Office equipment pre-boot method and apparatus based on user behavior prediction

Abstract

The application discloses an office equipment pre-starting method and device based on user behavior prediction, and a server method comprises the following steps: receiving a reservation request sent by a client, the request carrying a use time of a target office, an office ID, a user ID of a target user; when the time length between the current time and the use time is a preset time length, querying historical behavior data of the target user and current environment data of the target office according to the IDs; inputting the historical behavior data and the current environment data into a pre-trained equipment use preference parameter prediction model to output use preference parameters of each equipment in the target office; calculating optimal pre-starting time lengths of the equipment according to the use preference parameters and equipment performance parameters of the equipment; and encapsulating the use preference parameters of the equipment into starting instructions based on the optimal pre-starting time lengths and delivering the starting instructions to an equipment controller. By using the application, work efficiency and office comfort can be improved, and individual office needs of different users can be met.

Description

Technical Field

[0001] This application relates to the field of intelligent environmental sensing and control technology, and in particular to a method and apparatus for pre-starting office equipment based on user behavior prediction. Background Technology

[0002] In modern office environments, users need to wait for office equipment (such as air conditioners, humidifiers, and fresh air systems) to start and adjust to a suitable state when entering meeting rooms, offices, or other workspaces to meet the requirements of comfort and work efficiency. For example, in a meeting room, users need to wait for the air conditioner to adjust to a suitable temperature before the meeting can begin.

[0003] In related technologies, the startup and adjustment of office equipment rely on manual operation or a preset timer system to start the equipment at a fixed time. However, manual adjustment not only wastes time but also causes user discomfort during the waiting process, affecting work efficiency and comfort. Timer-based systems cannot meet the personalized requirements of different users for their office environment, thus reducing the flexibility of office equipment control. Summary of the Invention

[0004] This application provides a method and apparatus for pre-starting office equipment based on user behavior prediction. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.

[0005] In a first aspect, embodiments of this application provide a method for pre-starting office equipment based on user behavior prediction, applied to a server, the method comprising:

[0006] Receive a reservation request sent by the client. The reservation request carries the usage time and office ID of the target office, and the user ID of the target user.

[0007] If the time between the current moment and the time of use is a preset time, query the target user's historical behavior data and the target office's current environmental data based on the user ID and office ID.

[0008] Input historical behavior data and current environment data into a pre-trained device usage preference parameter prediction model, and output the usage preference parameters of each device in the target office corresponding to the target user.

[0009] Calculate the optimal pre-start time for each device based on the usage preference parameters and device performance parameters of each device;

[0010] Based on the optimal pre-startup duration, the usage preference parameters of each device are encapsulated into a startup command and sent to the device controller used to control each device.

[0011] Optionally, based on the usage preference parameters and device performance parameters of each device, the optimal pre-start time for each device is calculated, including:

[0012] The system acquires the current status of each device in real time and determines the target status corresponding to the usage preference parameters of each device.

[0013] Based on the current state, the target state, and the equipment performance parameters of each device, determine the theoretical shortest time required for each device to switch from the current state to the target state;

[0014] Based on the theoretical shortest duration of each device, construct the time interval for each device;

[0015] Obtain a preset objective function, which is used to search for the optimal pre-start time of each device. The optimal pre-start time is used to control the device to achieve the optimal balance between energy consumption and comfort.

[0016] Within the time interval of each device, the optimal pre-startup time that minimizes the output value of the preset objective function is searched to obtain the optimal pre-startup time for each device.

[0017] Optionally, based on the theoretical shortest duration of each device, a time interval is constructed for each device, including:

[0018] Get the current device type of each device;

[0019] Obtain the current buffer duration corresponding to the current device type from the pre-built mapping relationship between device type and buffer duration;

[0020] The current buffer duration corresponding to the current device type is added to the theoretical minimum duration of each device to obtain the initial pre-startup duration of each device.

[0021] The time interval for each device is obtained by combining the theoretical minimum time of each device with the initial pre-start time of each device.

[0022] Optionally, a preset objective function can be generated by following these steps:

[0023] Extract the rated power of each device from its performance parameters;

[0024] Based on the rated power of each device, we construct calculation expressions for energy consumption cost and comfort cost; energy consumption cost is used to reduce the time required for early start-up; comfort cost is used to ensure that the device is in the target state when the user arrives.

[0025] The expressions for calculating energy consumption costs and comfort costs are combined into a preset objective function; the expression for the preset objective function is as follows:

[0026] Cost ;

[0027] Cost The preset target function output value, For any pre-startup duration within the time interval, For energy consumption costs, For comfort costs;

[0028] The formula for calculating energy consumption cost is as follows:

[0029]

[0030] The longer the equipment is turned on in advance, the higher the energy consumption, as determined by a weighting coefficient. and rated power Measuring energy costs, The current cache time corresponding to the current weather type;

[0031] The formula for calculating the cost of comfort is as follows:

[0032] ;

[0033] In this case, if the device fails to reach the target state when the user arrives, a comfort penalty will be incurred, calculated using a weighted coefficient. The cost of comfort is measured by the difference between the actual state and the target state. The actual state when the user arrives. This is the target state.

[0034] Optionally, based on the current state, the target state, and the device performance parameters of each device, determine the theoretical minimum time required for each device to switch from the current state to the target state, including:

[0035] Identify the device type of each device;

[0036] If the equipment type is an air conditioning unit, mark the current status as the current temperature and the target status as the target temperature;

[0037] Extract the maximum cooling / heating rate from the equipment performance parameters of the air conditioning equipment;

[0038] Calculate the absolute value of the difference between the current temperature and the target temperature to obtain the temperature difference;

[0039] Calculate the ratio between the temperature difference and the maximum cooling / heating rate to obtain the theoretical shortest time required for the air conditioning equipment to switch from the current state to the target state.

[0040] Optionally, based on the optimal pre-boot duration, the usage preference parameters of each device are encapsulated into boot instructions, including:

[0041] By using the optimal pre-start duration, the usage time is pre-started and corrected to obtain the corrected pre-start time for each device;

[0042] When the pre-start time of each device is reached at the current time, the usage preference parameters of each device are encapsulated into a start command.

