Air conditioning system and control method thereof

Abstract

The invention provides an air conditioning system and a control method thereof, the air conditioning system comprises air conditioning equipment and a controller connected with the air conditioning equipment, and the controller is configured to determine a to-be-analyzed text based on input data of a user; acquiring environment data, wherein the environment data comprises a measured value of each environment parameter; determining a target cue word based on the to-be-analyzed text, the environment data and a cue word template, the cue word template comprising a target environment parameter and a weight corresponding to the target environment parameter, and the weight being used for reflecting an influence degree of the target environment parameter on a user intention; the target prompt word is used for guiding the large language model to analyze the user intention based on the to-be-analyzed text and the target environment parameters and generating a control instruction for controlling the air conditioner equipment based on the user intention; inputting the target cue word into the large language model to obtain a control instruction output by the large language model; and the air conditioning equipment is controlled based on the control instruction. The intelligence and accuracy of air conditioner control can be improved.

Description

Technical Field

[0001] This application relates to the field of air conditioning technology, and in particular to an air conditioning system and its control method. Background Technology

[0002] With the rapid development of artificial intelligence and the Internet of Things (IoT) technologies, smart home devices are becoming increasingly popular and are gradually becoming an important part of modern life. Among them, air conditioning systems, as an important component of smart home devices, are gradually becoming more widespread and are constantly being upgraded.

[0003] Currently, voice control based on fixed commands is mainly used to intelligently control air conditioning systems. However, this requires users to explicitly select control parameters such as mode and temperature, and it is difficult to recognize vague voice expressions or other inputs from users, thus limiting the effectiveness of intelligent control. Summary of the Invention

[0004] This application provides an air conditioning system and its control method to improve the intelligence and accuracy of air conditioning control.

[0005] In a first aspect, some embodiments of this application provide an air conditioning system, including: an air conditioning unit and a controller connected to the air conditioning unit;

[0006] The controller is configured to:

[0007] The text to be analyzed is determined based on user input data;

[0008] Acquire environmental data, which includes measured values ​​of various environmental parameters, to reflect the environmental conditions of the scene in which the air conditioning equipment is located;

[0009] The target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The prompt word template includes target environmental parameters and weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt word is used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent.

[0010] The target prompt word is input into the large language model to obtain the control command output by the large language model;

[0011] The air conditioning equipment is controlled based on the control commands.

[0012] Since environmental data includes measured values ​​of various environmental parameters, reflecting the actual environmental conditions of the scene where the air conditioning equipment is located, and the weights corresponding to the target environmental parameters in the prompt word template reflect the degree of influence of the target environmental parameters on the user's intention, the target prompt word is determined based on the environmental parameters, the text to be analyzed, and the prompt word template. This allows the large language model to combine the user's input and the degree of influence of the actual environmental conditions of the scene where the air conditioning equipment is located on the user's intention during the reasoning process based on the target prompt word. This enables a more accurate analysis of the user's actual intention to control the air conditioning equipment, improving the accuracy of the obtained control commands. Moreover, compared to directly recognizing the user's input data to determine the control commands, it can combine the influence of the environment on the user's intention, accurately identify fuzzy input data, and improve the intelligence effect and control accuracy.

[0013] In one possible implementation of the first aspect, the controller is further configured to: determine the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template before performing the operation.

[0014] Based on historical user data, the target environment parameters and the corresponding weights in the prompt word template are adjusted to obtain the adjusted prompt word template. The historical user data includes historical target prompt words corresponding to historical texts to be analyzed, historical control commands, and user feedback information on the historical control commands.

[0015] In one possible implementation of the first aspect, when the controller performs the task of determining the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured to:

[0016] If the text to be analyzed is not a mappable text, the target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The mappable text is text containing predetermined air conditioning equipment control operations.

[0017] In one possible implementation of the first aspect, after the controller performs the determination of the text to be analyzed based on the user's input data, it is further configured to:

[0018] If the text to be analyzed belongs to the mappable text, the text to be analyzed is mapped based on the set control instruction set to obtain the control instruction.

[0019] In one possible implementation of the first aspect, the large language model further outputs control suggestions corresponding to the control instructions, and the controller is configured to: control the air conditioning equipment based on the control instructions when performing such control.

[0020] The control suggestions are displayed on the set interactive interface, and upon receiving a confirmation instruction for the control suggestions, the air conditioning equipment is controlled based on the control instructions corresponding to the control suggestions.

[0021] In one possible implementation of the first aspect, when the controller executes a control instruction to input the target prompt word into the large language model and obtain the output of the large language model, it is configured to:

[0022] If the user intent is determined to be a control intent using the large language model, control instructions are generated based on the control intent.

[0023] Using the large language model, if the user intent is determined to be a non-control intent, control suggestions and control instructions corresponding to the control suggestions are generated based on the target environment parameters in the target prompt words and the weights corresponding to the target environment parameters.

[0024] In one possible implementation of the first aspect, when the controller, upon determining through the large language model that the user intent is not a control intent, generates control suggestions and control instructions corresponding to the control suggestions based on the target environment parameters in the target prompt words and the weights corresponding to the target environment parameters, is configured as follows:

[0025] Using the large language model, if the user intent is determined to be a non-control intent, control suggestions and corresponding control instructions are generated based on the target environment parameters in the target prompt words, the weights corresponding to the target environment parameters, and historical user data.

[0026] In one possible implementation of the first aspect, when the controller performs the task of determining the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured to:

[0027] The target prompt word is determined based on the text to be analyzed, the environmental data, the actual emotional information, and the prompt word template. The actual emotional information is used to reflect the user's current emotional state, and the target prompt word also includes the actual emotional information and the weight corresponding to the actual emotional information.

[0028] In one possible implementation of the first aspect, when the controller performs the task of determining the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured to:

[0029] The target prompt word is determined based on the text to be analyzed, the environmental data, the emotional needs information, and the prompt word template. The emotional needs information is used to reflect the user's emotional needs in the current scenario. The target prompt word also includes the emotional needs information and the weights corresponding to the emotional needs information.

[0030] Secondly, some embodiments of this application also provide a method for controlling an air conditioning system, the method comprising:

[0031] The text to be analyzed is determined based on user input data;

[0032] Acquire environmental data, which includes measured values ​​of various environmental parameters, to reflect the environmental conditions of the scene in which the air conditioning equipment of the air conditioning system is located;

[0033] The target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The prompt word template includes target environmental parameters and weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt word is used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent.

