Self-adaptive temperature control device and method for intelligent seat

Abstract

The invention discloses a self-adaptive temperature control device and method for an intelligent seat. The self-adaptive temperature control device comprises a seat body (1), a pipeline (2), temperature control equipment (3), camera equipment (5) and a controller. The pipeline (2) is mounted on the lower surface of a bottom plate of the seat body (1), and the pipeline (2) is connected with the temperature control equipment (3); and the temperature control equipment (3), the camera equipment (5) and the controller are in wireless / wired connection. The age of the user is accurately identified through the user classification model, and different gears are adaptively adjusted according to the age, so that adaptive cooling control according to different crowd characteristics and environment temperature characteristics is realized, and the temperature experience is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent applications, and particularly relates to an adaptive temperature control device and method for an intelligent seat. Background Art

[0002] In recent years, global warming has intensified, and extreme high-temperature weather has occurred frequently in domestic cities. Coupled with factors such as industrial pollution and automobile exhaust pollution, the urban heat island effect has been significantly enhanced. Accelerating the construction of cool cities has become a key task for major cities. For example, on November 3, 2021, the United Nations Environment Programme released a detailed guide to help cities around the world respond to climate change - "Beating the Heat: A Guidebook on Sustainable Urban Cooling" at the United Nations Climate Change Conference, and for the first time proposed the concept of "sustainable urban cooling". Solving the problem of urban high temperature requires implementing the "dual carbon" strategy and promoting "sustainable urban cooling" through technological innovation and engineering practice. Among them, urban public infrastructure has become a research hotspot, such as urban intelligent cooling furniture.

[0003] Intelligent cooling seats are typical products of urban intelligent cooling furniture. Existing cooling seats mostly adopt passive cooling. For example, the seats are made of stone materials. Although they have a certain cooling effect, they cannot meet the diverse cooling needs of the elderly, children, etc., cannot be adaptively adjusted according to the needs of different people, and the cooling effect is insufficient. During high-temperature periods, they cannot even be used directly. A few cooling seats that identify people through intelligent means have the defects of low recognition accuracy and single function, which greatly hinders the market application of intelligent cooling seats. Based on this, it is very necessary to design an intelligent cooling seat that can improve the accuracy of crowd recognition, thus meeting the cooling needs of different people and providing multi-functional services. Summary of the Invention

[0004] Aiming at the technical problem that the intelligent cooling seat in the prior art cannot accurately control the temperature according to the crowd, the present invention proposes an adaptive temperature control device and method for an intelligent seat.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] An adaptive temperature control device for an intelligent seat, comprising a seat body 1, a pipeline 2, a temperature control device 3, a camera device 5 and a controller;

[0007] The pipeline 2 is installed on the lower surface of the bottom plate of the seat body 1, and the pipeline 2 is connected to the temperature control device 3; the temperature control device 3, the camera device 5 and the controller are wirelessly / wiredly connected.

[0008] Preferably, it further comprises a temperature acquisition device 4 connected to the controller.

[0009] Preferably, the bottom plate includes an upper part and a lower part. The upper part is made of wood, and the lower part is made of stainless steel. A groove for installing the pipeline 2 is provided on the lower surface of the lower part.

[0010] Preferably, the temperature control device 3 includes a housing;

[0011] An air inlet pipeline 3-3 is provided on the left side of the housing, and a control valve 3-1 is provided in the air inlet pipeline 3-3; an air outlet pipeline 3-4 is provided on the right side of the housing; the air inlet pipeline 3-3 is connected to one end of the pipeline 2, and the air outlet pipeline 3-4 is connected to the other end of the pipeline 2;

[0012] A bearing 3-10 is installed in the first cavity of the housing. A first row of fans 3-5, a first rotary machine 3-6, a first metal grille 3-7, a second rotary machine 3-8, a second row of fans 3-12 and a second metal grille 3-13 are sequentially installed on the bearing 3-10 from left to right.

[0013] Preferably, the first rotary machine 3-6 includes a first motor and a plurality of metal fan blades, and circular ventilation openings are provided on the surface of the metal fan blades; a solid desiccant is embedded inside the first metal grille 3-7.

[0014] Preferably, the second rotary machine 3-7 includes a second motor, a water supply pipeline, and plastic fan blades, and cotton cloth pieces are evenly distributed on the surface of the plastic fan blades.

