A Prompt-based air conditioning control method
Through the Prompt-based air conditioning control method, the large language model is used to generate control instructions that adapt to complex environments, solving the problem of inefficiency of traditional air conditioning systems in complex environments and rapidly changing environments, and achieving rapid adjustment and high user comfort.
Patent Information
- Application Number
- CN202410924282.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-11
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-11
AI Technical Summary
Traditional air conditioning system control methods are difficult to quickly adjust control strategies when facing complex environments and rapid environmental changes, and require retraining of the model, resulting in inefficiency.
The air conditioning control method based on Prompt is adopted, and the air outlet temperature and air outlet speed of the air conditioning system are adjusted by obtaining environmental information, clustering environmental status, generating Prompt and inputting the trained large language model to generate control instructions that adapt to complex environments.
Without retraining the model, quickly generate control strategies that adapt to complex environments, improving the adaptability and user comfort of the air conditioning system and reducing the use of computing resources.
Smart Images

Figure CN118463348B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of air conditioning control, and in particular to an air conditioning control method based on Prompt. Background Art
[0002] There are two traditional air-conditioning system control methods. The first control method is based on preset rules, that is, multiple control rules are set and the air-conditioning system is controlled according to the set rules. When the control method of the air-conditioning system needs to be changed, the preset rules are changed first, and then the changed rules are executed. For example, the air-conditioning is controlled to perform corresponding operations according to a predetermined schedule or a fixed temperature.
[0003] The second control method is based on neural networks, which use data from different environments to train the neural network to optimize the control strategy of the air conditioning system. However, this method requires a long training time, is difficult to adapt to complex environments, and cannot adjust the control strategy of the air conditioning system in time when the environment changes rapidly.
[0004] Therefore, an air conditioning control method is needed that can quickly change the control strategy according to environmental changes and generate control strategies suitable for complex environments without retraining the model. Summary of the invention
[0005] In order to overcome the problems existing in the related art, the purpose of the present invention is to provide an air conditioning control method based on Prompt, which can quickly change the control strategy according to environmental changes, and can generate a control strategy suitable for complex environments without retraining the model.
[0006] A prompt-based air conditioning control method, comprising:
[0007] Acquire environmental information, cluster the environmental information, and obtain an environmental state;
[0008] Generate a prompt according to the environmental status, air conditioning status and user instructions;
[0009] Input the Prompt into a trained large language model to generate a control instruction;
[0010] The target air conditioner is controlled to execute the control instruction and adjust the environmental parameters.
[0011] In a preferred technical solution of the present invention, before generating a Prompt according to the environmental state, the air conditioning state and the user instruction, the method further includes:
[0012] Acquire input speech, and convert the input speech into a first text;
[0013] Acquire a second text, and combine the first text and the second text to obtain a combined text; wherein the second text is a text provided by a text input module;
[0014] The combined text is converted into user instructions.
[0015] In a preferred technical solution of the present invention, the generating of a Prompt according to the environmental state, the air conditioning state and the user instruction includes:
[0016] The environmental state, air conditioning state and user instructions are input into a Prompt generator, and the environmental state, air conditioning state and user instructions are processed into data structures to obtain a Prompt; wherein the Prompt includes the environmental state, the air conditioning state and the user instructions.
[0017] In a preferred technical solution of the present invention, clustering the environmental information to obtain the environmental state includes:
[0018] Preprocessing the environmental information to obtain preprocessed data;
[0019] Select K test sample points from the preprocessed data as cluster centers; wherein the abscissa of the test sample point is temperature, the ordinate of the test sample point is humidity, and each cluster center corresponds to an environmental state;
[0020] Calculate the distance between the test sample point and each of the cluster centers;
[0021] The test sample points are classified as the cluster center corresponding to the minimum value of the distance to obtain the environmental state.
[0022] In a preferred technical solution of the present invention, the step of inputting the Prompt into a trained large language model to generate a control instruction includes:
[0023] Input the Prompt into a trained large language model to determine the target air conditioner;
[0024] A control instruction corresponding to the target air conditioner is generated according to the environmental state, the air conditioner state and the user instruction.
[0025] In a preferred technical solution of the present invention, after the control target air conditioner executes the control instruction and adjusts the environmental parameters, the control target air conditioner further comprises:
[0026] The evaluation function is calculated according to the following formula:
[0027] ;
[0028] Wherein, L is the evaluation function, ai is the energy consumption of the i-th room, n is the total number of rooms, α is the adjustment parameter, T is the preset temperature, t i is the temperature of the ith room.