[0034] The target prompt word is input into the large language model to obtain the control command output by the large language model;

[0035] The air conditioning equipment is controlled based on the control commands.

[0036] Thirdly, embodiments of this application provide a control device, including a module for executing the control method of the air conditioning system in the second aspect.

[0037] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the control method for the air conditioning system described in the second aspect above.

[0038] Fifthly, embodiments of this application provide a computer program product that, when run on an air conditioning system, causes the air conditioning system to execute the control method for the air conditioning system described in the second aspect above.

[0039] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram illustrating the operational scenarios between an air conditioning system and a control device provided in some embodiments of this application;

[0042] Figure 2 This is a schematic diagram of the hardware configuration of an air conditioning system provided in some embodiments of this application;

[0043] Figure 3 This application provides a schematic diagram of the structure of an air conditioning system according to some embodiments;

[0044] Figure 4 A timing diagram illustrating a control method for an air conditioning system provided in some embodiments of this application;

[0045] Figure 5 A flowchart illustrating a control method for an air conditioning system provided in some embodiments of this application;

[0046] Figure 6 A flowchart illustrating a control method for an air conditioning system provided in some embodiments of this application;

[0047] Figure 7 A flowchart illustrating a control method for an air conditioning system provided in some embodiments of this application;

[0048] Figure 8 A timing interaction diagram of a control method for an air conditioning system provided in some embodiments of this application;

[0049] Figure 9 A flowchart illustrating a control method for an air conditioning system provided in some embodiments of this application;

[0050] Figure 10 A flowchart illustrating a control method for an air conditioning system provided in some embodiments of this application;

[0051] Figure 11 This is a schematic diagram of the structure of a control device for an air conditioning system provided in some embodiments of this application. Detailed Implementation

[0052] The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with this application. They are merely examples of systems and methods consistent with some aspects of this application as detailed in the claims.

[0053] It should be noted that the brief descriptions of terms in this application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.

[0054] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar or related objects or entities, and do not necessarily imply a specific order or sequence, unless otherwise specified. It should be understood that such terms can be used interchangeably where appropriate.

[0055] The terms “comprising” and “having”, and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device that includes a range of components is not necessarily limited to all of the components that are clearly listed, but may include other components that are not clearly listed or that are inherent to such product or device.

[0056] The term "module" refers to any known or subsequently developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functions associated with that element.

[0057] Figure 1 This is a schematic diagram illustrating the operational scenarios between an air conditioning system and control equipment provided in some embodiments of this application. For example... Figure 1 As shown, users can operate the air conditioning system 200 via touch operation, mobile terminal 300, and control device 100. For example, control device 100 can be a remote control, stylus, or handle.

[0058] In some embodiments, the control device 100 can be a remote control or a smart home controller, etc. For example, the control device is a remote control, and the communication methods between the remote control and the air conditioning system include, but are not limited to, infrared protocol communication, Bluetooth protocol communication, or other short-range communication methods, to control the air conditioning system 200 wirelessly or via wired means. Users can input user commands through buttons on the remote control, voice input, or control panel input to control the air conditioning system 200.

[0059] In some embodiments, a mobile terminal 300 (such as a tablet computer, computer, or mobile phone) can also be used to control the air conditioning system 200. For example, an application running on the mobile terminal 300 can be used to control the air conditioning system 200.

[0060] In some embodiments, the air conditioning system may receive instructions not through the aforementioned mobile terminal 300 or control device 100, but through buttons or other means provided on the air conditioning system.

[0061] Figure 2 Provided for some embodiments of this application Figure 1 Hardware configuration block diagram of the central control device. (Example) Figure 2 As shown, the control device 100 may include: a controller 110, a communication interface 130, a user input / output interface, a memory, and a power supply.

[0062] The control device 100 is configured to control the air conditioning system 200, and to receive user input operation commands and convert the operation commands into commands that the air conditioning system 200 can recognize and respond to, thus acting as an intermediary for interaction between the user and the air conditioning system 200.

[0063] In some embodiments, the control device 100 may be an intelligent device. For example, the control device 100 may be equipped with various applications for controlling the air conditioning system 200 according to user needs.

[0064] In some embodiments, such as Figure 1 As shown, a mobile terminal 300 or other smart electronic device can perform similar functions to control device 100 after an application for controlling the air conditioning system 200 is installed.

[0065] The controller 110 includes a processor 112, RAM 113, ROM 114, a communication interface 130, and a communication bus. The controller 110 is used to control the operation of the control device 100, as well as the communication and cooperation between internal components and the external and internal data processing functions.

[0066] Under the control of the controller 110, the communication interface 130 enables communication of control signals and data signals with the air conditioning system 200. The communication interface 130 may include at least one of other near-field communication modules such as WiFi chip 131, Bluetooth module 132, and NFC module 133.

[0067] User input / output interface 140, wherein the input interface includes at least one of other input interfaces such as microphone 141, touchpad 142, sensor 143, and button 144.

[0068] In some embodiments, the control device 100 includes at least one of a communication interface 130 and an input / output interface 140. The control device 100 is configured with the communication interface 130, such as a WiFi, Bluetooth, or NFC module, which can encode user input commands via the WiFi, Bluetooth, or NFC protocol and send them to the air conditioning system 200.

[0069] The memory 190 is used to store various operating programs, data, and applications for driving and controlling the control device 100 under the control of the controller. The memory 190 can also store various control signal instructions input by the user.

[0070] The power supply 180 is used to provide operating power support for the various components of the control device 100 under the control of the controller.

[0071] Figure 3 The diagram shows a schematic representation of an air conditioning system provided in some embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0072] See Figure 3 The air conditioning system includes an air conditioning unit 310 and a controller 320 connected to the air conditioning unit 310. It should be understood that the connection between the air conditioning unit 310 and the controller 320 can be a wired communication connection or a wireless communication connection.

[0073] The controller 320 is configured as follows:

[0074] The text to be analyzed is determined based on user input data;

[0075] Acquire environmental data, including measured values ​​of various environmental parameters, to reflect the environmental conditions of the scene where the air conditioning equipment is located.

[0076] Based on the above-mentioned text to be analyzed, the above-mentioned environmental data, and the prompt word template, the target prompt word is determined. The prompt word template includes the target environmental parameters and the weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt word is used to guide the large language model to analyze the user's intent based on the above-mentioned text to be analyzed and the above-mentioned target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent.

[0077] Input the target prompt words into the large language model to obtain the control commands output by the large language model;

[0078] The air conditioning equipment is controlled based on the aforementioned control commands.

