devices in smart homes that require high-speed data transmission. Bluetooth is a low-power, short-distance, high-reliability, low-cost wireless communication technology suitable for short-distance communication and data transmission between smart home devices.

[0136] Reference Figure 5 As shown, the control terminal performs data preprocessing on the home environment information, and the data to be learned specifically includes:

[0137] Remove duplicate data and extreme values ​​caused by failures;

[0138] The family environment information is normalized using a normalization formula to obtain normalized data;

[0139] Obtain the scaling function, wavelet mother function and adjustment factor;

[0140] Substitute the normalized data into the scaling function;

[0141] Substitute the normalized data into the wavelet mother function;

[0142] Based on the scale function, the wavelet mother function and the adjustment factor, a wavelet transform coefficient formula is constructed;

[0143] Determine whether the wavelet transform coefficient of the normalized data is higher than the coefficient setting threshold. If so, output the normalized data corresponding to the home environment information as the data to be learned; if not, output it as noise data;

[0144] The normalization formula is:

[0145]

[0146] In the formula, x i，j ' is the jth normalized data of the i-th group, x i，j is the jth family environment information of the i-th group, x i，min , x i，min are the maximum and minimum values ​​of the family environment information of group i, respectively;

[0147] The scaling function is:

[0148]

[0149] Where ψ(x') is the scaling function, j is the imaginary unit, and x' is the normalized data;

[0150] The wavelet mother function is:

[0151]

[0152] Where φ(x') is the wavelet mother function;

[0153] The wavelet transform coefficient formula is:

[0154]

[0155] Where WT(x') is the wavelet transform coefficient and τ is the adjustment factor.

[0156] Normalization is the process of changing a column of data to a fixed interval. Usually, this interval is a decimal between [0, 1] or [-1, 1]. It is mainly for the convenience of data processing. Mapping the data to the range of 0 to 1 for processing is more convenient and faster.

[0157] Reference Figure 6 As shown, the random forest classification algorithm is used to process the learning data to identify different states of the home environment and different behavior patterns of users, including:

[0158] S101: defining different states of a home environment and different behavior patterns of a user, wherein the different states of the home environment include weekdays, holidays, daytime, nighttime, sunny days, cloudy days, rainy days, summer, winter, and spring and autumn, and the different behavior patterns of a user include sleeping, exercising, cooking, leisure, and working;

[0159] S102: setting the to-be-learned data under different states of the home environment and different behavior modes of the user as a model training data set;

[0160] S103: Randomly select a subset of data to be learned that has not been selected from the model training data set as the root node of the decision tree;

[0161] S104: establishing a first leaf node at the root node according to the change in ambient temperature as a splitting criterion;

[0162] S105: establishing a second leaf node at the first leaf node according to the change of environmental humidity as a splitting criterion;

[0163] S106: establishing a third leaf node at the second leaf node according to the change in the ambient volume as a splitting criterion;

[0164] S107: establishing a fourth leaf node at the third leaf node according to the change in light intensity as a splitting criterion;

[0165] S108: establishing a fifth leaf node at the fourth leaf node according to the user behavior trajectory as a splitting criterion;

[0166] S109: Summarize the root node and all its leaf nodes and output them as a decision tree;

[0167] S110: Determine whether corresponding decision trees are established for all the data subsets to be learned in the model training data set. If so, summarize and output all decision trees as a random forest classification model. If not, return to step S103;

[0168] S111: Input the real-time home environment information into the random forest classification model to obtain the current home environment status and user behavior pattern.

[0169] Random forest is an algorithm that integrates multiple trees through the idea of ​​ensemble learning. Its basic unit is the decision tree. From an intuitive point of view, each decision tree is a classifier. Then for an input sample, N trees will have N classification results, and random forest integrates all classification voting results and designates the category with the most votes as the final output.