[0043] Optionally, a pre-trained device usage preference parameter prediction model is generated by following these steps:

[0044] Acquire multi-dimensional historical data on each user's office usage; the multi-dimensional historical data includes user behavior records, environmental status snapshots, and user and scene identifiers; user behavior records are the final preference parameter values ​​of each user after manually adjusting the equipment; environmental status snapshots are the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamps recorded by the system when the user manually adjusts the equipment; user and scene identifiers are each user's user ID, the office ID where the equipment is located, and the equipment type;

[0045] Each user's behavior record is used as a tag, and each user's environmental state snapshot, user and scene identifier are used as feature parameters.

[0046] Labels are used to associate and annotate the feature parameters to obtain multiple model training samples. Each model training sample is used to represent a reservation user adjustment event.

[0047] A neural network is used to create a model for predicting device usage preference parameters;

[0048] Based on each model training sample, machine learning is performed on the device preference parameter prediction model to output the model loss value;

[0049] If the model loss value reaches its minimum, a pre-trained device usage preference parameter prediction model is generated; or if the model loss value does not reach its minimum, the step of performing machine learning on the device usage preference parameter prediction model based on each model training sample continues.

[0050] Optionally, generate multi-dimensional historical data on each appointment user's office usage by following these steps:

[0051] During each user's use of the office, the current settings of each device in the office are monitored in real time.

[0052] In response to each user's parameter adjustment command, obtain the final preference parameter value of each user after manually adjusting the device, and record it as user behavior.

[0053] The system synchronously records the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamp of each user's office, serving as a snapshot of the environmental conditions.

[0054] Synchronously record each user's user ID, the office ID where the device is located, and the device type as user and scenario identifiers;

[0055] Store user behavior records, environmental status snapshots, and the relationship between users and scene identifiers for each user who made a reservation, to obtain multi-dimensional historical data on each user's use of the office.

[0056] Optionally, based on each model training sample, machine learning is performed on the device using the preference parameter prediction model, including:

[0057] Perform feature engineering and feature encoding operations on the environmental state snapshots, user and scene identifiers in each model training sample to obtain the feature data of each model training sample;

[0058] The feature data of each model training sample is input into the device to predict the model using preference parameters, and the model prediction result of each model training sample is obtained.

[0059] The model prediction result and the label of each training sample are substituted into the preset loss function to obtain the model loss value; the function expression of the preset loss function is:

[0060]

[0061] in, It is the loss value. It is the number of training samples for the model. It is the first The labels of the training samples for each model. It is the first The model prediction results for each training sample.

[0062] Secondly, embodiments of this application provide an office equipment pre-start device based on user behavior prediction, the device comprising:

[0063] The request receiving module is used to receive reservation requests sent by the client. The reservation request carries the usage time of the target office and the office ID, as well as the user ID of the target user.

[0064] The data query module is used to query the historical behavior data of the target user and the current environmental data of the target office based on the user ID and office ID when the time between the current time and the time of use is a preset time.

[0065] The preference parameter output module is used to input historical behavior data and current environment data into a pre-trained device usage preference parameter prediction model and output the usage preference parameters of each device in the target office corresponding to the target user.

[0066] The optimal pre-startup time calculation module is used to calculate the optimal pre-startup time for each device based on the usage preference parameters and device performance parameters of each device.

[0067] The instruction generation and distribution module is used to encapsulate the usage preference parameters of each device into a startup instruction based on the optimal pre-startup duration, and then distribute it to the device controller used to control each device.

[0068] The technical solutions provided in this application embodiment may include the following beneficial effects:

[0069] In this embodiment, on the one hand, based on the user's historical behavior data and current environmental data, and combined with a pre-trained device usage preference parameter prediction model, the system can automatically predict and adjust the device usage preference parameters. Each device will start up and adjust to the user's preferred state within the optimal pre-start time, eliminating the need for manual device adjustment, saving time, and improving work efficiency and office comfort. On the other hand, the system calculates the optimal pre-start time for each device based on each user's historical behavior data and current environmental data, so that device startup and adjustment are no longer fixed timed startup modes, but can be dynamically adjusted according to the optimal pre-start time of each device. Each user can be in an office environment that matches their preferences, greatly improving the flexibility of office equipment control and meeting the personalized office needs of different users.

[0070] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0071] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0072] Figure 1 This is a flowchart illustrating a method for pre-starting office equipment based on user behavior prediction, as provided in an embodiment of this application.

[0073] Figure 2 This is a schematic diagram of a reservation interface provided in an embodiment of this application;

[0074] Figure 3 This is a schematic diagram of a model prediction result provided in an embodiment of this application;

[0075] Figure 4 This is a schematic diagram of an optimal pre-startup duration search process provided in an embodiment of this application;

[0076] Figure 5 This is a schematic block diagram of an office equipment pre-start process based on user behavior prediction provided in an embodiment of this application;

[0077] Figure 6 This is a flowchart illustrating a model training method for a device usage preference parameter prediction model provided in an embodiment of this application.

[0078] Figure 7 This is a schematic diagram of the structure of an office equipment pre-start device based on user behavior prediction provided in an embodiment of this application;

[0079] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0080] The following description and accompanying drawings fully illustrate specific embodiments of this application to enable those skilled in the art to practice them.

[0081] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0082] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0083] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0084] Currently, the startup and adjustment of office equipment rely on manual operation or starting the equipment at fixed times through a preset timer system.

[0085] The inventors realized that manual adjustment not only wasted time but also caused user discomfort while waiting, affecting work efficiency and comfort. Timed adjustments could not meet the personalized requirements of different users for their office environment, thus reducing the flexibility of office equipment control.