[0079] Figure 4 The diagram illustrates the timing interaction of a control method for an air conditioning system according to some embodiments of this application. See also... Figure 4 The control method includes:

[0080] S401. Determine the text to be analyzed based on the user's input data.

[0081] The input data can be presented in various formats, including but not limited to text, voice, text and images, which can be set according to the actual application requirements.

[0082] It should be understood that when the controller acquires user input data, it can use data sent by the user to the controller through the terminal device, data entered by the user on the air conditioning system's interactive interface, or data collected by indoor voice acquisition devices (such as microphones or smart speakers). The method by which the controller acquires user input data can be set according to the actual application scenario, and no specific restrictions are imposed here.

[0083] It should be understood that when determining the text to be analyzed based on user input data, the text corresponding to the input data can be determined in a corresponding manner according to the current form of the input data, so as to obtain the required text to be analyzed.

[0084] For example, when the input data is voice data, the voice data can be converted into text through voice recognition and other methods to obtain the text to be analyzed.

[0085] It should be noted that in some embodiments, when the input data includes voice data, while converting the voice data into text, the user's emotional state can also be analyzed based on the voice data (such as analyzing the user's urgency based on the speech rate of the voice data), and the text to be analyzed can be determined based on the converted text and the emotional state.

[0086] In this embodiment of the application, the user's input data is converted into a unified text format to obtain the text to be analyzed, so that the user's intent can be better analyzed based on the semantic information expressed in the text to be analyzed.

[0087] S402. Obtain environmental data, including the measured values ​​of various environmental parameters, to reflect the environmental conditions of the scene where the air conditioning equipment is located.

[0088] It should be understood that environmental parameters include, but are not limited to, temperature, humidity, carbon dioxide concentration, air quality, ventilation volume, and radiant temperature. When acquiring environmental data, the controller can obtain the measured values ​​of one or more of the above environmental parameters according to user settings or actual needs to obtain the required environmental data.

[0089] As an example, environmental data can include measurements of temperature and humidity.

[0090] It should be understood that the measured values ​​of environmental parameters can be obtained through the sensors corresponding to the environmental parameter, or through third parties such as the bound meteorological service or environmental monitoring platform; different acquisition methods can be used to obtain the corresponding measured values ​​for different environmental parameters, and the specific method can be set according to the type of environmental parameter and actual application requirements.

[0091] In this embodiment of the application, since environmental parameters can reflect the actual environmental conditions of the scene where the air conditioning equipment is located, and the user's intention to control the air conditioning equipment has a significant impact, obtaining the environmental parameters and then combining them with the analysis of the user's intention can improve the accuracy of the intention analysis.

[0092] S403. Based on the above-mentioned text to be analyzed, the above-mentioned environmental data, and the prompt word template, determine the target prompt word. The prompt word template includes the target environmental parameters and the weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt word is used to guide the large language model to analyze the user's intent based on the above-mentioned text to be analyzed and the above-mentioned target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent.

[0093] It should be understood that the target environmental parameters can be environmental parameters determined based on user input or settings, or environmental parameters determined by the controller based on information such as actual application requirements; no specific restrictions are imposed here.

[0094] A cue word template is a predefined text framework used to guide the generative model to output text content that meets specific requirements according to predetermined logic and format. A cue word template typically includes a fixed part and a variable part; the fixed part is the predefined text content, which can be used to determine the basic semantic framework and grammatical rules of the cue words, while the variable part is reserved blank content that can be used to insert specific data or various values.

[0095] In this embodiment, the prompt word template is used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environment parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent. Correspondingly, the prompt word template may include, but is not limited to: analysis prompt information (used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environment parameters), control prompt information (used to guide the large language model to generate control instructions for the air conditioning equipment based on the user's intent), and output format information (used to indicate the content and format of the control instructions to be output, etc.).

[0096] As an example, a prompt template can be represented in the following form:

[0097] You are a smart air conditioner semantic analysis assistant. Please analyze the user's intent based on the following information:

[0098] Environmental conditions: room temperature (blank content 1, weight 1), humidity (blank content 2, weight 2), current weather (blank content 3, weight 3);

[0099] User input: blank content 4;

[0100] Please generate JSON format control instructions for the air conditioning equipment based on the user's intent. The fields of the output control instructions should include:

[0101] mode: Air conditioning mode, with possible values ​​including ["cooling", "heating", "dehumidify", "fan", "auto"].

[0102] Temperature: Recommended temperature (unit: degrees Celsius, integer, e.g., 24).

[0103] fan_speed: Fan speed setting, selectable values ​​are ["low", "medium", "high", "auto"].

[0104] dehumidify: Whether to enable the dehumidification function. The possible values ​​are true or false.

[0105] Suppose the text to be analyzed is: "It's too stuffy in the room, it feels like I can't breathe from the heat." Environmental data includes: current room temperature 35 degrees Celsius, current humidity 40%. When determining the target prompt, substitute the text to be analyzed into the corresponding blank content 4 in the prompt template, and substitute the environmental data into the blank content corresponding to the environmental conditions in the prompt template. The following target prompts can be obtained:

[0106] You are a smart air conditioner semantic analysis assistant. Please analyze the user's intent based on the following information:

[0107] Environmental conditions: room temperature (35 degrees Celsius, weight 1), humidity (40%, weight 2), current weather (none, weight 3);

[0108] User input: It's so stuffy in here, it feels like I can't breathe from the heat;

[0109] Please generate JSON format control instructions for the air conditioning equipment based on user intent. The output control instructions should include the following fields:

[0110] mode: Air conditioning mode, with possible values ​​including ["cooling", "heating", "dehumidify", "fan", "auto"].

[0111] Temperature: Recommended temperature (unit: degrees Celsius, integer, e.g., 24).

[0112] fan_speed: Fan speed setting, selectable values ​​are ["low", "medium", "high", "auto"].

[0113] dehumidify: Whether to enable the dehumidification function. The possible values ​​are true or false.

[0114] It is important to note that the weight values ​​corresponding to each target environment parameter in the prompt word template can be fixed or variable. That is, the weight values ​​corresponding to the target environment parameters can be preset values ​​or dynamically determined values ​​when the target prompt words are determined (such as dynamically calculated values ​​or values ​​obtained by dynamically adjusting preset values ​​according to needs).

