Air conditioning equipment and control method of air conditioning equipment

Abstract

The invention relates to air conditioning equipment and a control method of the air conditioning equipment and relates to the field of air conditioners, and the air conditioning equipment comprises a refrigerant circulation loop which comprises a compressor, a condenser, an expansion valve and an evaporator and is used for achieving refrigerating or heating circulation of refrigerants; the sound acquisition module is configured to receive a voice instruction; the environment information acquisition module is configured to acquire environment information; the controller is configured to obtain environment information after receiving a voice instruction input by a user, and extract voiceprint features and operation intention information from the voice instruction; identifying a target identity category of the user according to the voiceprint feature; and according to the target identity category, the operation intention information and the environment information, the air conditioner equipment is controlled to operate. By adopting the technical scheme of the invention, the use convenience can be improved.

Description

Technical Field

[0001] This application relates to the field of air conditioning equipment technology, and in particular to an air conditioning device and a control method for the air conditioning device. Background Technology

[0002] With the continuous improvement of modern living standards and the development of cooling and heating technologies, air conditioning equipment has become widely used. Current air conditioning equipment typically features multi-dimensional parameter adjustment functions and supports receiving user commands through various interactive methods such as remote control, voice, and mobile terminals.

[0003] In traditional technology, the adjustment of air conditioning equipment parameters strictly depends on user input. If no new parameter settings are entered at the time of startup, the air conditioning equipment will use the combination of operating parameters saved when it was last shut down.

[0004] However, when the usage scenario changes, users often need to reset multiple parameters such as temperature, operating mode, fan speed, and airflow direction to obtain a suitable indoor environment. This frequent and complex operation significantly reduces the ease of use of air conditioning equipment. Summary of the Invention

[0005] Therefore, it is necessary to provide an air conditioning device and a control method for the air conditioning device that can improve ease of use in response to the above-mentioned technical problems.

[0006] In a first aspect, this application provides an air conditioning device, comprising:

[0007] The refrigerant circulation loop, including the compressor, condenser, expansion valve and evaporator, is used to realize the refrigeration or heating cycle of the refrigerant;

[0008] The sound acquisition module is configured to receive voice commands;

[0009] The environmental information collection module is configured to collect environmental information;

[0010] The controller is configured as follows:

[0011] After receiving a voice command from the user, the system acquires environmental information and extracts voiceprint features and operation intent information from the voice command; it identifies the user's target identity category based on the voiceprint features; and it controls the operation of the air conditioning equipment based on the target identity category, operation intent information, and environmental information.

[0012] Technical Effects: Upon receiving a user's voice command, the system simultaneously extracts voiceprint features and operational intent information, and collects external environmental data. Then, it accurately identifies the user's target identity category through voiceprint features, and further integrates the target identity category, operational intent information, and external environmental data for comprehensive analysis. This captures the user's current actual needs and automatically generates and executes a combination of operating parameters that matches those needs, achieving a shift from passively executing user input to proactively adapting to actual requirements. When the same user issues power-on or simple adjustment commands in different scenarios, the air conditioning device can automatically match a suitable combination of parameters such as temperature, mode, and fan speed based on the user's comfort needs corresponding to their identity category, current specific intent, and real-time environmental status. This significantly reduces the tedious manual adjustments required due to historical parameter mismatches. This not only reduces interaction complexity but also allows users to obtain a personalized comfort environment that matches their identity category and real-time environment through simpler voice interaction, thus significantly improving the ease of use and smoothness of operation of the air conditioning device.

[0013] Secondly, this application provides a method for controlling an air conditioning device, including:

[0014] In response to user-inputted voice commands, the system acquires environmental information and extracts voiceprint features and operational intent information from the voice commands; it identifies the user's target identity category based on the voiceprint features; and it controls the operation of the air conditioning equipment based on the target identity category, operational intent information, and environmental information.

[0015] Technical Effects: Upon receiving a user's voice command, the system simultaneously extracts voiceprint features and operational intent information, and collects external environmental data. Then, it accurately identifies the user's target identity category through voiceprint features, and further integrates the target identity category, operational intent information, and external environmental data for comprehensive analysis. This captures the user's current actual needs and automatically generates and executes a combination of operating parameters that matches those needs, achieving a shift from passively executing user input to proactively adapting to actual requirements. When the same user issues power-on or simple adjustment commands in different scenarios, the air conditioning device can automatically match a suitable combination of parameters such as temperature, mode, and fan speed based on the user's comfort needs corresponding to their identity category, current specific intent, and real-time environmental status. This significantly reduces the tedious manual adjustments required due to historical parameter mismatches. This not only reduces interaction complexity but also allows users to obtain a personalized comfort environment that matches their identity category and real-time environment through simpler voice interaction, thus significantly improving the ease of use and smoothness of operation of the air conditioning device. Attached Figure Description