[0086] To address the existing technical problems, this application provides a method and apparatus for pre-starting office equipment based on user behavior prediction, thereby resolving the issues mentioned above. In the embodiments of this application, on the one hand, based on the user's historical behavior data and current environmental data, and combined with a pre-trained device usage preference parameter prediction model, the system can automatically predict and adjust the device's usage preference parameters. Each device will start up and adjust to the user's preferred state within the optimal pre-start time, eliminating the need for manual device adjustment, saving time, and improving work efficiency and office comfort. On the other hand, the system calculates the optimal pre-start time for each device based on each user's historical behavior data and current environmental data, so that device startup and adjustment are no longer fixed timed startup modes, but can be dynamically adjusted according to the optimal pre-start time of each device. Each user can be in an office environment that matches their preferences, greatly improving the flexibility of office equipment control and meeting the personalized office needs of different users. The following is a detailed description using exemplary embodiments.

[0087] The following will be combined with the appendix Figure 1 -Appendix Figure 6 This application provides a detailed description of the office equipment pre-start method based on user behavior prediction provided in its embodiments. This method can be implemented using a computer program and can run on an office equipment pre-start device based on user behavior prediction and the von Neumann architecture. The computer program can be integrated into an application or run as a standalone utility application.

[0088] Please see Figure 1 This document provides a flowchart illustrating a method for pre-starting office equipment based on user behavior prediction, applicable to the server side. For example... Figure 1 As shown, the method in this application embodiment includes the following steps:

[0089] S101, receive the reservation request sent by the client. The reservation request carries the usage time of the target office and the office ID, and the user ID of the target user.

[0090] In this context, the client refers to the terminal device or software used by the user to send requests to the server. In an office setting, the client can be an application on an employee's mobile phone, computer, or other smart device. The appointment request is a message sent by the client to the server, containing information about the user's desired office reservation. The target office's availability time refers to the specific time the user plans to use the target office. The office ID is a unique identifier for an office (number or code). The target user's user ID is a unique identifier for a user (number or code).

[0091] In some embodiments of this application, the user inputs reservation information via a client (such as a mobile app or computer client), including the available time of the target office, the office ID, and the user ID. The client encapsulates this information into a reservation request. The client sends the reservation request to the server via a network (such as HTTP / HTTPS protocol). The server receives the reservation request sent by the client, parses the request content, and extracts the available time of the target office, the office ID, and the user ID.

[0092] In one possible implementation, the user selects meeting room C201 (private office) on the booking interface. The booking interface is as follows: Figure 2 As shown. The user selects January 1, 2026 as the reservation date, with a start time of 09:00 and an end time of 17:00. The user enters that there will be one participant and the purpose of the meeting is a meeting. The user clicks the "Submit Reservation" button to send the reservation information to the server. The server receives the reservation request from the client, which includes the following information: Office ID is C201, the usage time is January 1, 2026, 09:00, and the user ID is "123456".

[0093] S102, when the duration between the current time and the usage time is a preset duration, query the target user's historical behavior data and the target office's current environmental data based on the user ID and office ID;

[0094] Here, "current time" refers to the actual time when the system receives the reservation request. "Usage time" refers to the specific time the user plans to use the office as specified in the reservation request. "Preset duration" is a pre-set time length used to execute the office equipment pre-start method based on user behavior prediction. "Historical behavior data of the target user" is a record of the target user's manual adjustments during past office use, such as manually set equipment parameters. "Current environmental data of the target office" is the environmental state of the target office at the current time, such as indoor and outdoor temperature and humidity, and light intensity.

[0095] In some embodiments of this application, the system calculates the time difference between the current time and the time of use. If the time difference equals a preset duration, the system begins the pre-start process. The system queries the target user's historical behavior data, including device parameter settings, based on the user ID. The system queries the target office's current environmental data, such as indoor and outdoor temperature and humidity, and light intensity, based on the office ID.

[0096] In one possible implementation, the server receives a reservation request from the client, containing the following information: Office ID is C201, usage time is January 1, 2026, 09:00, and User ID is "123456". The pre-start time is determined; for example, the current time is January 1, 2026, 08:00, which is one hour different from the usage time of 09:00. The preset duration is 30 minutes, so the system starts the pre-start process at 08:30. The system queries Xiao Zhang's historical behavior data, such as preferred indoor temperature of 23-25℃, humidity of 40-50%, and lighting brightness of 300-400 lx. The system queries the current environmental data of meeting room C201 to obtain indoor and outdoor temperature, humidity, light intensity, etc. The queried historical behavior data and current environmental data are shown in Table 1.

[0097] Table 1

[0098]

[0099] S103: Input historical behavior data and current environmental data into a pre-trained device usage preference parameter prediction model, and output the usage preference parameters of each device in the target office corresponding to the target user;

[0100] Among them, the device usage preference parameter prediction model is a pre-trained machine learning model used to predict users' device usage preferences based on historical behavioral data and current environmental data.

[0101] In some embodiments of this application, historical behavioral data of the target user and current environmental data of the target office are collected. The collected data is cleaned, formatted, and standardized to make it suitable for input into the prediction model. The preprocessed historical behavioral data and current environmental data are input into a pre-trained device usage preference parameter prediction model. The model predicts the target user's usage preference parameters for each device in the target office based on the input data. The model outputs the prediction results, including the target user's usage preference parameters for each device in the target office.

[0102] Among them, the prediction results are as follows: Figure 3As shown. The predicted target temperature for the air conditioning unit is 24°C, the predicted operating mode is cooling, and the predicted fan speed is medium. The predicted target humidity for the humidifier is 45%, and the predicted operating level is level 3. The predicted fan speed for the fresh air system is low, and the predicted operating time is 30 minutes. The predicted brightness for the lighting system is 350 lx, and the activation method is real-time activation upon user arrival.

[0103] It should be noted that the lighting equipment is marked as "on in real time" because it can reach the target brightness without preheating, which is consistent with the physical characteristics of lighting equipment. This application mainly optimizes equipment that requires pre-start (such as air conditioners, humidifiers, etc.).