[0029] In a preferred technical solution of the present invention, after calculating the evaluation function, the method further includes:
[0030] Matching the prompt voice and the prompt screen according to the value of the evaluation function;
[0031] Sending the prompt voice to the user;
[0032] The control display screen displays the prompt screen.
[0033] In a preferred technical solution of the present invention, the control target air conditioner executes the control instruction to adjust the environmental parameters, including:
[0034] Adjust the air outlet temperature and air outlet speed of the target air conditioner;
[0035] Detect whether the ambient temperature is within a first temperature range and whether the ambient humidity is within a first humidity range. If not, adjust the air outlet temperature and / or the air outlet speed of the target air conditioner until the ambient temperature is within the first temperature range and the ambient humidity is within the first humidity range.
[0036] In a preferred technical solution of the present invention, the step of adjusting the air outlet temperature and air outlet speed of the target air conditioner includes:
[0037] Detect whether the scenario mode sent by the user is received, and if so, match the air outlet temperature and air outlet speed corresponding to the target air conditioner according to the scenario mode.
[0038] In a preferred technical solution of the present invention, after the control display screen displays the prompt screen, the method further comprises:
[0039] Updating the environmental state to obtain an updated environmental state;
[0040] Updating the air-conditioning state to obtain an updated air-conditioning state;
[0041] The Prompt is updated according to the updated environmental status and the updated air-conditioning status.
[0042] The beneficial effects of the present invention are:
[0043] The present invention provides a Prompt-based air conditioning control method, which includes obtaining environmental information, which is diverse, for example, the environmental information includes environmental temperature and environmental humidity, clustering the environmental information, and obtaining an environmental state that meets the characteristics of the environmental information. Prompt is generated according to the environmental state, the air conditioning state and the user instruction, and the current environment of the air conditioner, the current state of the air conditioner and the operation that the user expects the air conditioner to perform are combined. Prompt is input into a trained large language model to generate a control instruction. The data in the Prompt has a structured characteristic, and the trained large language model can quickly convert the data in the Prompt into a control instruction that the air conditioner can execute. The control instruction corresponding to the Prompt carries the user's expectations. When the air conditioner is in a complex environment or the environment in which the air conditioner is located changes rapidly, the control instruction that adapts to the complex environment or the rapidly changing environment can be generated without retraining the model, and too much environmental data is not required, and the computing resources occupied are relatively small. The target air conditioner is controlled to execute the control instruction, and the environmental parameters are adjusted to improve the comfort of the user's environment. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 is a flow chart of the Prompt-based air conditioning control method provided by the present invention;
[0045] Figure 2 It is a schematic diagram of a flow chart of generating a user instruction provided by the present invention;
[0046] Figure 3 It is a schematic diagram of the process of clustering environmental information provided by the present invention;
[0047] Figure 4 It is a schematic diagram of a process of sending a prompt voice and a prompt screen to a user provided by the present invention;
[0048] Figure 5 It is a schematic diagram of the process of adjusting environmental parameters provided by the present invention. DETAILED DESCRIPTION
[0049] The preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.
[0050] Example 1
[0051] like Figure 1 As shown, this embodiment provides an air conditioning control method based on Prompt, including:
[0052] S1: Acquire environmental information, cluster the environmental information, and obtain an environmental state.
[0053] Clustering is to divide a data set into different classes according to a specific standard such as distance, so that the similarity of data samples in the same class is as large as possible, and the difference of data samples in different classes is as large as possible. That is, after the clustering operation, data samples in the same class are clustered together as much as possible, and data samples in different classes are separated as much as possible.
[0054] Environmental information includes ambient temperature, ambient humidity, building type, room area and weather conditions. Building types include large offices, small offices and hospitals. Weather conditions include hot and dry, hot and humid, and warm and dry. Users can configure the required environmental information in the air conditioning system.
[0055] S2: Generate a Prompt according to the environmental status, air conditioning status and user instructions.
[0056] The environment status, air conditioning status and user instructions are input into the prompt generator to generate a prompt, which includes a description of the current status, control instructions, demonstration of related status and other specific context information.
[0057] The content of the Prompt generator is a typical three-part structure. The core is the logic of loop optimization. A prompt word optimization function is set in the Prompt, which requires the user to continuously input content. The prompt word optimization function outputs structured data.