[0016] Figure 1 A schematic diagram of the hardware configuration of the refrigerant circulation loop provided in some embodiments of this application;

[0017] Figure 2 This is a schematic diagram of the hardware configuration of an air conditioning device provided in some embodiments of this application;

[0018] Figure 3 This is a schematic diagram of the hardware configuration of an air conditioning device provided in other embodiments of this application;

[0019] Figure 4 A flowchart illustrating a method for controlling an air conditioning device according to some embodiments of this application;

[0020] Figure 5 A timing diagram illustrating a method for controlling an air conditioning device according to some embodiments of this application;

[0021] Figure 6 The following are schematic diagrams illustrating home Internet of Things (IoT) scenarios provided in some embodiments of this application;

[0022] Figure 7 This is a flowchart illustrating the process of controlling the operation of an air conditioning device according to some embodiments of this application;

[0023] Figure 8 A flowchart illustrating the process of controlling an air conditioning device to operate according to a target parameter set, provided in some embodiments of this application;

[0024] Figure 9 A flowchart illustrating the implementation of parameter fusion process provided in some embodiments of this application;

[0025] Figure 10 A flowchart illustrating the process of determining the target parameter set provided in some embodiments of this application;

[0026] Figure 11 A flowchart illustrating the weight matrix update process provided in some embodiments of this application;

[0027] Figure 12 A flowchart illustrating the weight matrix update process provided in other embodiments of this application;

[0028] Figure 13 A flowchart illustrating the implementation of the intelligent power-on process provided in some embodiments of this application;

[0029] Figure 14 A flowchart illustrating the process of determining a target identity category provided in some embodiments of this application;

[0030] Figure 15 This is a flowchart illustrating a method for controlling an air conditioning device according to other embodiments of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0032] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0034] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0035] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0036] In the description of the embodiments in this application, the term "and / or" is merely a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. In addition, the character " / " in this document generally indicates that the related objects before and after are in an "or" relationship, and the term "multiple" refers to two or more (including two).

[0037] In this application embodiment, air conditioning equipment refers to intelligent devices with temperature and / or humidity regulation functions, such as intelligent air conditioning equipment.

[0038] In this application, the air conditioning equipment can achieve cooling or heating functions through its internal refrigerant circulation loop. The refrigerant circulation includes a series of processes, mainly involving four major processes: compression, condensation, expansion, and evaporation, and ultimately achieves air temperature regulation through heat exchange between the refrigerant and the air.

[0039] like Figure 1 As shown, the refrigerant circulation loop 10 includes a compressor 102, a condenser 104, an expansion valve 106, and an evaporator 108.

[0040] Compressor 102 compresses the refrigerant gas under high temperature and high pressure and discharges the compressed refrigerant gas. The discharged refrigerant gas flows into condenser 104. Condenser 104 condenses the compressed refrigerant into a liquid phase, and heat is released to the surrounding environment through the condensation process.

[0041] Expansion valve 106 expands the high-temperature, high-pressure liquid refrigerant condensed in condenser 104 into a low-temperature, low-pressure liquid refrigerant. Evaporator 108 evaporates the refrigerant that expands in expansion valve 106 and returns the low-temperature, low-pressure refrigerant gas to compressor 102.

[0042] Throughout the cycle, the air conditioning unit can regulate the indoor temperature through heat exchange between the refrigerant and the indoor air.

[0043] Air conditioning equipment typically consists of two parts: an outdoor unit and an indoor unit.

[0044] Air conditioning equipment typically includes an outdoor unit and an indoor unit. The outdoor unit may include a compressor and an outdoor heat exchanger, while the indoor unit may include an indoor heat exchanger. An expansion valve may be located in either the indoor or outdoor unit.

[0045] Indoor and outdoor heat exchangers function as either condensers or evaporators. When the indoor heat exchanger is used as a condenser, the air conditioning unit functions as a heater in heating mode; when the indoor heat exchanger is used as an evaporator, the air conditioning unit functions as a cooler in cooling mode.

[0046] Figure 2 Block diagrams of air conditioning equipment provided in some embodiments of this application. For example... Figure 2 As shown in the figure, the air conditioning equipment 20 of this application embodiment includes a refrigerant circulation loop 10, an outdoor fan 202, an indoor fan 206, and a controller 204.

[0047] The refrigerant circulation loop 10 includes a compressor 102, a condenser 104, an expansion valve 106, and an evaporator 108, which are used to realize the cooling or heating cycle of the refrigerant.

[0048] The outdoor fan 202 is used to drive outdoor air through the outdoor heat exchanger by rotation, so that the refrigerant can exchange heat with the outdoor air. The outdoor heat exchanger is used as a condenser 102 in the refrigeration cycle and as an evaporator 108 in the heating cycle.