[0104] S104, Calculate the optimal pre-start time for each device based on the usage preference parameters and device performance parameters of each device;

[0105] Equipment performance parameters refer to the technical specifications and performance indicators of the equipment, such as the cooling / heating rate of an air conditioner and the humidification rate of a humidifier. These parameters describe the time and resources required for the equipment to reach a specific operating state and are key factors in calculating the pre-start-up time. The optimal pre-start-up time is the time the equipment needs to start up in advance to ensure it reaches its ideal state when the user needs it. The optimal pre-start-up time ensures that the equipment is adjusted to the user's preferred state when the user arrives or needs to use it, while avoiding unnecessary energy waste.

[0106] In some embodiments of this application, the specific process of calculating the optimal pre-start time of each device based on the usage preference parameters and device performance parameters of each device includes: acquiring the current state of each device in real time and determining the target state corresponding to the usage preference parameters of each device; determining the theoretical shortest time required for each device to switch from the current state to the target state based on the current state, the target state, and the device performance parameters of each device; constructing the time interval of each device based on the theoretical shortest time of each device; obtaining a preset objective function, which is used to search for the optimal pre-start time of each device, and the optimal pre-start time is used to control the device to achieve the optimal balance between energy consumption and comfort; and searching for the optimal pre-start time that minimizes the output value of the preset objective function within the time interval of each device, thereby obtaining the optimal pre-start time of each device.

[0107] The current state of each device refers to its actual operating state at the current moment, such as the current temperature and humidity of an air conditioner. Since devices may be subject to administrator adjustments, real-time acquisition of the device's current state is necessary. The target state corresponding to the user preference parameters is the specific parameters the device should achieve based on user preferences, such as temperature and humidity. Device performance parameters describe the device's performance, such as the cooling / heating rate of an air conditioner and the humidification rate of a humidifier. The theoretical minimum time is the shortest time required for the device to adjust from its current state to the target state. The time interval is a time range set for device pre-start based on the theoretical minimum time. The preset objective function is a mathematical model used to optimize the device pre-start time, considering energy consumption and comfort. The optimal pre-start time is the device pre-start time that achieves the best balance between energy consumption and comfort, ensuring the device reaches its ideal state when needed by the user while minimizing energy consumption.

[0108] Specifically, the process of determining the theoretical shortest time required for each device to switch from the current state to the target state, based on the current state, the target state, and the device performance parameters of each device, includes: identifying the device type of each device; if the device type is an air conditioning device, marking the current state as the current temperature and the target state as the target temperature; extracting the maximum cooling / heating rate from the device performance parameters of the air conditioning device; calculating the absolute value of the difference between the current temperature and the target temperature to obtain the temperature difference; and calculating the ratio between the temperature difference and the maximum cooling / heating rate to obtain the theoretical shortest time required for the air conditioning device to switch from the current state to the target state.

[0109] Specifically, the process of constructing the time interval for each device based on its theoretical minimum duration includes: obtaining the current device type of each device; obtaining the current buffer duration corresponding to the current device type from the pre-built mapping relationship between device type and buffer time; adding the current buffer duration corresponding to the current device type to the theoretical minimum duration of each device to obtain the initial pre-startup duration of each device; and combining the theoretical minimum duration of each device with the initial pre-startup duration of each device to form a time interval to obtain the time interval for each device.

[0110] Specifically, the process of generating the preset objective function includes: extracting the rated power of each device from its performance parameters; constructing calculation expressions for energy consumption cost and comfort cost based on the rated power of each device; using energy consumption cost to reduce the time required for early startup; using comfort cost to ensure that the device reaches the target state when the user arrives; and combining the calculation expressions for energy consumption cost and comfort cost into the preset objective function.

[0111] Specifically, the preset objective function expression is as follows:

[0112] Cost ;

[0113] Cost The preset target function output value, For any pre-startup duration within the time interval, For energy consumption costs, For comfort costs;

[0114] The formula for calculating energy consumption cost is as follows:

[0115]

[0116] The longer the equipment is turned on in advance, the higher the energy consumption, as determined by a weighting coefficient. and rated power Measuring energy costs, The current cache time corresponding to the current weather type;

[0117] The formula for calculating the cost of comfort is as follows:

[0118] ;

[0119] In this case, if the device fails to reach the target state when the user arrives, a comfort penalty will be incurred, calculated using a weighted coefficient. The cost of comfort is measured by the difference between the actual state and the target state. The actual state when the user arrives. This is the target state.

[0120] In one possible implementation, the system monitors the current indoor temperature in real time as 28°C. Based on user preference, the target temperature for the air conditioner is 24°C. It takes 2 minutes for the air conditioner to adjust from 28°C to 24°C ((28°C - 24°C) ÷ 2°C / minute). The system sets a pre-start time interval of 2 to 5 minutes for the air conditioner, considering the theoretical minimum duration and buffer time. The system constructs an objective function, considering energy consumption (the longer the pre-start time, the higher the energy consumption) and comfort (ensuring the air conditioner reaches 24°C when the user arrives). Within the 2-5 minute time interval, the system uses an optimization algorithm to find the optimal pre-start time. Assume the optimal pre-start time found is 3 minutes. The system starts the air conditioner 3 minutes before the user arrives, adjusting it to 24°C in advance.

[0121] For example Figure 4 As shown, Figure 4This is a schematic diagram illustrating the optimal pre-start time search process provided in this application. First, the current operating status of the device (such as an air conditioner, humidifier, etc.), including temperature and humidity, is obtained. Based on user preferences, the target state the device needs to achieve is determined. Based on the current and target states, the theoretical shortest time required for the device to adjust from the current state to the target state is calculated. The buffer time corresponding to the device type is obtained, and the theoretical shortest time and the buffer time are added to obtain the initial pre-start time. The theoretical shortest time and the initial pre-start time are combined to form the device's time interval. The rated power of the device is extracted from its performance parameters, and an expression for calculating energy consumption cost and comfort cost is constructed based on the rated power. The expressions for energy consumption cost and comfort cost are combined into a preset objective function, which is used to find the optimal balance between energy consumption and comfort. The optimal pre-start time is searched within the set time interval. The output value of the objective function is calculated under different pre-start times. The pre-start time that minimizes the output value of the objective function is found. The optimal pre-start time is determined. The calculated optimal pre-start time is applied to device control.