[0058] In the case of less interactive data of the real-world environment provided to Prompt, Prompt is generated by combining user needs and actual data using a Prompt generator, which can quickly adapt to complex and rapidly changing environments.
[0059] S3: Input the Prompt into the trained large language model to generate a control instruction.
[0060] LLM (Large Language Model) is a deep learning model trained based on massive text data. LLM can not only generate natural language text, but also deeply understand the meaning of text. LLM is used to process various natural language tasks, such as text summarization, question answering, and translation. The large language model can use the GPT-3 model or the GPT-4 model, which is not limited here.
[0061] The large language model is used as the decision maker, and generates control actions for the target air conditioner according to the information contained in the Prompt.
[0062] S4: Control the target air conditioner to execute the control instruction and adjust the environmental parameters.
[0063] The target air conditioner is controlled to adjust the air outlet temperature and air outlet speed, and to adjust the ambient temperature and ambient humidity. Preferably, the control instructions, environmental status, air conditioner status and user instructions are stored in the air conditioning system. After executing the control instructions and adjusting the environmental parameters, the operation results are fed back to the user by voice or screen display, and the operation results include the time when the control instructions are started to be executed, and the changes in the ambient temperature and ambient humidity over time after the control instructions are executed.
[0064] The present embodiment provides a Prompt-based air conditioning control method, which includes obtaining environmental information, which is diverse. For example, the environmental information includes environmental temperature and environmental humidity, clustering the environmental information, and obtaining an environmental state that meets the characteristics of the environmental information. Prompt is generated according to the environmental state, air conditioning state, and user instructions, combined with the current environment of the air conditioner, the current state of the air conditioner, and the operation that the user expects the air conditioner to perform. Prompt is input into a trained large language model to generate a control instruction. The data in Prompt has a structured characteristic, and the trained large language model can quickly convert the data in Prompt into a control instruction that can be executed by the air conditioner. The control instruction corresponding to the Prompt carries the user's expectations. When the air conditioner is in a complex environment or the environment in which the air conditioner is located changes rapidly, the control instruction that adapts to the complex environment or the rapidly changing environment can be generated without retraining the model, and too much environmental data is not required, and the computing resources occupied are relatively small. The target air conditioner is controlled to execute the control instruction, adjust the environmental parameters, and improve the user's comfort.
[0065] Example 2
[0066] like Figure 2 As shown, this embodiment provides an air conditioning control method based on Prompt. This embodiment only describes the differences from Embodiment 1. Before generating Prompt according to the environmental state, air conditioning state and user instruction, it also includes:
[0067] S11': Acquire input speech, and convert the input speech into a first text.
[0068] S12': Acquire a second text, and combine the first text and the second text to obtain a combined text; wherein the second text is a text provided by a text input module.
[0069] S13': Convert the combined text into a user instruction.
[0070] The air conditioning system includes an input module, a speech recognition module and a text recognition module. The input module classifies the input user speech and / or control text, sends the user speech to the speech recognition module, and sends the control text to the text recognition module. The speech recognition module uses a neural network for end-to-end speech recognition.
[0071] The air conditioning system can communicate with the user's terminal device and receive the second text sent by the terminal device, and the terminal device includes a mobile phone and a computer. The input voice includes the user's expectations for the environmental parameters of the current environment, and the input voice is converted into the first text, and the user instruction is generated by combining the first text and the second text.
[0072] If only the input voice is received, the input voice is converted into the first text, and the first text is converted into the user instruction. If only the second text sent by the terminal device is received, the second text is converted into the user instruction.
[0073] The generating of a Prompt according to the environmental state, the air conditioning state and the user instruction includes:
[0074] The environmental state, air conditioning state and user instructions are input into a Prompt generator, and the environmental state, air conditioning state and user instructions are processed into data structures to obtain a Prompt; wherein the Prompt includes the environmental state, the air conditioning state and the user instructions.
[0075] The prompt generator performs data structuring processing on the environment status, air conditioning status and user instructions, so that parameters related to the environment status, parameters related to the air conditioning status and user instructions can be more easily recognized by the large language model, thereby improving the data processing efficiency of the large language model.
[0076] This embodiment converts the input voice into a first text when simultaneously receiving an input voice and a second text sent by a user terminal, combines the first text and the second text to obtain a combined text, and converts the combined text into a user instruction. It can comprehensively adjust the environmental parameters of the user's environment according to the user's needs and improve the user's comfort.