[0015] Preferably, a spacer metal sheet 3-11 is provided between the first row of fans 3-5 and the first rotary machine 3-6.

[0016] Preferably, a return water pipe 3-9 is provided on the lower side of the first cavity inside the housing, and a return air pipeline 3-2 protrudes from the lower side of the housing, that is, a second cavity is formed between the first cavity and the return air pipeline 3-2.

[0017] The present invention also provides an adaptive temperature control method for an intelligent seat, including the following steps:

[0018] S1: Real-time obtain the image information of the camera device;

[0019] S2: Input the image information into the trained user classification model to output the user classification;

[0020] S3: Perform temperature adaptive adjustment according to the user classification and the temperature of the seat.

[0021] Preferably, S2 includes:

[0022] S2-1: First, download 10,000 pictures of people's faces and appearances at different age periods from the network. Divide users into 5 categories according to age, including a1, a2, a3, a4, and a5, and store them as the training dataset U.

[0023] S2-2: Import the training dataset U into the BiGRU model for training to obtain the user classification model.

[0024] S2-2-1: Initialize the parameters of the gray wolf population, calculate the fitness of the gray wolf population for the first time, take the gray wolf with the best fitness value as the α-layer wolf, the gray wolf with the second-best fitness as the β-layer wolf, and the gray wolf with the worst fitness as the δ-layer wolf.

[0025] S2-2-2: Update the positions of the gray wolves:

[0026]

[0027] In formula (1), θ’ α represents the position of the updated α-layer wolf; θ α (t) is the position of the α-layer wolf at the t-th iteration, θ’ β represents the position of the updated β-layer wolf; θ β (t) is the position of the β-layer wolf at the t-th iteration; θ’ δ represents the position of the updated δ-layer wolf; θ δ (t) is the position of the δ-layer wolf at the t-th iteration;

[0028] A 1 、A 2 、A 3 、C 1 、C 2 、C 3 represent the cooperation coefficients of the α-layer, β-layer, and δ-layer wolves respectively; θ w (t) represents the position of the w-layer wolf at time t; θ w (t + 1) represents the position of the w-layer wolf at time t + 1;

[0029] S2-2-3: Update λ:

[0030]

[0031] In formula (2), λ represents the convergence factor; t represents the number of iterations; T max represents the maximum number of iterations. S2-2-4: Update A and C according to λ:

[0032] A = 2λr 1 - λ, C = 2r 2 (3)

[0033] In formula (3), both A and C represent vector coefficients; r 1 , r 2 represent random numbers between [0, 1];

[0034] S2-2-5: Calculate the offspring gray wolves:

[0035]

[0036] In formula (4), is the d-th dimension offspring after the i-th parent, is the d-th dimension offspring after the j-th parent, r 3 , r 4 are random numbers between [0, 1], C 4 , C 5 are random numbers between [-1, 1]; θ i,d represents the d-th dimension position data of the i-th parent in the gray wolf population; θ j,d represents the d-th dimension position data of the j-th parent in the gray wolf population; φ CO represents the set of offspring gray wolves; represents the n M th offspring gray wolf;

[0037] S2-2-6: Calculate the fitness values of each gray wolf for the second time. When the fitness value calculated for the gray wolf θ k (k ∈ {1, 2,..., n M}) is lower than that of the gray wolf θ CO in the φ k CO population, update the gray wolf θ k to θ k CO , otherwise do not adjust. Keep performing repeated operations. When the maximum number of iterations T max is reached, stop the operation to obtain the BiGRU model with the optimal hyperparameter values, that is, the user classification model;

[0038] S2-3: Input the real-time acquired image information into the trained user classification model and output the user classification.

[0039] In summary, due to the adoption of the above technical solutions, compared with the prior art, the present invention has at least the following beneficial effects:

[0040] The present invention accurately identifies the user's age through the constructed user classification model, and adaptively adjusts different gears and background colors according to the age, thus realizing the adaptive cooling control according to different population characteristics and environmental temperature characteristics, and greatly improving the temperature experience. Description of the Drawings

[0041] Figure 1 Schematic diagram of an adaptive temperature control device for a smart seat according to an exemplary embodiment of the present invention.

[0042] Figure 2 Schematic diagram of a temperature control device according to an exemplary embodiment of the present invention.