[0122] S105, based on the optimal pre-startup time, encapsulates the usage preference parameters of each device into a startup command and sends it to the device controller used to control each device.

[0123] In some embodiments of this application, the specific process of encapsulating the usage preference parameters of each device into a startup instruction based on the optimal pre-startup duration and sending it to the device controller used to control each device includes: using the optimal pre-startup duration to pre-startup correct the usage time to obtain the corrected pre-startup time of each device; when the current time reaches the corrected pre-startup time of each device, encapsulating the usage preference parameters of each device into a startup instruction.

[0124] The pre-start correction adjusts the original usage time based on the optimal pre-start duration to determine the actual start-up time required for the device. The corrected pre-start time is the actual start-up time after the pre-start correction. User preference parameters are the user's personal settings for the device's operating status, such as temperature, humidity, and brightness.

[0125] For example Figure 5 As shown, Figure 5This application provides a schematic block diagram of an office equipment pre-start process based on user behavior prediction. The server receives a reservation request from the client, including the target office's usage time, office ID, and user ID. Based on the user ID and office ID, it queries the target user's historical behavior data and the target office's current environmental data. The retrieved historical behavior data and current environmental data are input into a pre-trained device usage preference parameter prediction model. The model outputs the target user's usage preference parameters for each device in the target office. Based on the device usage preference parameters and device performance parameters, the optimal pre-start time for each device is calculated to ensure the device is in an ideal state upon user arrival while minimizing energy consumption. A preset objective function (considering energy consumption and comfort) is used to determine the optimal pre-start time. The device usage preference parameters are encapsulated into start commands. The optimal pre-start time is used to pre-start the usage time, resulting in the corrected pre-start time for each device. When the current time reaches the corrected pre-start time for each device, the encapsulated start command is sent to the device controller.

[0126] In this embodiment, on the one hand, based on the user's historical behavior data and current environmental data, and combined with a pre-trained device usage preference parameter prediction model, the system can automatically predict and adjust the device usage preference parameters. Each device will start up and adjust to the user's preferred state within the optimal pre-start time, eliminating the need for manual device adjustment, saving time, and improving work efficiency and office comfort. On the other hand, the system calculates the optimal pre-start time for each device based on each user's historical behavior data and current environmental data, so that device startup and adjustment are no longer fixed timed startup modes, but can be dynamically adjusted according to the optimal pre-start time of each device. Each user can be in an office environment that matches their preferences, greatly improving the flexibility of office equipment control and meeting the personalized office needs of different users.

[0127] Please see Figure 6 This is a flowchart illustrating a method for generating a user state determination model, as provided in an embodiment of this application. Figure 6 As shown, the method in this application embodiment may include the following steps:

[0128] S201, acquire multi-dimensional historical data of each user's office usage; multi-dimensional historical data includes user behavior records, environmental status snapshots, and user and scene identifiers; user behavior records are the final preference parameter values ​​of each user after manually adjusting the equipment; environmental status snapshots are the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamps recorded by the system when the user manually adjusts the equipment; user and scene identifiers are each user's user ID, the office ID where the equipment is located, and the equipment type;

[0129] In some embodiments of this application, the specific process of generating multi-dimensional historical data on each user's use of the office includes: during each user's use of the office, real-time monitoring of the current settings parameters of each device in the office; in response to each user's parameter adjustment command, obtaining the final preference parameter value of each user after manually adjusting the device, as a user behavior record; synchronously recording the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamp of each user's office, as an environmental state snapshot; synchronously recording each user's user ID, the office ID of the device, and the device type, as user and scene identifiers; storing the user behavior record, environmental state snapshot, and the association between the user and scene identifiers for each user, thus obtaining multi-dimensional historical data on each user's use of the office.

[0130] S202, use the user behavior records of each reservation user as tags, and use the environmental state snapshot, user and scene identifiers of each reservation user as feature parameters;

[0131] S203, labels are used to associate and annotate the feature parameters to obtain multiple model training samples. Each model training sample is used to represent a reservation user adjustment event.

[0132] S204, uses a neural network to create a device usage preference parameter prediction model;

[0133] S205, based on each model training sample, performs machine learning on the device use preference parameter prediction model and outputs the model loss value;

[0134] In some embodiments of this application, the specific process of performing machine learning on the device usage preference parameter prediction model based on each model training sample includes: performing feature engineering on the environmental state snapshot and user and scene identifiers in each model training sample to obtain feature data for each model training sample; inputting the feature data of each model training sample into the device usage preference parameter prediction model to obtain the model prediction result for each model training sample; substituting the model prediction result and the label of each model training sample into a preset loss function to obtain the model loss value; wherein, the function expression of the preset loss function is:

[0135]

[0136] in, It is the loss value. It is the number of training samples for the model. It is the first The labels of the training samples for each model. It is the first The model prediction results for each training sample.

[0137] S206, if the model loss value reaches the minimum, generate a pre-trained device usage preference parameter prediction model; or if the model loss value does not reach the minimum, continue to perform the step of machine learning on the device usage preference parameter prediction model based on each model training sample.

[0138] In this embodiment, on the one hand, based on the user's historical behavior data and current environmental data, and combined with a pre-trained device usage preference parameter prediction model, the system can automatically predict and adjust the device usage preference parameters. Each device will start up and adjust to the user's preferred state within the optimal pre-start time, eliminating the need for manual device adjustment, saving time, and improving work efficiency and office comfort. On the other hand, the system calculates the optimal pre-start time for each device based on each user's historical behavior data and current environmental data, so that device startup and adjustment are no longer fixed timed startup modes, but can be dynamically adjusted according to the optimal pre-start time of each device. Each user can be in an office environment that matches their preferences, greatly improving the flexibility of office equipment control and meeting the personalized office needs of different users.