[0115] For example, assuming that in the above prompt word template, the values ​​of weight 1, weight 2, and weight 3 are 0.4, 0.3, and 0.3 respectively, when determining the target prompt word based on the text to be analyzed, environmental data, and the prompt word template, if the corresponding user input data includes voice data, the values ​​of weight 1, weight 2, and weight 3 can be adjusted according to the voice information such as tone, speed, and intonation of the voice data to obtain adjusted values, such as the adjusted values ​​of weight 1, weight 2, and weight 3 being 0.5, 0.2, and 0.3 respectively.

[0116] In some embodiments, since different users often have different physical sensations and preferences, their control preferences for air conditioning will vary under the same or different environmental conditions. Therefore, in order to improve the accuracy of the generated control commands, each user can pre-set the weight values ​​corresponding to each target environmental parameter. When determining the target prompt word, the user identifier of the user corresponding to the text to be analyzed (i.e., the user corresponding to the input data, assuming to be called the target user) can be used to determine the pre-set weight values ​​of the target user, and then substitute them into the prompt word template to determine the weight values ​​corresponding to each target environmental parameter in the target prompt word.

[0117] In this embodiment, since environmental conditions usually directly affect user intent, and the degree of influence of different environmental parameters on user intent usually varies, target environmental parameters are introduced into the prompt word template. At the same time, the weights corresponding to each target environmental parameter reflect the degree of influence of different target environmental parameters on user intent. Then, the target prompt words are determined by combining actual environmental data and the user's text to be analyzed. This allows the large language model to comprehensively analyze the user's actual intent to control the air conditioning equipment by combining the user's actual input, the actual environmental conditions of the scene where the air conditioning equipment is located, and the degree of influence of different target environmental parameters on user intent, thereby improving the accuracy and reliability of the generated control commands.

[0118] S404. Input the target prompt words into the large language model to obtain the control instructions output by the large language model.

[0119] Large language models are natural language processing models based on deep learning. They learn language rules by pre-training on massive amounts of text data and have the ability to understand, generate, and reason about text.

[0120] In this embodiment, the large language model used can be a general model pre-trained on massive text data, or it can be a model fine-tuned based on the target prompt word sample data and the corresponding control instruction label data, so that it can perform better in the control instruction generation task and improve the accuracy of the final generated control instructions.

[0121] S405. Control the air conditioning equipment based on the above control instructions.

[0122] by Figure 4 For example, when the controller controls the air conditioning equipment based on the generated control command, it can send the control command directly to the air conditioning equipment, and the air conditioning equipment will operate according to the control command, thereby realizing the control of the air conditioning equipment in the air conditioning system.

[0123] It should be understood that in scenarios such as smart buildings, the air conditioning system can be a multi-split air conditioning system (i.e., an air conditioning system where the air conditioning equipment is connected to multiple indoor units through one outdoor unit) or a centralized control system (i.e., an air conditioning system that includes multiple air conditioning units). For example, such as Figure 3 As shown, the air conditioning system includes two air conditioning units, and the controller is connected to each air conditioning unit respectively.

[0124] When the controller controls the air conditioning equipment based on the control command, it can first determine the control target (i.e., the air conditioning equipment to be controlled or the indoor unit of the air conditioning equipment to be controlled), and based on the control target, send the control command to the corresponding air conditioning equipment or indoor unit, thereby accurately controlling the corresponding air conditioning equipment or indoor unit to operate according to the control command.

[0125] It should be understood that when determining the control target, it can be based on the location information of the corresponding user (such as the location of the user's terminal device) or it can be based on the text to be analyzed. The specific implementation method of determining the control target can be set according to the actual application requirements.

[0126] In some embodiments, the controller can perform corresponding protocol conversion processing (such as infrared code conversion processing) on ​​the control command according to the communication protocol between the controller and the air conditioning equipment to be controlled, and send the converted control command to the air conditioning equipment, thereby ensuring that the control command is accurately sent to the air conditioning equipment and correctly parsed by the air conditioning equipment.

[0127] In this embodiment, since the environmental data includes the measured values ​​of various environmental parameters, it can reflect the actual environmental conditions of the scene where the air conditioning equipment is located. The weights corresponding to the target environmental parameters in the prompt word template can reflect the degree of influence of the target environmental parameters on the user's intention. Therefore, the target prompt word is determined based on the environmental parameters, the text to be analyzed, and the prompt word template. This allows the large language model to combine the user's input and the degree of influence of the actual environmental conditions of the scene where the air conditioning equipment is located on the user's intention during the reasoning process based on the target prompt word. This enables a more accurate analysis of the user's actual intention to control the air conditioning equipment, improving the accuracy of the obtained control commands. Compared with directly recognizing the user's input data to determine the control commands, this approach can combine the influence of the environment on the user's intention, accurately identify fuzzy input data, and improve the intelligence effect and control accuracy.

[0128] Figure 5 The illustration shows a control method for an air conditioning system according to some embodiments. See also Figure 5 Before the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is also configured to perform the following steps:

[0129] Based on historical user data, the target environment parameters and their corresponding weights in the above-mentioned prompt word template are adjusted to obtain the adjusted prompt word template. The historical user data includes historical target prompt words corresponding to historical texts to be analyzed, historical control commands, and user feedback information on the historical control commands.

[0130] Optionally, user feedback on historical control commands can be used to reflect the user's actual adjustment status of the historical control command, the user's satisfaction with the historical control command and / or adjustment suggestions, and the degree of matching between the historical control command and the corresponding historical text to be analyzed (or historical input data), etc. The specific settings can be configured according to actual application requirements.

[0131] It should be understood that in other embodiments, historical user data may also include, but is not limited to, input data corresponding to historical text to be analyzed, user intent, the scene in which the air conditioning equipment is located, and the actual execution status of control commands.

[0132] In some embodiments, historical user data (let's call it first historical user data) may include historical target prompts, historical control instructions, and feedback information corresponding to historical text to be analyzed within a first time period in the past (e.g., the past week). By adjusting the prompt template using more recent historical user data, the target environment parameters and their corresponding weights contained in the prompt template can more accurately reflect the degree of influence of the target environment parameters on user intent, thereby improving the accuracy of subsequent user intent analysis.

[0133] In some embodiments, the prompt word template on the controller can be adjusted every preset time interval (e.g., 10 days). In this case, the historical user data can include the time period from the last adjustment to the current time (i.e., the preset time period is the time period starting from the last adjustment and ending at the current time, and its duration is equal to the preset time period), the historical target prompt words corresponding to the historical text to be analyzed on the controller, the historical control instructions, and the feedback information.

[0134] It should be understood that the controller's adjustment of the target environment parameters in the prompt word template may include adding and / or deleting. That is, the controller may add target environment parameters contained in the prompt word template, or it may delete existing target environment parameters in the prompt word template.