[0043] Figure 3 Schematic diagram of an adaptive temperature control method for a smart seat according to an exemplary embodiment of the present invention. Detailed implementation manners

[0044] The present invention will be further described in detail below in conjunction with embodiments and specific implementation manners. However, this should not be construed as limiting the scope of the above-mentioned subject matter of the present invention to the following embodiments. Any technology implemented based on the content of the present invention belongs to the scope of the present invention.

[0045] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings. These are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.

[0046] As Figure 1 shown, the present invention provides an adaptive temperature control device for a smart seat, including a seat body 1, a pipeline 2, a temperature control device 3, a camera device 5 and a controller.

[0047] In this embodiment, the seat body 1 includes a bracket, armrests, a bottom plate and a backrest; among them, the bracket can be made of metal material, and the armrests, the bottom plate and the backrest can be made of wood material.

[0048] In this embodiment, the armrests, the bottom plate and the backrest are all arranged above the two brackets; among them, a storage opening is also arranged on the armrests.

[0049] In this embodiment, the bottom plate includes two upper and lower parts. The upper part is made of wood, and the lower part is made of stainless steel. A groove for installing the pipeline 2 is arranged in the lower part.

[0050] In this embodiment, the pipeline 2 can be made of copper pipelines.

[0051] As Figure 2As shown, the temperature control device 3 (which can be installed on both sides of the seat) includes a housing, a control valve 3-1, a return air pipeline 3-2, an air inlet pipeline 3-3, an air outlet pipeline 3-4, a first row of fans 3-5, a first rotary machine 3-6, a first metal grille 3-7, a second rotary machine 3-8, a return water pipe 3-9, a mounting bearing 3-10, a spacer metal sheet 3-11, a second row of fans 3-12, and a second metal grille 3-13.

[0052] The air inlet pipeline 3-3 is located on the left side of the housing, and the control valve 3-1 is provided in the air inlet pipeline 3-3; the air outlet pipeline 3-4 is displaced to the right side of the housing; the mounting bearing 3-10 is fixed in the first cavity inside the housing and is located in the center. The first row of fans 3-5, the first rotary machine 3-6, the first metal grille 3-7, the second rotary machine 3-8, the second row of fans 3-12, and the second metal grille 3-13 are successively installed on the mounting bearing 3-10 from left to right.

[0053] The air inlet pipeline 3-3 is connected to one end of the pipeline 2, and the air outlet pipeline 3-4 is connected to the other end of the pipeline 2, so that the air circulates in the pipeline 2; at the same time, there is an external air inlet here in the air inlet pipeline 3-3, which can be opened when the device pressure is insufficient to improve the air circulation efficiency.

[0054] Among them, the control valve 3-1 includes a first gear, a second gear, and a third gear. Among them, the opening degree of the control valve in the third gear is the smallest, and the opening degree of the control valve in the first gear is the largest, and the ventilation volume is the largest at this time.

[0055] The first rotary machine 3-6 is composed of a first motor and a plurality of metal fan blades. There are circular ventilation openings on the surface of the metal fan blades; a solid desiccant is inlaid inside the metal grille 3-7. When the air passes through the first row of fans 3-5, the first rotary machine 3-6, and the first metal grille 3-7, it can play a role in dehumidification.

[0056] The second rotary machine 3-8 includes a second motor, a water supply pipeline, and plastic fan blades. Uniformly distributed cotton cloth pieces are installed on the surface of the plastic fan blades, which are mainly used to adsorb water, so that the air evaporates and cools down when passing through the wetted cotton cloth. The cotton cloth pieces are glued to the plastic fan blades by strong glue. The water supply pipeline (connected to an external water storage device) is installed on the top of the first cavity, such as a micro drip nozzle, which can be opened or closed.

[0057] In this embodiment, when the water supply pipeline is opened, a small amount of water is sprayed into the first cavity, and the second rotary machine 3-8 is used to rotate and accelerate evaporation and cooling; then the second row of fans 3-12 extracts the cooled air, and after drying through the second metal grille 3-13 (equipped with desiccant), it enters the pipeline 2 from the air outlet pipeline 3-4.

[0058] There is a spaced metal sheet 3-11 (containing desiccant for drying) provided between the first row of fans 3-5 and the first rotary machine 3-6; there is also a spaced metal sheet 3-11 provided between the metal grille 3-7 and the second rotary machine 3-8; there is also a spaced metal sheet 3-11 provided between the second rotary machine 3-8 and the second row of fans 3-12.