[0170] Reference Figure 7 As shown, a long short-term memory neural network model is constructed. Based on the gated recurrent unit, the family environment information at the future moment is predicted, including:

[0171] According to the gated recurrent unit formula, the input gate, forget gate and output gate functions are calculated respectively;

[0172] Construct a long short-term memory neural network model based on input gate, forget gate and output gate functions;

[0173] Input the real-time home environment information into the long short-term memory neural network model to obtain the home environment information at the future moment;

[0174] The gated recurrent unit formula is:

[0175]

[0176] In the formula, α t , β t , γ t are the input gate, forget gate and output gate functions respectively, W α , W β , W γ are the parameter matrices of the input gate, forget gate, and output gate from time t-1 to time t, U α , U β , U γ are the parameter matrices from the input layer to the current layer of the input gate, forget gate, and output gate, respectively, x' t-1 , x' t are the data to be learned at time t-1 and time t respectively, and tanh(y) is the hyperbolic tangent function;

[0177] The hyperbolic tangent function is:

[0178]

[0179] Where y is the input data of the hyperbolic tangent function, and tanh(y) is the output data of the hyperbolic tangent function.

[0180] Recurrent neural networks are often used in time series related tasks. Common variants include long short-term memory networks and gated recurrent units. The traditional neural network structure is feedforward, and there is no connection between the nodes of each layer. The parameters of each layer of the recurrent neural network are shared. The output of the current layer not only considers the output of the previous layer, but also the output of the hidden layer at the previous moment.

[0181] Reference Figure 8 As shown in the figure, based on the current home environment status, user behavior patterns and future home environment information, the fuzzy logic control method is used to adaptively control the smart home terminal devices, including:

[0182] Determine that the input variables of the fuzzy logic control system are ambient temperature, ambient humidity, ambient volume, light intensity and user behavior trajectory, and the output variables are the operating parameters of each smart home terminal device;

[0183] The input variables are converted into fuzzy quantities using normal membership functions;

[0184] Obtain at least one fuzzy rule, determine the relationship between the fuzzy quantity of the input variable and the fuzzy quantity of the output variable, and obtain the fuzzy quantity of the output variable;

[0185] Based on the centroid method formula, the fuzzy quantity of the output variable is defuzzified using the centroid method to obtain the exact value of the output variable;

[0186] Output the precise value of the output variable as a control instruction to each smart home terminal device;

[0187] The normal type membership function is:

[0188]

[0189] In the formula, μ is the fuzzy value of the input variable, z is the input variable, a is the preset fuzzy threshold, and σ is an adjustable parameter;

[0190] The centroid method formula is:

[0191]

[0192] Where v' is the exact value of the output variable, v is the fuzzy value of the output variable, and μ(v) is the fuzzy value of the corresponding input variable.

[0193] Fuzzy logic control is a kind of control that uses language to summarize the operator's control strategy and uses language variables and fuzzy set theory to form a control algorithm. The most important feature of fuzzy logic control is that it does not require the establishment of an accurate mathematical model of the controlled object. It only requires the operator's experience and data to be summarized into a more complete language control rule, so that the control object with characteristics such as uncertainty, imprecision, noise, nonlinearity, time-varying, and time-delay can be controlled. The fuzzy logic control system is highly robust and is particularly suitable for the control of nonlinear, time-varying, and hysteresis systems.

[0194] Reference Fig. 9 As shown, based on the real-time home environment information, the user end performs linkage control on the smart home terminal devices, including:

[0195] The user terminal presets at least one scene mode and sets the operating parameters of each device in each scene mode, wherein the scene modes include leaving home mode, returning home mode, sleeping mode and dining mode;

[0196] The user presets the automatic control conditions of the equipment, and the system automatically triggers the linkage control of the equipment according to the preset conditions;

[0197] Determine whether the ambient temperature is higher than the preset temperature and the human body sensor detects that the user is in the room. If so, the air conditioner is automatically turned on. If not, no output is made.

[0198] Determine whether the ambient humidity is higher than the preset humidity. If so, put away the clothes drying rack, close the window and turn on the dryer. If not, do not output anything.

[0199] The user end controls the smart furniture terminal device through preset voice commands;

[0200] The user can remotely view the status of home appliances in real time and perform linkage control via smartphones and tablets.

[0201] The above control method realizes the intelligent interconnection and linkage control of home appliances. This control method not only improves the comfort and convenience of users' home life, but also promotes the efficient use of household energy and environmental protection and energy saving.

[0202] Furthermore, the present solution also proposes a computer-readable storage medium on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned smart home terminal device linkage control method based on the Internet of Things technology is executed.