[0139] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0140] Please see Figure 7 This illustration shows a schematic diagram of an office equipment pre-start device based on user behavior prediction, provided in an exemplary embodiment of this application. This office equipment pre-start device based on user behavior prediction can be implemented as all or part of an electronic device through software, hardware, or a combination of both. The device 1 includes a request receiving module 10, a data query module 20, a preference parameter output module 30, an optimal pre-start duration calculation module 40, and an instruction generation and distribution module 50.

[0141] The request receiving module 10 is used to receive the reservation request sent by the client. The reservation request carries the usage time of the target office and the office ID, and the user ID of the target user.

[0142] The data query module 20 is used to query the historical behavior data of the target user and the current environmental data of the target office based on the user ID and office ID when the time between the current time and the time of use is a preset time.

[0143] The preference parameter output module 30 is used to input historical behavior data and current environment data into a pre-trained device usage preference parameter prediction model and output the usage preference parameters of each device in the target office corresponding to the target user.

[0144] The optimal pre-startup time calculation module 40 is used to calculate the optimal pre-startup time of each device based on the usage preference parameters and device performance parameters of each device.

[0145] The instruction generation and distribution module 50 is used to encapsulate the usage preference parameters of each device into a startup instruction based on the optimal pre-startup duration and distribute it to the device controller used to control each device.

[0146] It should be noted that the above embodiments of the office equipment pre-start device based on user behavior prediction, when executing the office equipment pre-start method based on user behavior prediction, are only illustrative examples of the above-described functional module divisions. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the office equipment pre-start device based on user behavior prediction and the office equipment pre-start method embodiments provided above belong to the same concept, and their implementation process is detailed in the method embodiments, which will not be repeated here.

[0147] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0148] In this embodiment, on the one hand, based on the user's historical behavior data and current environmental data, and combined with a pre-trained device usage preference parameter prediction model, the system can automatically predict and adjust the device usage preference parameters. Each device will start up and adjust to the user's preferred state within the optimal pre-start time, eliminating the need for manual device adjustment, saving time, and improving work efficiency and office comfort. On the other hand, the system calculates the optimal pre-start time for each device based on each user's historical behavior data and current environmental data, so that device startup and adjustment are no longer fixed timed startup modes, but can be dynamically adjusted according to the optimal pre-start time of each device. Each user can be in an office environment that matches their preferences, greatly improving the flexibility of office equipment control and meeting the personalized office needs of different users.

[0149] This application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implement the office equipment pre-start method based on user behavior prediction provided in the above-described method embodiments.

[0150] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to execute the office equipment pre-start method based on user behavior prediction of the above-described method embodiments.

[0151] Please see Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 8 As shown, the electronic device 1000 may include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002.

[0152] The communication bus 1002 is used to realize the connection and communication between these components.

[0153] The user interface 1003 may include a display screen and a camera. Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface.

[0154] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0155] The processor 1001 may include one or more processing cores. The processor 1001 connects to various parts within the electronic device 1000 using various interfaces and lines. It executes various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and by calling data stored in the memory 1005. Optionally, the processor 1001 may be implemented using at least one hardware form selected from Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The processor 1001 may integrate one or more of the following: a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed on the screen; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip, without being integrated into the processor 1001.

[0156] The memory 1005 may include random access memory (RAM) or read-only memory. Optionally, the memory 1005 may include a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 1005 may also be at least one storage system located remotely from the aforementioned processor 1001. Figure 8 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an office equipment pre-boot application based on user behavior prediction.

[0157] exist Figure 8 In the illustrated electronic device 1000, the user interface 1003 is mainly used to provide an input interface for the user and to obtain user input data; while the processor 1001 can be used to call the office device pre-boot application based on user behavior prediction stored in the memory 1005, and specifically perform the following operations:

[0158] Receive a reservation request sent by the client. The reservation request carries the usage time and office ID of the target office, and the user ID of the target user.

[0159] If the time between the current moment and the time of use is a preset time, query the target user's historical behavior data and the target office's current environmental data based on the user ID and office ID.

[0160] Input historical behavior data and current environment data into a pre-trained device usage preference parameter prediction model, and output the usage preference parameters of each device in the target office corresponding to the target user.

[0161] Calculate the optimal pre-start time for each device based on the usage preference parameters and device performance parameters of each device;

[0162] Based on the optimal pre-startup duration, the usage preference parameters of each device are encapsulated into a startup command and sent to the device controller used to control each device.

[0163] In one embodiment, when the processor 1001 calculates the optimal pre-startup time for each device based on the usage preference parameters and device performance parameters of each device, it specifically performs the following operations:

[0164] The system acquires the current status of each device in real time and determines the target status corresponding to the usage preference parameters of each device.

[0165] Based on the current state, the target state, and the equipment performance parameters of each device, determine the theoretical shortest time required for each device to switch from the current state to the target state;

[0166] Based on the theoretical shortest duration of each device, construct the time interval for each device;

[0167] Obtain a preset objective function, which is used to search for the optimal pre-start time of each device. The optimal pre-start time is used to control the device to achieve the optimal balance between energy consumption and comfort.

[0168] Within the time interval of each device, the optimal pre-startup time that minimizes the output value of the preset objective function is searched to obtain the optimal pre-startup time for each device.

[0169] In one embodiment, when the processor 1001 constructs the time intervals for each device based on the theoretical shortest duration of each device, it specifically performs the following operations:

[0170] Get the current device type of each device;

[0171] Obtain the current buffer duration corresponding to the current device type from the pre-built mapping relationship between device type and buffer duration;

[0172] The current buffer duration corresponding to the current device type is added to the theoretical minimum duration of each device to obtain the initial pre-startup duration of each device.

[0173] The time interval for each device is obtained by combining the theoretical minimum time of each device with the initial pre-start time of each device.

[0174] In one embodiment, when the processor 1001 executes the function to generate a preset target function, it performs the following operations:

[0175] Extract the rated power of each device from its performance parameters;

[0176] Based on the rated power of each device, we construct calculation expressions for energy consumption cost and comfort cost; energy consumption cost is used to reduce the time required for early start-up; comfort cost is used to ensure that the device is in the target state when the user arrives.