[0059] The spaced metal sheet 3-11 includes a first sheet and a second sheet. The first sheet is fixed to the upper side of the first cavity of the housing, and the second sheet is fixed to the lower side of the first cavity of the housing. There is a gap between the first sheet and the second sheet to ensure that air passes through.

[0060] A water return pipe 3-9 is provided on the lower side of the first cavity inside the housing, and a return air pipeline 3-2 protrudes from the lower side of the housing, that is, there is a second cavity between the first cavity and the return air pipeline 3-2.

[0061] The function of the water return pipe 3-9 is to introduce the excess water in the first cavity into the second cavity. A liquid level sensor and a valve are provided in the second cavity. When the water volume is lower than the threshold, the valve is closed, and when the water volume is higher than the threshold, the valve is opened to drain the excess water to the outside or a water storage device.

[0062] In this embodiment, a first L-shaped bracket is provided on the first side (such as the left side) of the seat body 1, and a camera device 5 is provided at the top of the first bracket; the camera device 5 is used to collect the image information of the user using the seat and send the image information to the controller (which can be located in the control center room) wirelessly or wiredly.

[0063] In this embodiment, a temperature acquisition device 4, such as an infrared thermometer gun, is further provided in the middle of the first bracket; the temperature acquisition device 4 is used to measure the temperature Tem of the seat body 1 and send the temperature Tem to the controller wirelessly or wiredly every interval of t time periods.

[0064] In this embodiment, a second bracket is further provided on the second side of the seat body 1, and a flexible screen is provided at the top of the second bracket. The flexible screen is located directly above the seat body 1 and is used to display different background colors, including black background color, yellow background color, and light white background color, which can be adjusted as needed. The light absorption effects of different background colors are different. Black has more heat than yellow, and yellow has more heat than light white.

[0065] In this embodiment, the controller is used to identify the user classification on the seat body 1 according to the image information, so as to adjust the gear of the temperature control device 3 and change the background color of the flexible screen at the same time.

[0066] Based on the above device, as Figure 3 shown, the present invention also provides an adaptive temperature control method for an intelligent seat, specifically including the following steps:

[0067] S1: Obtain the image information of the camera device in real time.

[0068] S2: Identify the user classification according to the image information.

[0069] S2-1: First, download 10,000 pictures of people's faces and appearances at different age periods from the network, and classify users into 5 categories according to age, including a1 (children), a2 (teenagers), a3 (youth), a4 (middle-aged), and a5 (elderly), and store them as the training data set U.

[0070] S2-2: Import the training data set U into the BiGRU (Bidirectional Gated Recurrent Unit) model for training to obtain the user classification model.

[0071] S2-2-1: Initialize the parameters of the gray wolf population, calculate the fitness of the gray wolf population for the first time, and take the gray wolf with the best fitness value as the α-layer wolf, the gray wolf with the second-best fitness (the second) as the β-layer wolf, and the gray wolf with the worst fitness as the δ-layer wolf.

[0072] Among them, the number of the gray wolf population is n M , the maximum number of iterations is T max , the population dimension is D, and the initial gray wolf population is

[0073] S2-2-2: Update the position of the gray wolf:

[0074]

[0075] In formula (1), θ’ α represents the position of the updated α-layer wolf; θ α (t) is the position of the α-layer wolf at the t-th iteration, and θ’ β represents the position of the updated β-layer wolf; θ β (t) is the position of the β-layer wolf at the t-th iteration; θ’ δ represents the position of the updated δ-layer wolf; θ δ (t) is the position of the δ-layer wolf at the t-th iteration;

[0076] A 1 、A 2 、A 3 、C 1 、C 2 、C 3 represent the cooperation coefficients of the α-layer, β-layer, and δ-layer wolves respectively; θ w (t) represents the position of the w-layer wolf at time t; θ w (t + 1) represents the position of the w-layer wolf at time t + 1;

[0077] S2-2-3: Update λ:

[0078]

[0079] In formula (2), λ represents the convergence factor; t represents the number of iterations; T max represents the maximum number of iterations. S2-2-4: Update A and C according to λ:

[0080] A = 2λr 1 -λ, C = 2r 2 (3)