[0135] As an example, when the controller adjusts the target environment parameters and their corresponding weights in the prompt word template based on historical user data, it can use machine learning algorithms (such as clustering learning or deep learning networks) to analyze the changing trends of user intent based on historical user data, thereby analyzing the correlation between the changes in various environment parameters and the changes in user intent. Then, based on the learned correlation, the controller adjusts the target environment parameters and their corresponding weights in the prompt word template to obtain the adjusted prompt word template.

[0136] As another example, the controller can also perform statistical analysis on historical user data to determine the causal relationships and statistical patterns among various data points in the historical user data. Based on the information obtained from the analysis, the target environment parameters and their corresponding weights in the prompt word template can be adjusted to obtain an adjusted prompt word template. Furthermore, adjusting the prompt word template based on statistical analysis can reduce the demand on the controller's computing resources and improve the controller's reliability and response speed.

[0137] In this embodiment, historical user data that reflects user satisfaction with historical control commands is acquired. Based on this historical user data, the target environment parameters included in the prompt word template and their corresponding weights are adjusted to better suit the user's personalized habits and needs, thereby improving the accuracy of analyzing user intent based on target prompt words.

[0138] Figure 6 A flowchart illustrating a control method for an air conditioning system according to some embodiments is shown. See also Figure 6 When the controller determines the target prompt word based on the above-mentioned text to be analyzed, the above-mentioned environmental data, and the prompt word template, it is configured to perform the following steps:

[0139] If the text to be analyzed is not a mappable text, the target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The mappable text is text containing predetermined air conditioning equipment control operations.

[0140] It should be understood that the predetermined air conditioning equipment control operations include, but are not limited to, temperature setting operations, air conditioning switch operations, mode selection operations, and fan speed setting operations, as well as other air conditioning equipment control-related operations.

[0141] In some embodiments, if the control operation in the text to be analyzed belongs to the target type and the control parameters of the control operation can be determined based on the text to be analyzed, the text to be analyzed can be determined as a mappable text. This ensures that when a user needs to perform certain specific control operations, the actual control parameters can be clearly determined, thereby improving the accuracy of air conditioning control and user experience.

[0142] For example, suppose the target type includes parameter setting types (such as temperature setting or fan speed setting), and suppose the text to be analyzed is: "Turn down the air conditioner temperature". That is, the control operation in the text to be analyzed is temperature setting, which belongs to the parameter setting type. However, the text to be analyzed does not contain a specific temperature value (i.e., a specific control parameter). Therefore, the text to be analyzed can be determined to be unmappable text.

[0143] In this embodiment of the application, when the text to be analyzed does not contain the predetermined control operation of the control device, the target prompt word is determined based on the text to be analyzed, environmental data and prompt word template, so as to generate appropriate control instructions by analyzing the target prompt word through a large language model, thereby improving the recognition accuracy of fuzzy expressions input by users and the control accuracy of air conditioning equipment.

[0144] In some embodiments, after the controller performs the above-described determination of the text to be analyzed based on user input data, it is further configured to perform the following steps:

[0145] If the text to be analyzed belongs to the mappable text, the text to be analyzed is mapped based on the set control instruction set to obtain the control instructions.

[0146] It should be understood that a control instruction set is typically used to describe how text is mapped to specific control instructions. In the embodiments of this application, the control instruction set may include, but is not limited to, information such as the description, content, and mapping rules of the corresponding control instructions for each control operation, where the mapping rules refer to the specific rules for mapping text to control instructions.

[0147] Optionally, when mapping the text to be analyzed based on the set control instruction, keywords can first be extracted from the text through one or more methods such as entity recognition or semantic recognition. Then, the extracted keywords are matched based on the mapping rules in the control instruction set to determine the corresponding control instruction. It should be understood that the extracted keywords include, but are not limited to, entity keywords, operation keywords (i.e., words related to control operations), and parameter keywords (such as words related to temperature parameters such as 21 degrees).

[0148] See Figure 7 After obtaining the control instructions corresponding to the text to be analyzed through mapping processing, the controller can control the air conditioning equipment according to the control instructions without performing large language model analysis and other steps, which can improve the control efficiency of the air conditioning system and reduce unnecessary resource consumption.

[0149] In this embodiment of the application, if the text to be analyzed is a mappable text, that is, the text to be analyzed contains air conditioning equipment control operations, then the text to be analyzed is directly mapped to the corresponding control instructions based on the set control instruction set, without the need to generate target prompt words and perform model processing, thereby improving the control efficiency of the air conditioning system and reducing the resource occupation of the controller.

[0150] Figure 8 The diagram illustrates the timing interaction of a control method for an air conditioning system according to some embodiments of this application. See also... Figure 8 The aforementioned large language model also outputs control suggestions corresponding to the aforementioned control commands. When the controller executes control of the aforementioned air conditioning equipment based on the aforementioned control commands, it is configured to perform the following steps:

[0151] The above control suggestions are displayed on the set interactive interface, and upon receiving a confirmation instruction for the above control suggestions, the air conditioning equipment is controlled based on the control instructions corresponding to the above control suggestions.

[0152] It should be understood that control recommendations may include, but are not limited to, detailed descriptions of control instructions, control effects, and precautions.

[0153] It should be understood that the set interactive interface can be the user interface of the air conditioning system or the user interface of the user's corresponding terminal device. This application embodiment does not impose specific limitations on this.

[0154] The above confirmation instruction is used to instruct the user to confirm the use of the control instruction corresponding to the control suggestion to control the air conditioning equipment.

[0155] Figure 8 The following example illustrates how a controller can control an air conditioning unit based on a control suggestion sent by a user through a terminal device. When the user instruction is a control instruction (i.e., the user instruction instructs the air conditioning unit to be controlled based on the control suggestion), the controller can control the air conditioning unit based on the control instruction corresponding to the control suggestion.

[0156] It should be understood that when the user instruction is not a control instruction, the controller can process it accordingly based on the actual content of the user instruction.

[0157] For example, when a user instruction indicates that the value of a certain control parameter (such as temperature) in the control suggestion should be adjusted to the target value, the controller can modify the control instruction corresponding to the control suggestion according to the user instruction, obtain the modified control instruction, and control the air conditioning equipment based on the control instruction corresponding to the modified control suggestion.

[0158] For example, if the user command indicates that no action should be taken, the controller does not need to control the air conditioning equipment.