[0203] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0204] To sum up, the advantages of the present invention are: the Internet of Things technology enables smart home devices to be intelligently adjusted according to environmental changes and user needs. The smart home system can also provide personalized health management, entertainment and family atmosphere adjustment services according to the health conditions, preferences and living habits of family members, which significantly improves the convenience, comfort, safety and energy saving and environmental protection of home life. By controlling the control terminal and the user end to control each smart home terminal device in a linked manner, the stability of the system and the user experience are improved.

[0205] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A linkage control method for smart home terminal devices based on Internet of Things technology, characterized in that: include: Build smart home control architecture; Collect home environment information in real time through sensors equipped on smart home terminal devices; The home environment information is transmitted to the home gateway through the Internet of Things communication technology, and the home gateway summarizes the home environment information and transmits it to the control terminal; The control terminal performs data preprocessing on the historical home environment information to obtain the data to be learned; The random forest classification algorithm is used to process the learning data, and the random forest classification model is obtained to recognize different states of the home environment and different behavior patterns of users in real time; Construct a long short-term memory neural network model based on gated recurrent units to predict home environment information at future moments; Based on the current home environment status, user behavior patterns and future home environment information, the fuzzy logic control method is used to adaptively control the smart home terminal devices; Based on real-time home environment information, the user side performs linkage control on smart home terminal devices.

2. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 1 is characterized in that: The construction of the smart home control architecture specifically includes: Set up device nodes on each smart home terminal device; All device nodes are combined to form a terminal node; Set the data transmission center as the home gateway; Combining home gateways with terminal nodes to form the underlying structure of the smart home control architecture; Set up a control terminal to process data transmitted by the home gateway and centrally control the terminal nodes; Set up the user end to manually control the smart home terminal devices; The control terminal and the user end are combined to form the upper structure of the smart home control architecture.

3. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 2 is characterized in that: The real-time collection of home environment information by sensors equipped by smart home terminal devices specifically includes: Collect ambient temperature change data through temperature sensors; Collect environmental humidity change data through humidity sensors; Collect environmental volume change data through volume sensors; Collect light intensity change data through light sensors; Collect user behavior trajectory data through human body sensors.

4. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 3 is characterized in that: The transmitting of the home environment information to the home gateway through the Internet of Things communication technology, and the home gateway summarizing and transmitting the home environment information to the control terminal specifically includes: Determine whether the smart home terminal device node is less than the node preset threshold. If so, no output is made. If not, the home environment information is transmitted to the home gateway using ZigBee wireless communication technology. Determine whether the distance between the smart home terminal device and the home gateway is less than a preset distance threshold. If so, use Bluetooth wireless communication technology to transmit the home environment information to the home gateway. If not, use WiFi wireless communication technology to transmit the home environment information to the home gateway. Use WiFi wireless communication technology to aggregate and transmit home environment information to the control terminal.

5. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 4 is characterized in that: The control terminal performs data preprocessing on the home environment information to obtain the data to be learned, specifically including: Remove duplicate data and extreme values ​​caused by failures; The family environment information is normalized using a normalization formula to obtain normalized data; Obtain the scaling function, wavelet mother function and adjustment factor; Substitute the normalized data into the scaling function; Substitute the normalized data into the wavelet mother function; Based on the scale function, the wavelet mother function and the adjustment factor, a wavelet transform coefficient formula is constructed; Determine whether the wavelet transform coefficient of the normalized data is higher than the coefficient setting threshold. If so, output the normalized data corresponding to the home environment information as the data to be learned; if not, output it as noise data; The normalization formula is: In the formula, x i，j ' is the jth normalized data of the i-th group, x i，j is the jth family environment information of the i-th group, x i，min , x i，min are the maximum and minimum values ​​of the family environment information of group i, respectively; The scaling function is: Where ψ(x') is the scaling function, j is the imaginary unit, and x' is the normalized data; The wavelet mother function is: Where φ(x') is the wavelet mother function; The wavelet transform coefficient formula is: Where WT(x') is the wavelet transform coefficient and τ is the adjustment factor.

6. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 5 is characterized in that: The use of the random forest classification algorithm to process the learning data and identify different states of the home environment and different behavior patterns of the user specifically includes: S101: defining different states of a home environment and different behavior patterns of a user, wherein the different states of the home environment include weekdays, holidays, daytime, nighttime, sunny days, cloudy days, rainy days, summer, winter, and spring and autumn, and the different behavior patterns of a user include sleeping, exercising, cooking, leisure, and working; S102: setting the to-be-learned data under different states of the home environment and different behavior modes of the user as a model training data set; S103: Randomly select a subset of data to be learned that has not been selected from the model training data set as the root node of the decision tree; S104: establishing a first leaf node at the root node according to the change in ambient temperature as a splitting criterion; S105: establishing a second leaf node at the first leaf node according to the change of environmental humidity as a splitting criterion; S106: establishing a third leaf node at the second leaf node according to the change in the ambient volume as a splitting criterion; S107: establishing a fourth leaf node at the third leaf node according to the change in light intensity as a splitting criterion; S108: establishing a fifth leaf node at the fourth leaf node according to the user behavior trajectory as a splitting criterion; S109: Summarize the root node and all its leaf nodes and output them as a decision tree; S110: Determine whether corresponding decision trees are established for all the data subsets to be learned in the model training data set. If so, summarize and output all decision trees as a random forest classification model. If not, return to step S103; S111: Input the real-time home environment information into the random forest classification model to obtain the current home environment status and user behavior pattern.

7. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 6 is characterized in that: The construction of the long short-term memory neural network model based on the gated recurrent unit to predict the home environment information at the future moment specifically includes: According to the gated recurrent unit formula, the input gate, forget gate and output gate functions are calculated respectively; Construct a long short-term memory neural network model based on input gate, forget gate and output gate functions; Input the real-time home environment information into the long short-term memory neural network model to obtain the home environment information at the future moment; The gated recurrent unit formula is: In the formula, α t , β t , γ t are the input gate, forget gate and output gate functions respectively, W α , W β , W γ are the parameter matrices of the input gate, forget gate, and output gate from time t-1 to time t, U α , U β , U γ are the parameter matrices from the input layer to the current layer of the input gate, forget gate, and output gate, respectively, x' t-1 , x' t are the data to be learned at time t-1 and time t respectively, and tanh(y) is the hyperbolic tangent function; The hyperbolic tangent function is: Where y is the input data of the hyperbolic tangent function, and tanh(y) is the output data of the hyperbolic tangent function.

8. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 7 is characterized in that: The adaptive control of the smart home terminal device using the fuzzy logic control method based on the current home environment state, user behavior pattern and future home environment information specifically includes: Determine that the input variables of the fuzzy logic control system are ambient temperature, ambient humidity, ambient volume, light intensity and user behavior trajectory, and the output variables are the operating parameters of each smart home terminal device; The input variables are converted into fuzzy quantities using normal membership functions; Obtain at least one fuzzy rule, determine the relationship between the fuzzy quantity of the input variable and the fuzzy quantity of the output variable, and obtain the fuzzy quantity of the output variable; Based on the centroid method formula, the fuzzy quantity of the output variable is defuzzified using the centroid method to obtain the exact value of the output variable; Output the precise value of the output variable as a control instruction to each smart home terminal device; The normal type membership function is: In the formula, μ is the fuzzy value of the input variable, z is the input variable, a is the preset fuzzy threshold, and σ is an adjustable parameter; The centroid method formula is: Where v' is the exact value of the output variable, v is the fuzzy value of the output variable, and μ(v) is the fuzzy value of the corresponding input variable.

9. The method for controlling the linkage of smart home terminal devices based on the Internet of Things technology according to claim 8, characterized in that: The user end performs linkage control on the smart home terminal device based on the real-time home environment information specifically includes: The user terminal presets at least one scene mode and sets the operating parameters of each device in each scene mode, wherein the scene modes include leaving home mode, returning home mode, sleeping mode and dining mode; The user presets the automatic control conditions of the equipment, and the system automatically triggers the linkage control of the equipment according to the preset conditions; Determine whether the ambient temperature is higher than the preset temperature and the human body sensor detects that the user is in the room. If so, the air conditioner is automatically turned on. If not, no output is made. Determine whether the ambient humidity is higher than the preset humidity. If so, put away the clothes drying rack, close the window and turn on the dryer. If not, do not output anything. The user end controls the smart furniture terminal device through preset voice commands; The user can remotely view the status of home appliances in real time and perform linkage control via smartphones and tablets.

Images