[0081] In formula (3), A and C represent vector coefficients; r 1 and r 2 represent random numbers between [0, 1];

[0082] When |A| > 1, the grey wolf moves away from the prey and looks for better prey. When |A| < 1, the grey wolf must attack the prey. Attack means the prey is the optimal solution at this time, and moving away from the prey means continuing to perform the step to find the optimal solution;

[0083] S2-2-5: Calculate the offspring grey wolves:

[0084]

[0085] In formula (4), is the d-th dimension offspring after the i-th parent, is the d-th dimension offspring after the j-th parent, r 3 and r 4 are random numbers between [0, 1], C 4 and C 5 are random numbers between [-1, 1]; θ i,d represents the d-th dimension position data of the i-th parent in the grey wolf population; θ j,d represents the d-th dimension position data of the ji-th parent in the grey wolf population; φ CO represents the set of offspring grey wolves; represents the n M th offspring grey wolf.

[0086] S2-2-6: Calculate the fitness values of each grey wolf for the second time. When the fitness value calculated for the grey wolf θ k (k ∈ {1, 2,..., n M}) is lower than that of the grey wolf θ CO in the φ k CO population, update the grey wolf θ k to θ k CO, otherwise no adjustment is made, and the iterative calculation is continuously performed. When the maximum number of iterations T max is reached, the operation is stopped. At this time, the optimal individual a-layer wolf is the optimal solution. After decoding the a-layer wolf value, the optimal hyperparameter values of the BiGRU model are obtained. Finally, the training dataset U is imported into the BiGRU model with the optimal hyperparameter values and trained to obtain the trained age recognition and detection BiGRU model, that is, the user classification model.

[0087] S2-3: Input the image information obtained in real time into the trained user classification model and output the user classification.

[0088] In this embodiment, when identifying the user classification, the principle of taking the higher value is adopted: that is, if a3 (youth) and a5 (elderly) exist simultaneously among the users, the output user classification is a5, which can take care of the older users and improve safety.

[0089] S3: Perform temperature adaptive adjustment according to the user classification and the temperature Tem sent by the temperature acquisition device.

[0090] S3-1: When the user classification is a1 or a5, set the temperature threshold T1. When Tem≥T1, perform the first temperature adjustment, that is, adjust the control valve of the temperature control device 3 to the third gear, and the flexible screen displays black;

[0091] S3-2: When the user classification is a2, set the temperature threshold T2. When Tem≥T2, perform the second temperature adjustment, that is, adjust the control valve of the temperature control device 3 to the second gear, and the flexible screen displays yellow;

[0092] S3-3: When the user classification is a3 or a4, set the temperature threshold T3. When Tem≥T3, perform the third temperature adjustment, that is, adjust the control valve of the temperature control device 3 to the first gear, and the flexible screen displays light white.

[0093] In this embodiment, T1>T2>T3.

[0094] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present invention, and in practical applications, various changes can be made in form and details without departing from the spirit and scope of the present invention.

Claims

1. An adaptive temperature control device for a smart seat, characterized in that: It comprises a seat body (1), a pipe (2), a temperature control device (3), a camera device (5) and a controller; The pipe (2) is installed on the lower surface of the bottom plate of the seat body (1), and the pipe (2) is connected to the temperature control device (3); the temperature control device (3), the camera device (5) and the controller are connected wirelessly / wired.

2. The adaptive temperature control device for a smart seat according to claim 1, characterized in that: It also includes a temperature collection device (4) connected to the controller.

3. The adaptive temperature control device for a smart seat according to claim 1, characterized in that: The bottom plate comprises an upper part and a lower part, the upper part is made of wood, the lower part is made of stainless steel, and the lower surface of the lower part is provided with a groove for installing a pipe (2).

4. The adaptive temperature control device for a smart seat according to claim 1, characterized in that: The temperature control device (3) comprises a housing; An air inlet pipeline (3-3) is arranged on the left side of the shell, and a control valve (3-1) is arranged in the air inlet pipeline (3-3); an air outlet pipeline (3-4) is arranged on the right side of the shell; the air inlet pipeline (3-3) is connected to one end of the pipeline (2), and the air outlet pipeline (3-4) is connected to the other end of the pipeline (2); A bearing (3-10) is installed in the first cavity of the shell, and a first exhaust fan (3-5), a first rotating machine (3-6), a first metal grille (3-7), a second rotating machine (3-8), a second exhaust fan (3-12) and a second metal grille (3-13) are installed on the bearing (3-10) in sequence from left to right.