[0159] In this embodiment, the large language model generates control instructions and corresponding control suggestions at the same time. The controller can display the control suggestions in a set interactive interface. The air conditioning equipment is controlled only after the user confirms that the control instructions corresponding to the control suggestions are used to control the air conditioning equipment. This ensures that the control of the air conditioning equipment meets the user's expectations and improves the accuracy and reliability of the air conditioning system control.

[0160] Figure 9 A schematic flowchart illustrating a control method for an air conditioning system according to some embodiments of this application is shown. See also Figure 9 When the controller executes the control command output by inputting the target prompt word into the large language model and obtaining the control command output by the large language model, it is configured to perform the following steps:

[0161] Based on the above large language model, if the user intent is determined to be a control intent, control instructions are generated based on the control intent.

[0162] Using the aforementioned large language model, when the user intent is determined to be a non-control intent, control suggestions and corresponding control instructions are generated based on the target environment parameters and their corresponding weights in the target prompt words.

[0163] The aforementioned control intent refers to the intention to control the air conditioning equipment. It can be understood that when a user's intent is to arbitrarily control the air conditioning equipment in the air conditioning system, the user's intent can be identified as the control intent.

[0164] For example, suppose the text to be analyzed, determined based on user input data, is: "Stocks have surged." Environmental data includes: temperature 19℃, humidity 80%. Since this text is unrelated to weather or user experience, when the large language model analyzes user intent based on the corresponding target cue words, it may directly obtain the user intent: "Stocks have surged," or it may be unable to clearly identify the user intent. In this case, the user intent can be considered a non-control intent. The large language model can then generate control suggestions and corresponding control instructions based on the target environmental parameters and their corresponding weights.

[0165] In some embodiments, when the user intent is a non-control intent containing target keywords, control suggestions and corresponding control instructions can be generated based on the user intent, target environmental parameters, and the weights corresponding to the target environmental parameters. Target keywords include, but are not limited to, climate-related words (such as weather, temperature, or humidity), user feeling descriptions (such as dampness, stuffiness, or shortness of breath), and user action-related words (such as exercise or sleeping).

[0166] In some embodiments, when the user's intent is not a control intent, the relevant control parameters (such as temperature) in the control suggestions generated by the large language model can be empty, i.e., the current control parameters are not set. When the controller displays the control suggestions on the interactive interface, it can prompt the user to input control parameters, and then adjust the control instructions generated by the large language model according to the user's instructions for the control suggestions (which may include the control parameters input by the user) and the control suggestions to obtain the final control instructions.

[0167] It should be noted that the target environment parameters in the target prompt words usually have corresponding measurement values. When large language models perform various analyses based on the target environment parameters in the target prompt words, they usually combine the measurement values ​​of the target environment parameters for analysis.

[0168] In this embodiment, after the large language model obtains the user intent based on the text to be analyzed, target environment parameters, and the target environment parameters, if it determines that the user intent is a control intent, it can directly generate control instructions for the air conditioning equipment based on the control intent. If the user intent is a non-control intent unrelated to the control of the air conditioning equipment, in order to ensure the control accuracy of the air conditioning equipment and reduce invalid processing, the large language model can analyze feasible control suggestions and corresponding control instructions based on the target environment parameters in the target prompt words and the weights corresponding to the target environment parameters. This allows the controller to confirm with the user whether to perform corresponding control on the air conditioning equipment based on the generated control suggestions, instead of directly canceling the processing. This improves the recognition effect of fuzzy user expressions while enhancing the intelligence and reliability of the air conditioning system.

[0169] Figure 10 The diagram illustrates the timing interaction of a control method for an air conditioning system according to some embodiments of this application. See also... Figure 10 When the controller, through the large language model, determines that the user intent is not a control intent, and generates control suggestions and corresponding control instructions based on the target environment parameters and their corresponding weights in the target prompt words, it is configured to execute the following steps:

[0170] Using the aforementioned large language model, if the user intent is determined to be a non-control intent, then based on the target environment parameters in the target prompt words, the weights corresponding to the target environment parameters, and historical user data, control suggestions and corresponding control instructions are generated.

[0171] In this embodiment, historical user data (let's call it second historical user data) may include historical target prompts, historical control commands, and feedback information corresponding to historical texts to be analyzed within a second past time period (e.g., the past month). By adjusting the prompt template using more recent historical user data, the target environment parameters and their corresponding weights contained in the prompt template can more accurately reflect the degree of influence of the target environment parameters on user intent, thereby improving the accuracy of subsequent user intent analysis.

[0172] It is important to note that the first and second time periods can be the same or different time periods. In some embodiments, the first time period is shorter than the second time period. By setting a shorter first time period, the prompt word template is updated using the most recent historical user data, ensuring that the prompt word template closely matches the user's personalized habits and needs. Furthermore, when necessary, control suggestions and commands can be generated based on the larger volume of second historical user data, improving the accuracy and reliability of the generated control suggestions and commands. Moreover, based on the larger volume of second historical information, the large language model can better analyze the temporal and periodic characteristics of the user's historical control intentions and the context of user interactions, establishing long-term user preference behavior patterns, thereby generating more accurate and appropriate control suggestions and commands.

[0173] It should be understood that the specific reasoning method by which the large language model generates control suggestions and corresponding control instructions based on the target environment parameters in the target prompt words, the weights of the target environment parameters, and historical user data can be set according to actual application requirements.

[0174] For example, a large language model can first determine the target environment parameter (including its measured value) from historical user data that meets the requirements (e.g., the first similarity is greater than the similarity threshold, such as 0.8) and obtain the reference target environment parameter; then, based on the control instructions corresponding to the reference target environment parameter, determine the initial control suggestion; then, based on the difference between the weights corresponding to the current target environment parameter and the weights corresponding to the reference target environment parameter, adjust the initial control suggestion to obtain the final control suggestion, and generate the control instructions corresponding to the control suggestion.

[0175] It should be understood that, in other embodiments, when the large language model determines that the user intent is a non-control intent, it can also generate control suggestions and the target environment of the control instructions corresponding to the control suggestions based on the user intent, target environment parameters, the weights corresponding to the target environment parameters, and historical user data. This allows the large language model to combine the analysis of the similarity between the data corresponding to historical non-control intents (let's call them reference intents, whose similarity to the current non-control intent is greater than or equal to a threshold) in the historical user data and the current data to determine appropriate control suggestions.

[0176] For example, a large language model can generate a current control suggestion based on the overall similarity (let's call it the second similarity) between the target environment parameters and their corresponding weights in the current target prompt and the target environment parameters and their corresponding weights in the target prompt corresponding to the target intent. This similarity is based on the historical control instructions corresponding to the target intent with the highest similarity.