[0077] Example 3
[0078] like Figure 3 As shown, this embodiment provides an air conditioning control method based on Prompt. This embodiment only describes the differences from Embodiment 1. The environmental information is clustered to obtain the environmental state, including:
[0079] S12: Preprocessing the environmental information to obtain preprocessed data.
[0080] S13: Select K test sample points from the preprocessed data as cluster centers; wherein the horizontal coordinate of the test sample point is temperature, the vertical coordinate of the test sample point is humidity, and each cluster center corresponds to an environmental state.
[0081] S14: Calculate the distance between the test sample point and each of the cluster centers.
[0082] S15: Classify the test sample points as the cluster center corresponding to the minimum value of the distance to obtain the environmental state.
[0083] Before step S12, the method further includes step S11: acquiring environmental information.
[0084] Preprocessing includes removing abnormal test sample points and normalizing normal test sample points. Abnormal test sample points include test sample points with large deviations from the mean of all test sample points. Abnormal test sample points are removed and then normalized so that the normalization process is not affected by abnormal test sample points and the remaining normal test sample points are unified to the same scale.
[0085] After some test sample points are allocated, the cluster center of each cluster will be recalculated based on the existing test sample points in the cluster. Multiple allocation thresholds can be set. When the test sample points are greater than or equal to different allocation thresholds, the cluster center of each cluster is recalculated. For example, there are 100 test sample points in total, and 3 allocation thresholds are set. The first allocation threshold is 30, the second allocation threshold is 50, and the third allocation threshold is 80. After clustering 30 test sample points, the cluster centers of each cluster are updated for the first time; after clustering 50 test sample points, the cluster centers of each cluster are updated for the second time; after clustering 80 test sample points, the cluster centers of each cluster are updated for the third time.
[0086] The test sample points are located in a two-dimensional space. The Euclidean distances between the test sample points and each cluster center are calculated. All Euclidean distances are sorted in descending order, the minimum distance value is screened out, and the test sample points are classified as the cluster center corresponding to the minimum distance value.
[0087] The step of inputting the Prompt into a trained large language model to generate a control instruction includes:
[0088] S31: Input the Prompt into the trained large language model to determine the target air conditioner.
[0089] S32: Generate a control instruction corresponding to the target air conditioner according to the environmental state, the air conditioner state and the user instruction.
[0090] Users can remotely control the air conditioner through mobile phone APP or other networked devices, such as opening and closing in advance, etc. The air conditioning system transmits the control command to the target air conditioner through the network to achieve remote control. The use of trained large language models to generate control instructions allows the system to achieve a balance between comfort and energy consumption in different environments, while also having the characteristics of strong adaptability and robustness.
[0091] This embodiment clusters the environmental information to obtain the environmental state, including preprocessing the environmental information to obtain preprocessed data. K test sample points are selected from the preprocessed data as cluster centers, wherein the horizontal coordinate of the test sample point is temperature, the vertical coordinate of the test sample point is humidity, and each cluster center corresponds to an environmental state. The distance between the test sample point and each cluster center is calculated, and the test sample point is classified as the cluster center corresponding to the minimum distance value to obtain the environmental state. Clustering is an unsupervised learning method, which uses clustering to divide all test sample points into multiple categories of environmental states. The similarity of test sample points corresponding to one cluster center, i.e., one category of environmental states, is higher, and the similarity of test sample points corresponding to different cluster centers, i.e., different categories of environmental states, is lower.
[0092] Example 4
[0093] This embodiment provides an air conditioning control method based on Prompt. This embodiment only describes the differences from Embodiment 1. After the control target air conditioning executes the control instruction and adjusts the environmental parameters, it also includes:
[0094] The evaluation function is calculated according to the following formula:
[0095] ;
[0096] Wherein, L is the evaluation function, a i is the energy consumption of the i-th room, n is the total number of rooms, α is the adjustment parameter, T is the preset temperature, t i is the temperature of the ith room.
[0097] Environmental parameters include ambient temperature and ambient humidity. The ambient temperature needs to be kept at 22 degrees Celsius while minimizing energy consumption. The air conditioning action is encoded as a real number ranging from -1 to 1. Negative values indicate cooling mode, and positive values indicate heating mode. The absolute value of the air conditioning action corresponds to the degree of opening of the valve, and the absolute value of the air conditioning action is proportional to the energy consumption, that is, the larger the absolute value of the air conditioning action, the greater the energy consumption.