5. The adaptive temperature control device for a smart seat as claimed in claim 4, characterized in that: The first rotating machine (3-6) comprises a first motor and a plurality of metal fan blades, wherein the surfaces of the metal fan blades are provided with circular vents; and a solid desiccant is embedded inside the first metal grille (3-7).

6. The adaptive temperature control device for a smart seat according to claim 4, characterized in that: The second rotating machine (3-7) comprises a second motor, a water supply pipeline, and a plastic fan blade, and cotton cloth pieces are evenly distributed on the surface of the plastic fan blade.

7. The adaptive temperature control device for a smart seat according to claim 4, characterized in that: A spacer metal sheet (3-11) is provided between the first exhaust fan (3-5) and the first rotating machine (3-6).

8. The adaptive temperature control device for a smart seat as claimed in claim 4, characterized in that: A water return pipe (3-9) is arranged at the lower side of the first cavity inside the shell, and an air return line (3-2) is protrudingly arranged at the lower side of the shell, that is, a second cavity is formed between the first cavity and the air return line (3-2).

9. An adaptive temperature control method for a smart seat based on the device according to any one of claims 1 to 8, characterized in that: The following steps are involved: S1: Acquire image information of the camera device in real time; S2: Input the image information into the trained user classification model to output the user classification; S3: Adaptive temperature adjustment based on user classification and seat temperature.

10. The adaptive temperature control method for a smart seat according to claim 9, characterized in that: S2 includes: S2-1: First, download 10,000 pictures of people’s faces and clothing at different ages from the Internet, and divide the users into five categories according to their ages, including a1, a2, a3, a4, and a5, and store them as the training data set U; S2-2: Import the training data set U into the BiGRU model for training, thereby obtaining a user classification model; S2-2-1: Initialize the parameters of the gray wolf population, calculate the first gray wolf population fitness, and take the gray wolf with the best fitness value as the α-layer wolf, the gray wolf with the second best fitness value as the β-layer wolf, and the gray wolf with the worst fitness value as the δ-layer wolf; S2-2-2: Updated the position of the gray wolf: In formula (1), θ, α represents the updated position of the wolf in layer α; θ α (t) is the position of the wolf in layer α at the tth iteration, θ, β represents the updated position of the wolf in layer β; θ β (t) is the position of the wolf in the β layer at the tth iteration; θ, δ represents the updated position of the wolf in layer δ; θ δ (t) is the position of the wolf in layer δ at the tth iteration; A1, A2, A3, C1, C2, C3 represent the synergy coefficients of α-layer, β-layer and δ-layer wolves respectively; θ w (t) represents the position of the wolf in layer w at time t; θ w (t+1) represents the position of the wolf in layer w at time t+1; S2-2-3: Update λ: In formula (2), λ represents the convergence factor; t represents the number of iterations; T max Indicates the maximum number of iterations; S2-2-4: Update A and C according to λ: A=2λr1-λ,C=2r2(3) In formula (3), A and C represent vector coefficients; r1 and r2 represent random numbers between [0,1]; S2-2-5: Calculate the offspring gray wolf: In formula (4), is the d-th dimension offspring after the ith parent, is the d-dimensional offspring after the j-th parent, r3 and r4 are random numbers between [0,1], C4 and C5 are random numbers between [-1,1]; θ i,d represents the d-dimensional position data of the i-th parent in the gray wolf population; θ j,d represents the d-dimensional position data of the j-th parent in the gray wolf population; φ CO represents the set of offspring gray wolves; Indicates the nth M Offspring gray wolf; S2-2-6: Calculate the fitness value of each gray wolf for the second time. When the gray wolf θ in the φ population k (k∈{1,2,...,n M }) The calculated fitness value is lower than φ CO Gray wolf θ in the population k CO When the gray wolf θ k Update to θ k CO Otherwise, no adjustment is made and the operation is repeated continuously. When the maximum number of iterations T is reached max When , the calculation is stopped and the BiGRU model with the optimal hyperparameter value is obtained, that is, the user classification model; S2-3: Input the real-time acquired image information into the trained user classification model and output the user classification.

Images