[0177] In this embodiment, since historical user data can better reflect the actual control status of the air conditioning equipment by the user under various environmental conditions, and the user intent obtained by the large language model analysis is a non-control intent, that is, when the large language model cannot clearly identify the user's control intent for the air conditioning equipment, the feasible control suggestions are comprehensively analyzed by combining the target environment parameters in the target prompt words, the weights corresponding to the target environment parameters, and historical user data. Therefore, more accurate and reliable control suggestions and control instructions can be generated, which further improves the recognition effect of the user's fuzzy expression, while improving the intelligence effect and reliability of the air conditioning system.

[0178] In some embodiments, when the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured to perform the following steps:

[0179] The target prompt words are determined based on the text to be analyzed, the environmental data, the actual emotional information, and the prompt word template. The actual emotional information is used to reflect the current emotional state of the user. The target prompt words also include the actual emotional information and the weights corresponding to the actual emotional information.

[0180] It should be understood that a user's emotional state includes, but is not limited to, states such as irritability, fatigue, or sadness. A user may be in one or more states, and correspondingly, actual emotional information may include one or more states.

[0181] It should be understood that actual emotional information can be obtained from user input data and the text to be analyzed, or it can be obtained from user physiological state data (such as heart rate and blood pressure detected by the user's wearable device), or it can be obtained from user input data, the text to be analyzed, and physiological state data, etc. The specific settings can be determined according to the actual application requirements.

[0182] It should be understood that the weight corresponding to the actual emotional information in the target prompt can be a pre-set fixed weight, or a weight dynamically determined based on the state reflected by the emotional state information. Different states may correspond to different weights. Correspondingly, the value of the weight corresponding to the actual emotional information in the prompt template can be a pre-set fixed value, or it can be blank content.

[0183] For example, suppose the weight corresponding to irritability is 0.2, the weight corresponding to fatigue is 0.18, and the weight corresponding to sadness is 0.15; suppose the emotional state information reflects that the user's current emotional state is fatigue, then the weight corresponding to the emotional state information can be 0.18. The weight of 0.18 corresponding to the current emotional state: fatigue can be substituted into the position of the weight value corresponding to the emotional state information in the prompt word template.

[0184] In this embodiment, since a user's emotional state may affect their intention to control the air conditioning equipment, and the acquired emotional state information can reflect the user's current emotional state, and the weight corresponding to the emotional state information can reflect the degree of influence of the user's emotional state on the user's intention, when the controller determines the target prompt word, it determines the target prompt word containing the emotional state information and its corresponding weight. This allows the large language model to more comprehensively analyze the user's intention based on the target prompt word, from aspects such as user input data, environmental conditions, and user emotions and their degree of influence, thereby improving the accuracy of user intention recognition and improving the accuracy of the final generated control command and the intelligent effect of the air conditioning system.

[0185] In some embodiments, when the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured to perform the following steps:

[0186] Based on the above-mentioned text to be analyzed, the above-mentioned environmental data, the above-mentioned emotional needs information, and the above-mentioned prompt word template, the above-mentioned target prompt words are determined. The above-mentioned emotional needs information is used to reflect the user's emotional needs in the current scenario. The above-mentioned target prompt words also include the above-mentioned emotional needs information and the weights corresponding to the above-mentioned emotional needs information.

[0187] The aforementioned emotional needs may include, but are not limited to, emotional needs (which may refer to the expected emotion, such as happiness), state needs (which may refer to the expected state, such as relaxation), and control needs (which may refer to the expected control effect, such as the flexibility of controlling air conditioning equipment).

[0188] It should be noted that in some embodiments, emotional needs information can reflect the user's current or future (e.g., in the next hour) emotional needs in the current scenario.

[0189] In some embodiments, a user's emotional needs in the current scenario can be reflected through behavioral information, including but not limited to information on work, entertainment, rest, or exercise.

[0190] It should be understood that the controller can determine the emotional needs information based on the user's current scenario and actual emotional information, or based on the user's current scenario and input data, or based on the user's current scenario, input data, actual emotional information, and historical data corresponding to each data (such as historical scenarios, historical actual emotional information, and historical emotional needs information). The specific settings can be configured according to actual application requirements.

[0191] In some embodiments, the controller can analyze the user's emotional needs in the current scenario using a Naive Bayes algorithm to obtain emotional need information.

[0192] It should be understood that the weight corresponding to the emotional need information in the target prompt can be a pre-set fixed weight, or a weight dynamically determined based on the emotional need reflected by that emotional need information. Different emotional needs may correspond to different weights. Correspondingly, the value of the weight corresponding to the emotional need information in the prompt template can be a pre-set fixed value, or it can be blank content.

[0193] In this embodiment, since a user's emotional needs in the current scenario can reflect information such as the user's expectations, goals, and motivations to a certain extent, and the weights corresponding to the emotional needs information can reflect the degree of influence of the user's emotional needs in the current scenario on the user's intentions, when the controller determines the target prompt word, it can determine the target prompt word containing the emotional needs information and its corresponding weights. This allows the large language model to analyze the user's intentions more comprehensively from aspects such as user input data, environmental conditions, and the user's emotional needs and their degree of influence based on the target prompt word, thereby improving the accuracy of user intention recognition and improving the accuracy of the final generated control commands and the intelligent effect of the air conditioning system.

[0194] In some embodiments, the controller can determine the target prompt word based on the text to be analyzed, environmental data, actual emotional information, emotional needs information, and prompt word template, thereby further improving the comprehensiveness of user intent analysis by the large language model, and thus improving the reliability of the generated control commands and the intelligent effect of the air conditioning system.

[0195] In some embodiments, the environment in which the air conditioning system or air conditioning equipment is located may also include various other systems such as a fresh air system, a lighting system, or a curtain control system. Before determining the target prompt word, the controller of the air conditioning system can also obtain the operating information of the associated system (such as the fresh air system). This operating information can be used to reflect the operating status of the equipment corresponding to the associated system. At this time, the controller can determine the target prompt word by combining the operating information of the associated system. Correspondingly, the determined target prompt word may include the operating information of each associated system. The target prompt word may also include linkage control prompt information. This linkage control prompt information is used to guide the large language model to analyze whether to link control of the associated system by combining the user intent and the operating information of the associated system. When determining the associated system to be linked for control, the controller generates an associated control command for controlling the equipment corresponding to the associated system. It should be understood that after receiving the associated control command, the controller can send the associated control command to the corresponding associated system or the central controller, thereby realizing the linkage control of the equipment corresponding to the associated system.