[0049] The indoor fan 206 is used to drive indoor air through the indoor heat exchanger by rotation, so that the refrigerant and the indoor air can exchange heat. The indoor heat exchanger is used as an evaporator 108 in the refrigeration cycle and as a condenser 102 in the heating cycle.

[0050] As an example, in the refrigeration cycle, the low-temperature, low-pressure refrigerant gas is compressed in the compressor 102 of the outdoor unit, becoming a high-temperature, high-pressure gas. The high-temperature, high-pressure gas flows into the outdoor heat exchanger of the outdoor unit, where it acts as a condenser 104, releasing heat to the outdoor air with the help of an outdoor fan, and condensing into a medium-temperature, high-pressure liquid. The medium-temperature, high-pressure liquid flows through the expansion valve 106, where its pressure and temperature drop sharply, becoming a low-temperature, low-pressure vapor-liquid mixture. The low-temperature, low-pressure refrigerant enters the indoor heat exchanger of the indoor unit, where it acts as an evaporator 108, absorbing heat from the indoor air with the help of an indoor fan, and evaporating into a low-temperature, low-pressure gas, thereby lowering the indoor temperature. Subsequently, the gas returns to the compressor 102, starting the next cycle.

[0051] As an example, in the heating cycle, the low-temperature, low-pressure refrigerant gas is compressed in the compressor 102 of the outdoor unit, becoming a high-temperature, high-pressure gas. The high-temperature, high-pressure gas flows into the indoor heat exchanger of the indoor unit, where it acts as a condenser 104, releasing heat to the indoor air with the help of an indoor fan, condensing into a medium-temperature, high-pressure liquid, thereby raising the indoor temperature. The medium-temperature, high-pressure liquid flows through the expansion valve 106, where the pressure and temperature drop sharply, becoming a low-temperature, low-pressure vapor-liquid mixture. The low-temperature, low-pressure refrigerant enters the outdoor heat exchanger of the outdoor unit, where it acts as an evaporator 108, absorbing heat from the outdoor air with the help of an outdoor fan, and evaporating into a low-temperature, low-pressure gas. Subsequently, the gas returns to the compressor 102, starting the next cycle.

[0052] In some embodiments, such as Figure 3 As shown, the air conditioning equipment 20 includes a refrigerant circulation loop 10, a sound acquisition module 208, an environmental information acquisition module 210, and a controller 204.

[0053] The sound acquisition module 208 can refer to a device capable of receiving sound wave signals and converting them into electrical signals, such as a microphone array or microphone. The sound acquisition module 208 can capture ambient sounds, including user voice commands. For example, when a user says "Turn on, set to 26 degrees," the sound acquisition module 208 converts this sound signal into audio data that can be processed by the controller.

[0054] The environmental information acquisition module 210 can refer to a sensor or sensor group used to detect and collect physical quantities of the environment in which the air conditioning equipment is located, including at least one of temperature sensors, humidity sensors, etc.

[0055] As an example, the environmental information acquisition module 210 can acquire environmental information in real time, periodically, or triggeredly, and store it in a specific path of the memory. The controller 204 can retrieve the required environmental information from the specific path of the memory as needed.

[0056] As another example, the environmental information acquisition module 210 can respond to the environmental information acquisition command sent by the controller 204, acquire environmental information, and return the acquired environmental information to the controller 204.

[0057] The controller 204 is configured to: after receiving a voice command input by a user, acquire environmental information and extract voiceprint features and operation intent information from the voice command; identify the user's target identity category based on the voiceprint features; and control the operation of the air conditioning equipment based on the target identity category, operation intent information, and environmental information.

[0058] In some embodiments, such as Figure 4 As shown, controller 204 is configured to perform the following steps:

[0059] Step 402: After receiving the user's voice command, obtain environmental information and extract voiceprint features and operation intent information from the voice command;

[0060] Step 404: Identify the user's target identity category based on voiceprint features;

[0061] Step 406: Control the operation of the air conditioning equipment based on the target identity category, operation intent information, and environmental information.

[0062] Voice commands can refer to voice signals issued by users to control air conditioning equipment, which contain specific operational intentions.

[0063] Voiceprint features can refer to individual distinguishable speech parameter features that reflect the biological characteristics or pronunciation habits of the speaking user, including at least one of the following: spectral features, formant trajectories, fundamental frequency variation patterns, etc.

[0064] As an example, the method for extracting voiceprint features from speech commands includes: first, preprocessing and framing the audio signal of the speech command; then, performing a Fast Fourier Transform (FFT) on each frame to obtain spectral data; next, mapping the spectral data to the Mel frequency scale using a Mel-scale filter bank; taking the logarithm of the output of the Mel-scale filter bank and performing a Discrete Cosine Transform (DCT); and finally, extracting the first 12-20 Mel-frequency cepstral coefficients to form a Mel-frequency cepstral coefficient feature sequence. Mel-frequency cepstral coefficients can reflect the speaker's vocal tract characteristics and pronunciation habits; therefore, the Mel-frequency cepstral coefficient feature sequence can be used as voiceprint features.