[0177] The calculation expressions for energy consumption costs and comfort costs are combined into a preset objective function.

[0178] In one embodiment, when the processor 1001 determines the theoretical shortest time required for each device to switch from the current state to the target state based on the current state, the target state, and the device performance parameters of each device, it specifically performs the following operations:

[0179] Identify the device type of each device;

[0180] If the equipment type is an air conditioning unit, mark the current status as the current temperature and the target status as the target temperature;

[0181] Extract the maximum cooling / heating rate from the equipment performance parameters of the air conditioning equipment;

[0182] Calculate the absolute value of the difference between the current temperature and the target temperature to obtain the temperature difference;

[0183] Calculate the ratio between the temperature difference and the maximum cooling / heating rate to obtain the theoretical shortest time required for the air conditioning equipment to switch from the current state to the target state.

[0184] In one embodiment, when the processor 1001 executes a boot instruction that encapsulates the usage preference parameters of each device based on the optimal pre-boot duration, it specifically performs the following operations:

[0185] By using the optimal pre-start duration, the usage time is pre-started and corrected to obtain the corrected pre-start time for each device;

[0186] When the pre-start time of each device is reached at the current time, the usage preference parameters of each device are encapsulated into a start command.

[0187] In one embodiment, when the processor 1001 executes the generation of a pre-trained device-use preference parameter prediction model, it specifically performs the following operations:

[0188] Acquire multi-dimensional historical data on each user's office usage; the multi-dimensional historical data includes user behavior records, environmental status snapshots, and user and scene identifiers; user behavior records are the final preference parameter values ​​of each user after manually adjusting the equipment; environmental status snapshots are the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamps recorded by the system when the user manually adjusts the equipment; user and scene identifiers are each user's user ID, the office ID where the equipment is located, and the equipment type;

[0189] Each user's behavior record is used as a tag, and each user's environmental state snapshot, user and scene identifier are used as feature parameters.

[0190] Labels are used to associate and annotate the feature parameters to obtain multiple model training samples. Each model training sample is used to represent a reservation user adjustment event.

[0191] A neural network is used to create a model for predicting device usage preference parameters;

[0192] Based on each model training sample, machine learning is performed on the device preference parameter prediction model to output the model loss value;

[0193] If the model loss value reaches its minimum, a pre-trained device usage preference parameter prediction model is generated; or if the model loss value does not reach its minimum, the step of performing machine learning on the device usage preference parameter prediction model based on each model training sample continues.

[0194] In one embodiment, when the processor 1001 generates multi-dimensional historical data on each user's office usage, it specifically performs the following operations:

[0195] During each user's use of the office, the current settings of each device in the office are monitored in real time.

[0196] In response to each user's parameter adjustment command, obtain the final preference parameter value of each user after manually adjusting the device, and record it as user behavior.

[0197] The system synchronously records the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamp of each user's office, serving as a snapshot of the environmental conditions.

[0198] Synchronously record each user's user ID, the office ID where the device is located, and the device type as user and scenario identifiers;

[0199] Store user behavior records, environmental status snapshots, and the relationship between users and scene identifiers for each user who made a reservation, to obtain multi-dimensional historical data on each user's use of the office.

[0200] In one embodiment, when processor 1001 performs machine learning on the device using preference parameters to predict the model based on each model training sample, it specifically performs the following operations:

[0201] Perform feature engineering and feature encoding operations on the environmental state snapshots, user and scene identifiers in each model training sample to obtain the feature data of each model training sample;

[0202] The feature data of each model training sample is input into the device to predict the model using preference parameters, and the model prediction result of each model training sample is obtained.

[0203] The model prediction results and labels of each training sample are substituted into a preset loss function to obtain the model loss value.

[0204] In this embodiment, on the one hand, based on the user's historical behavior data and current environmental data, and combined with a pre-trained device usage preference parameter prediction model, the system can automatically predict and adjust the device usage preference parameters. Each device will start up and adjust to the user's preferred state within the optimal pre-start time, eliminating the need for manual device adjustment, saving time, and improving work efficiency and office comfort. On the other hand, the system calculates the optimal pre-start time for each device based on each user's historical behavior data and current environmental data, so that device startup and adjustment are no longer fixed timed startup modes, but can be dynamically adjusted according to the optimal pre-start time of each device. Each user can be in an office environment that matches their preferences, greatly improving the flexibility of office equipment control and meeting the personalized office needs of different users.

[0205] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The pre-boot program for office equipment based on user behavior prediction can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The storage medium for the pre-boot program for office equipment based on user behavior prediction can be a magnetic disk, optical disk, read-only memory, or random access memory, etc.

[0206] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A method for pre-starting office equipment based on user behavior prediction, characterized in that, Applied to the server side, the method includes: Receive a reservation request sent by the client, the reservation request carrying the usage time of the target office and the office ID, and the user ID of the target user; When the duration between the current moment and the usage moment is a preset duration, query the target user's historical behavior data and the target office's current environmental data based on the user ID and office ID; The historical behavior data and current environment data are input into a pre-trained device usage preference parameter prediction model, which outputs the usage preference parameters of each device in the target office corresponding to the target user. Based on the usage preference parameters and device performance parameters of each device, the optimal pre-start time for each device is calculated, including: acquiring the current state of each device in real time and determining the target state corresponding to the usage preference parameters of each device; determining the theoretical shortest time required for each device to switch from the current state to the target state based on the current state, the target state, and the device performance parameters of each device; constructing the time interval for each device based on the theoretical shortest time; obtaining a preset objective function, which is used to search for the optimal pre-start time for each device, and the optimal pre-start time is used to control the device to achieve an optimal balance between energy consumption and comfort; and searching for the optimal pre-start time that minimizes the output value of the preset objective function within the time interval of each device, thereby obtaining the optimal pre-start time for each device. The preset objective function is generated according to the following steps: extracting the rated power of each device from its performance parameters; constructing calculation expressions for energy consumption cost and comfort cost based on the rated power of each device; the energy consumption cost is used to reduce the time required for early startup; the comfort cost is used to ensure that the device reaches the target state when the user arrives; and combining the calculation expressions for energy consumption cost and comfort cost into the preset objective function; wherein the function expression of the preset objective function is: ； in, Output the value of the preset objective function. For any pre-start duration within the time interval, For energy consumption costs, For comfort costs; The energy consumption cost is calculated using the following expression: ； The longer the equipment is turned on in advance, the higher the energy consumption, as determined by a weighting coefficient. and rated power Measuring energy costs, The current cache time corresponding to the current weather type; The formula for calculating the comfort cost is as follows: ； In this case, if the device fails to reach the target state when the user arrives, a comfort penalty will be incurred, calculated using a weighted coefficient. The cost of comfort is measured by the difference between the actual state and the target state. The actual state when the user arrives. The target state; Based on the optimal pre-startup duration, the usage preference parameters of each device are encapsulated into a startup command and sent to the device controller used to control each device.