[0098] Under the condition of taking into account both comfort and energy consumption, the above evaluation function is used to evaluate the effect of the air conditioning system on adjusting environmental parameters. The preset temperature T can be set as needed. In this embodiment, the preset temperature T is 22 degrees Celsius as an example.
[0099] like Figure 4 As shown, after calculating the evaluation function, it also includes:
[0100] S51': Matching the prompt voice and the prompt screen according to the value of the evaluation function.
[0101] S52': Send the prompt voice to the user.
[0102] S53': controlling the display screen to display the prompting picture.
[0103] The value of the evaluation function has a range, which is divided into M intervals. Different intervals represent the evaluation of the comprehensive ability of the air conditioner to adjust environmental parameters and air conditioning energy consumption. Each interval corresponds to a set of prompt voices and prompt screens, and there are M groups of prompt voices and prompt screens in M intervals.
[0104] Send a prompt voice to the user terminal, and the voice player of the user terminal plays the prompt voice. The voice player can be a speaker or a headset. The air conditioner can be provided with a display screen, or an external display screen can be connected to control the display screen to display the prompt screen.
[0105] In this embodiment, after the target air conditioner starts to execute the control instruction and starts to adjust the environmental parameters such as the ambient temperature, the value of the evaluation function is calculated. The evaluation function combines the temperature of n rooms and the energy consumption of n rooms. The energy consumption of the target air conditioner and the comfort of the environment can be evaluated simultaneously by using the evaluation function.
[0106] Example 5
[0107] like Figure 5 As shown, this embodiment provides an air conditioning control method based on Prompt. This embodiment only describes the differences from Embodiment 1. The control target air conditioning executes the control instruction to adjust the environmental parameters, including:
[0108] S41: Adjust the air outlet temperature and air outlet speed of the target air conditioner.
[0109] Detect whether the scenario mode sent by the user is received, and if so, match the air outlet temperature and air outlet speed corresponding to the target air conditioner according to the scenario mode.
[0110] Multiple scene modes are preset, such as sleep mode, work mode and sports mode, each of which corresponds to different air outlet temperatures and air outlet speeds. Users can select the appropriate scene mode, and the air conditioning system automatically sets the air conditioning parameters according to the scene mode to provide a comfortable environment. The currently selected scene mode is included in the prompt, so that the large language model can generate control instructions that meet the scene requirements.
[0111] Users can set the air temperature and air speed through the APP or control panel. The air conditioning system adds user preference information to the prompt so that the large language model can consider personalized needs when generating control instructions. The large language model can learn the user's usage habits, automatically adjust personalized parameters, and provide a more intimate control experience.
[0112] The air conditioning system is equipped with temperature sensors and humidity sensors to monitor changes in environmental parameters in real time. The air conditioning system adjusts the outlet temperature and air speed of the target air conditioner according to the ambient temperature collected by the temperature sensor and the ambient humidity collected by the humidity sensor to adapt to environmental changes and provide a comfortable experience.
[0113] The air conditioning system sets target parameters in the prompt, such as minimizing energy consumption while satisfying user comfort. The large language model generates control instructions that take into account both comfort and energy efficiency based on the target parameters. The air conditioning system monitors the operating status of the target air conditioner in real time and diagnoses common faults, such as abnormal cooling and sensor failures. The fault information is included in the prompt for the large language model to provide troubleshooting and repair suggestions. The air conditioning system automatically performs fault repair operations based on the suggestions, reducing the time spent by users on troubleshooting and repairing faults, and reducing users' maintenance costs. The large language model generates an air conditioning control strategy that takes into account both comfort and energy efficiency based on the target parameters. The air conditioning system executes the optimized control instructions, saving energy while providing users with a good user experience.
[0114] S42: Detect whether the ambient temperature is within the first temperature range and whether the ambient humidity is within the first humidity range. If not, adjust the air outlet temperature and / or the air outlet speed of the target air conditioner until the ambient temperature is within the first temperature range and the ambient humidity is within the first humidity range.
[0115] If the ambient temperature is within the first temperature range and the ambient humidity is within the first humidity range, the air outlet temperature and air outlet speed of the target air conditioner are kept unchanged; if the ambient temperature is within the first temperature range and the ambient humidity is not within the first humidity range, the air outlet temperature of the target air conditioner is kept unchanged and the air outlet speed of the target air conditioner is adjusted; if the ambient temperature is not within the first temperature range and the ambient humidity is within the first humidity range, the air outlet speed of the target air conditioner is kept unchanged and the air outlet temperature of the target air conditioner is adjusted.