[0196] In some embodiments, the controller can also combine data such as the operating information of the associated system, the text to be analyzed, and the target environment parameters to analyze user intent, thereby improving the comprehensiveness of user intent analysis.

[0197] The above text combined Figures 4 to 10 The control method of the air conditioning system according to the embodiments of this application is described in detail below. Figure 11 This application describes an embodiment of the apparatus. It should be understood that the control device of the air conditioning system in the embodiments of this application can execute the various air conditioning system control methods described in the foregoing embodiments of this application. That is, the specific working processes of the various products described below can be referred to the corresponding processes in the foregoing method embodiments.

[0198] Figure 11 A schematic diagram of the structure of a control device for an air conditioning system provided in some embodiments of this application is shown. See also Figure 11 The device 1100 includes: a text to be analyzed determination module, an environmental data acquisition module 1120, a target prompt word determination module 1130, a generation module 1140, and a control module 1150. Among them,

[0199] The text to be analyzed determination module 1110 is used to determine the text to be analyzed based on user input data.

[0200] The environmental data acquisition module 1120 is used to acquire environmental data, including the measured values ​​of various environmental parameters, which are used to reflect the environmental conditions of the scene in which the air conditioning equipment of the air conditioning system is located.

[0201] The target prompt word determination module 1130 is used to determine target prompt words based on the text to be analyzed, the environmental data, and the prompt word template. The prompt word template includes target environmental parameters and the weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt words are used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent.

[0202] The generation module 1140 is used to input the target prompt words into the large language model and obtain the control instructions output by the large language model.

[0203] The control module 1150 is used to control the air conditioning equipment based on the control commands mentioned above.

[0204] Each unit module of the control device 1100 of the air conditioning system can execute the corresponding steps in the above method embodiment. Therefore, each unit module will not be described in detail here. Please refer to the description of the corresponding steps above for details.

[0205] This application also provides a computer-readable storage medium storing a computer program, which, when executed by an air conditioning system, can implement the steps in the above-described method embodiments.

[0206] This application provides a computer program product that, when run on an air conditioning system, enables the controller to implement the steps of the control methods described in the above embodiments.

[0207] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0208] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0209] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0210] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0211] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0212] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An air conditioning system, characterized in that, include: Air conditioning equipment; The controller connected to the air conditioning unit is configured as follows: The text to be analyzed is determined based on user input data; Acquire environmental data, which includes measured values ​​of various environmental parameters, to reflect the environmental conditions of the scene in which the air conditioning equipment is located; The target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The prompt word template includes target environmental parameters and weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt word is used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent. The target prompt word is input into the large language model to obtain the control command output by the large language model; The air conditioning equipment is controlled based on the control commands.

2. The air conditioning system as described in claim 1, characterized in that, Before the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is also configured to: Based on historical user data, the target environment parameters and the corresponding weights in the prompt word template are adjusted to obtain the adjusted prompt word template. The historical user data includes historical target prompt words corresponding to historical texts to be analyzed, historical control commands, and user feedback information on the historical control commands.

3. The air conditioning system as described in claim 1, characterized in that, When the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured as follows: If the text to be analyzed is not a mappable text, the target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The mappable text is text containing predetermined air conditioning equipment control operations.

4. The air conditioning system as described in claim 3, characterized in that, After the controller executes the process of determining the text to be analyzed based on user input data, it is further configured to: If the text to be analyzed belongs to the mappable text, the text to be analyzed is mapped based on the set control instruction set to obtain the control instruction.

5. The air conditioning system as described in claim 1, characterized in that, The large language model also outputs control suggestions corresponding to the control commands. When the controller executes control of the air conditioning equipment based on the control commands, it is configured as follows: The control suggestions are displayed on the set interactive interface, and upon receiving a confirmation instruction for the control suggestions, the air conditioning equipment is controlled based on the control instructions corresponding to the control suggestions.

6. The air conditioning system as described in claim 5, characterized in that, When the controller executes the control command that inputs the target prompt word into the large language model and obtains the output of the large language model, it is configured as follows: If the user intent is determined to be a control intent using the large language model, control instructions are generated based on the control intent. Using the large language model, if the user intent is determined to be a non-control intent, control suggestions and control instructions corresponding to the control suggestions are generated based on the target environment parameters in the target prompt words and the weights corresponding to the target environment parameters.

7. The air conditioning system as described in claim 6, characterized in that, When the controller, upon determining through the large language model that the user intent is not a control intent, generates control suggestions and corresponding control commands based on the target environment parameters in the target prompt words and the weights corresponding to those parameters, it is configured as follows: Using the large language model, if the user intent is determined to be a non-control intent, control suggestions and corresponding control instructions are generated based on the target environment parameters in the target prompt words, the weights corresponding to the target environment parameters, and historical user data.

8. The air conditioning system according to any one of claims 1 to 7, characterized in that, When the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured as follows: The target prompt word is determined based on the text to be analyzed, the environmental data, the actual emotional information, and the prompt word template. The actual emotional information is used to reflect the user's current emotional state, and the target prompt word also includes the actual emotional information and the weight corresponding to the actual emotional information.

9. The air conditioning system according to any one of claims 1 to 7, characterized in that, When the controller determines the target prompt word based on the text to be analyzed, the environmental data, and the prompt word template, it is configured as follows: The target prompt word is determined based on the text to be analyzed, the environmental data, the emotional needs information, and the prompt word template. The emotional needs information is used to reflect the user's emotional needs in the current scenario. The target prompt word also includes the emotional needs information and the weights corresponding to the emotional needs information.

10. A control method for an air conditioning system, characterized in that, include: The text to be analyzed is determined based on user input data; Acquire environmental data, which includes measured values ​​of various environmental parameters, to reflect the environmental conditions of the scene in which the air conditioning equipment of the air conditioning system is located; The target prompt word is determined based on the text to be analyzed, the environmental data, and the prompt word template. The prompt word template includes target environmental parameters and weights corresponding to the target environmental parameters. The weights are used to reflect the degree of influence of the target environmental parameters on the user's intent. The target prompt word is used to guide the large language model to analyze the user's intent based on the text to be analyzed and the target environmental parameters, and to generate control instructions for controlling the air conditioning equipment based on the user's intent. The target prompt word is input into the large language model to obtain the control command output by the large language model; The air conditioning equipment is controlled based on the control commands.

Images