2. The method according to claim 1, characterized in that, The step of constructing the time interval for each device based on the theoretical shortest duration of each device includes: Obtain the current device type of each device; Obtain the current buffer duration corresponding to the current device type from the pre-built mapping relationship between device type and buffer time; The current buffer duration corresponding to the current device type is added to the theoretical minimum duration of each device to obtain the initial pre-startup duration of each device. The theoretical minimum duration of each device is combined with the initial pre-start duration of each device to form a time interval, thus obtaining the time interval of each device.

3. The method according to claim 1, characterized in that, The step of determining the theoretical shortest time required for each device to switch from the current state to the target state based on the current state, the target state, and the device performance parameters of each device includes: Identify the device type of each device; When the device type is an air conditioning device, the current state is marked as the current temperature and the target state is marked as the target temperature; The maximum cooling / heating rate is extracted from the equipment performance parameters of the air conditioning equipment. Calculate the absolute value of the difference between the current temperature and the target temperature to obtain the temperature difference; Calculate the ratio between the temperature difference and the maximum cooling / heating rate to obtain the theoretical shortest time required for the air conditioning equipment to switch from the current state to the target state.

4. The method according to claim 1, characterized in that, Based on the optimal pre-startup duration, the usage preference parameters of each device are encapsulated into a startup instruction, including: Using the optimal pre-start duration, the usage time is pre-started and corrected to obtain the corrected pre-start time for each device; When the current time reaches the pre-start time of each device, the usage preference parameters of each device are encapsulated into a start command.

5. The method according to any one of claims 1-4, characterized in that, Generate a pre-trained device usage preference parameter prediction model by following these steps: Acquire multi-dimensional historical data on each user's use of the office; the multi-dimensional historical data includes user behavior records, environmental status snapshots, and user and scene identifiers; the user behavior records are the final preference parameter values ​​of each user after manually adjusting the device; the environmental status snapshots are the office ID, indoor and outdoor temperature and humidity, light intensity, and timestamps recorded by the system when the user manually adjusts the device; the user and scene identifiers are each user's user ID, the office ID where the device is located, and the device type; The user behavior records of each reservation user are used as tags, and the environmental state snapshot, user and scene identifiers of each reservation user are used as feature parameters. The feature parameters are associated and labeled using the tags to obtain multiple model training samples. Each model training sample is used to represent a reservation user adjustment event. A neural network is used to create a model for predicting device usage preference parameters; Based on the training samples of each model, machine learning is performed on the device using the preference parameter prediction model, and the model loss value is output. If the model loss value reaches its minimum, a pre-trained device preference parameter prediction model is generated; or if the model loss value does not reach its minimum, the step of performing machine learning on the device preference parameter prediction model based on each model training sample continues.

6. The method according to claim 5, characterized in that, Generate multi-dimensional historical data on each user's office usage by following these steps: During each user's use of the office, the current settings parameters of each device in the office are monitored in real time. In response to the parameter adjustment command of each reservation user, the final preference parameter value of each reservation user after manual adjustment of the device is obtained as a user behavior record; The office ID, indoor and outdoor temperature and humidity, light intensity, and timestamp of each appointment user's office are recorded synchronously as an environmental status snapshot; Synchronously record each user's user ID, the office ID where the device is located, and the device type as user and scenario identifiers; The system stores the user behavior records, environmental state snapshots, and the association between the user and the scene identifier for each user who made a reservation, thereby obtaining multi-dimensional historical data on each user's use of the office.

7. The method according to claim 5, characterized in that, The step of performing machine learning on the device using a preference parameter prediction model based on each model training sample includes: Feature engineering and feature encoding operations are performed on the environmental state snapshots, user and scene identifiers in each model training sample to obtain the feature data of each model training sample; The feature data of each model training sample is input into the device to predict the model using preference parameters, thereby obtaining the model prediction result for each model training sample; The model prediction result of each training sample and the label of each training sample are substituted into a preset loss function to obtain the model loss value; wherein, the function expression of the preset loss function is: ； in, It is the loss value. It is the number of training samples for the model. It is the first The labels of the training samples for each model. It is the first The model prediction results for each training sample.

8. An office equipment pre-start device based on user behavior prediction implemented using the method according to any one of claims 1-7, characterized in that, The device includes: The request receiving module is used to receive reservation requests sent by the client. The reservation request carries the usage time and office ID of the target office and the user ID of the target user. The data query module is used to query the historical behavior data of the target user and the current environmental data of the target office based on the user ID and office ID when the time between the current time and the time of use is a preset time. The preference parameter output module is used to input the historical behavior data and current environment data into a pre-trained device usage preference parameter prediction model, and output the usage preference parameters of each device in the target office corresponding to the target user. The optimal pre-startup time calculation module is used to calculate the optimal pre-startup time of each device based on the usage preference parameters and device performance parameters of each device. The instruction generation and distribution module is used to encapsulate the usage preference parameters of each device into a startup instruction based on the optimal pre-startup duration, and distribute it to the device controller used to control each device.

Images