[0116] After the control display screen displays the prompt screen, the method further includes:
[0117] S54': Update the environmental state to obtain an updated environmental state.
[0118] S55': updating the air-conditioning state to obtain an updated air-conditioning state.
[0119] S56': Update the Prompt according to the updated environmental status and the updated air-conditioning status.
[0120] The air conditioning system records user operation history and environmental change data, and uses machine learning algorithms to analyze user usage habits and preferences. Based on the learned user profile, the air conditioning system can automatically adjust the control strategy to provide a personalized air conditioning experience. User profile information is incorporated into Prompt, allowing the large language model to generate control suggestions that are closer to user needs.
[0121] The air conditioning system can work with other smart devices in the home, such as curtains and lights. Prompt contains comprehensive information about the home environment, such as temperature and light, for the large language model to formulate an overall optimized control strategy. The air conditioning system interacts with other devices through network protocols to achieve coordinated control of the target air conditioning and the home environment.
[0122] The air conditioning system uses natural language processing technologies, such as named entity recognition and semantic role labeling, to deeply understand the semantic meaning of the user's natural language instructions. The prompt contains semantic information about the user's instructions, such as the purpose and context of the instructions, to help the large language model generate accurate control commands. The air conditioning system is able to handle complex semantic expressions, not just simple keyword matching.
[0123] The air conditioning system can interact with users through voice and text to improve the user experience. The air conditioning system analyzes the user's voice intonation, text expression and other non-verbal signals to identify the user's emotional state, including happiness, anxiety and fatigue, and uses this as a basis for more considerate interaction. Specific implementations include:
[0124] Voice emotion analysis: The air conditioning system uses emotion recognition algorithms to analyze the emotional color in the user's voice, such as changes in speaking speed, pitch and volume, to understand the user's emotions.
[0125] Text sentiment analysis: Perform sentiment analysis on the text input by the user, including word choice, tone, and expression, so as to infer the user's emotional state.
[0126] Emotional feedback: Based on the user’s emotional state, the air conditioning system responds with a corresponding tone and wording to better meet the user’s emotional needs and improve the interactive experience.
[0127] The air conditioning system can predict future environmental changes based on historical data and adjust the target air conditioner in advance. The air conditioning system predicts future environmental changes based on historical environmental data and machine learning algorithms, and adjusts the working mode of the target air conditioner in advance to ensure that corresponding adjustments are made before the environment changes. Specific implementations include:
[0128] Data analysis: The system collects and analyzes historical environmental data, such as temperature, humidity, and weather, to understand patterns of environmental change.
[0129] Large model prediction function: adjust the large model based on historical data to predict future environmental change trends.
[0130] Predictive adjustment: The system adjusts the working mode of the target air conditioner in advance according to the prediction results of the large model to adapt to future environmental changes and improve comfort and energy efficiency.
[0131] The air conditioning system supports multiple interaction modes, including voice, gesture and touch. Users can choose the appropriate interaction mode according to their needs, and the air conditioning system will generate corresponding control instructions based on different inputs. Specific implementations include:
[0132] Voice interaction: Users can control through voice commands, and the system converts voice commands into control commands through voice recognition technology.
[0133] Gesture interaction: Users can control the device through gestures, such as sliding gestures to adjust the air temperature or air speed.
[0134] Touch interaction: The air conditioning system is equipped with a touch screen interface, through which users can operate, such as selecting modes and adjusting parameters.
[0135] The air conditioning system ensures user privacy. For example, voice commands will not be recorded and saved, and personal preferences will not be leaked. User data is encrypted for storage and transmission to prevent illegal access to data. The access rights of each module within the air conditioning system to user data are restricted to ensure that data is only accessed when necessary.
[0136] After the control display screen of this embodiment displays the prompt screen, the environment state is updated to obtain an updated environment state. The air conditioning state is updated to obtain an updated air conditioning state, and the Prompt is updated according to the updated environment state and the updated air conditioning state. According to the feedback information, the air conditioning system can adjust the next operation or update the state of the air conditioning system.
[0137] Example 6
[0138] This embodiment provides an electronic device, which includes a memory and a processor.
[0139] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0140] The memory may include various types of storage units such as system memory, read-only memory (ROM), and permanent storage.
[0141] The memory stores executable codes, and when the executable codes are processed by the processor, the processor can execute part or all of the Prompt-based air conditioning control method mentioned above.
[0142] Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values of the parts and steps set forth in these embodiments do not limit the scope of the application. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a restriction. Therefore, other examples of exemplary embodiments can have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, and therefore, once a certain item is defined in an accompanying drawing, it does not need to be further discussed in the subsequent accompanying drawings.
[0143] It should be understood that spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figures is inverted, a device described as "above" or "above other devices or structures" would then be positioned "below" or "below the other devices or structures." Thus, the exemplary term "above" can include both "above" and "below." The device may also be oriented in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used herein shall be interpreted accordingly.
[0144] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only to facilitate the distinction between corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of this application.
[0145] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A prompt-based air conditioning control method, characterized in that: include: Acquire environmental information, cluster the environmental information, and obtain environmental status; wherein the environmental information includes environmental temperature, environmental humidity, building type, room area, and weather conditions; Generate a prompt according to the environmental status, air conditioning status and user instructions; Input the Prompt into a trained large language model to generate a control instruction; Controlling the target air conditioner to execute the control instruction and adjust the environmental parameters; The clustering of the environmental information to obtain the environmental state includes: Preprocessing the environmental information to obtain preprocessed data; Select K test sample points from the preprocessed data as cluster centers; wherein the abscissa of the test sample point is temperature, the ordinate of the test sample point is humidity, and each cluster center corresponds to an environmental state; Calculate the distance between the test sample point and each of the cluster centers; Classifying the test sample point as the cluster center corresponding to the minimum value of the distance to obtain the environmental state; After the controlled target air conditioner executes the control instruction and adjusts the environmental parameters, the control method further includes: The evaluation function is calculated according to the following formula: ; Wherein, L is the evaluation function, a i is the energy consumption of the i-th room, n is the total number of rooms, α is the adjustment parameter, T is the preset temperature, t i is the temperature of the ith room.
2. The Prompt-based air conditioning control method according to claim 1, characterized in that: Before generating a Prompt according to the environmental state, the air conditioning state and the user instruction, the method further includes: Acquire input speech, and convert the input speech into a first text; Acquire a second text, and combine the first text and the second text to obtain a combined text; wherein the second text is a text provided by a text input module; The combined text is converted into user instructions.
3. The Prompt-based air conditioning control method according to claim 1, characterized in that: The generating of a Prompt according to the environmental state, the air conditioning state and the user instruction includes: The environmental state, air conditioning state and user instructions are input into a Prompt generator, and the environmental state, air conditioning state and user instructions are processed into data structures to obtain a Prompt; wherein the Prompt includes the environmental state, the air conditioning state and the user instructions.
4. The Prompt-based air conditioning control method according to claim 3, characterized in that: The step of inputting the Prompt into a trained large language model to generate a control instruction includes: Input the Prompt into a trained large language model to determine the target air conditioner; A control instruction corresponding to the target air conditioner is generated according to the environmental state, the air conditioner state and the user instruction.
5. The Prompt-based air conditioning control method according to claim 1, characterized in that: After calculating the evaluation function, it also includes: Matching the prompt voice and the prompt screen according to the value of the evaluation function; Sending the prompt voice to the user; The control display screen displays the prompt screen.
6. The air conditioning control method based on Prompt according to claim 1, characterized in that: The control target air conditioner executes the control instruction to adjust the environmental parameters, including: Adjust the air outlet temperature and air outlet speed of the target air conditioner; Detect whether the ambient temperature is within a first temperature range and whether the ambient humidity is within a first humidity range. If not, adjust the air outlet temperature and / or the air outlet speed of the target air conditioner until the ambient temperature is within the first temperature range and the ambient humidity is within the first humidity range.
7. The Prompt-based air conditioning control method according to claim 6, characterized in that: The step of adjusting the air outlet temperature and air outlet speed of the target air conditioner includes: Detect whether the scenario mode sent by the user is received, and if so, match the air outlet temperature and air outlet speed corresponding to the target air conditioner according to the scenario mode.
8. The Prompt-based air conditioning control method according to claim 5, characterized in that: After the control display screen displays the prompt screen, the method further includes: Updating the environmental state to obtain an updated environmental state; Updating the air-conditioning state to obtain an updated air-conditioning state; The Prompt is updated according to the updated environmental status and the updated air-conditioning status.