[0065] In some feasible embodiments, Mel-frequency cepstral coefficients typically reflect the static characteristics of the speaker's vocal tract shape. The first and second differences of the Mel-frequency cepstral coefficient feature sequence can be further calculated; these differences reflect dynamic characteristics. Therefore, the Mel-frequency cepstral coefficient feature sequence, the first and second differences can be concatenated into a high-dimensional feature vector, which can then be used as the voiceprint feature.

[0066] As another example, methods for extracting voiceprint features from voice commands include using a pre-trained deep neural network model to extract voiceprint features from voice commands and output the voiceprint features.

[0067] Operation intent information can refer to the operation or state that the user currently wants the air conditioning device to perform, which can be parsed from the semantic content of the voice command. For example, the operation intent can be extracted from the voice command "adjust to 26 degrees" as "set the temperature to 26℃"; and the operation intent can be extracted from the voice command "too hot" as "lower the temperature".

[0068] As an example, one method for extracting operational intent information from voice commands includes: first, converting the voice command into a text sequence using speech recognition technology; then, matching the text sequence against a predefined intent template library. The intent template library predefines various operational intents and their associated keywords or sentence patterns. The system determines the operational intent type by searching the text sequence for specific keywords or matching specific sentence patterns. If the parameter value corresponding to the operational intent type is extracted from the text sequence, both the operational intent type and the corresponding parameter value can be used as the operational intent information. If the parameter value corresponding to the operational intent type is not extracted from the text sequence, the operational intent type can be used as the operational intent information.

[0069] As another example, extracting operational intent information from voice commands includes inputting a text embedding vector, converted using speech recognition technology, into a trained deep learning model. The deep learning model then outputs the intent classification result and its associated parameter values. The intent classification result and its associated parameter values ​​can be used as operational intent information. In some feasible implementations, the deep learning model can employ a bidirectional LSTM (Long Short-Term Memory) + CRF (Conditional Random Field) model.

[0070] Environmental information may include at least one of the following: temperature, humidity, season, weather, and real-time time of the user's geographical location.

[0071] In some embodiments, such as Figure 5 As shown, the sound acquisition module can collect sound signals and convert them into audio data, which is then sent to the controller. The controller identifies whether there are voice commands in the audio data. After recognizing a voice command, it acquires environmental information collected by the environmental information acquisition module and processes the audio data corresponding to the voice command to extract voiceprint features and operation intent information. After extracting the voiceprint features, it compares the voiceprint features with a first voiceprint feature model library pre-built and stored inside the controller. By calculating the similarity or distance between the voiceprint features and each voiceprint feature model in the first voiceprint feature model library, it determines the most matching identity category and identifies this identity category as the target identity category for the current user. Subsequently, the controller uses the target identity category, operation intent information, and environmental information as input parameters, generates a running control command based on the preset control logic, and sends the running control command to the corresponding actuator. The actuator drives its associated physical components to perform actions based on the received running control command, providing a temperature environment that matches the user's needs.

[0072] The first voiceprint feature model library stores voiceprint feature models corresponding to various identity categories. These voiceprint feature models can be pre-built based on a large number of voice samples corresponding to different identity categories.

[0073] In some feasible embodiments, the preset control logic may include at least one of rule table lookup, fuzzy control algorithm or threshold judgment mechanism.

[0074] In some feasible embodiments, the actuator may include at least one of a compressor, a fan, and a deflector.

[0075] In some feasible embodiments, the operation control commands may include at least one of the following: adjusting the compressor frequency, fan speed, air guide vane angle, and switching operating modes.

[0076] In some feasible embodiments, such as Figure 6 As shown, the air conditioning unit 20 can be connected to a home IoT network, which in turn connects to at least one other home appliance 30. After identifying the user's target identity category based on voiceprint characteristics, the controller can be further configured to: generate linkage control commands based on the target identity category; and send linkage control commands to the other home appliances via the home IoT network. For example, when the target identity category is identified as a child, linkage control commands can be generated for the television and electric kettle, and sent to the television and electric kettle via the home IoT network to switch them to a disabled state. This disabled state can be unlocked by a non-child user, for example, by receiving a voice command from a non-child user instructing them to unlock.

[0077] In some feasible embodiments, to accommodate the actual needs of different users, a control mode switch or a control mode selection interface can be set. The control mode selection interface can be displayed to the user through the air conditioning unit's display interface, speakers, or other output modules, or through application software on the user terminal associated with the air conditioning unit. In response to the user's selection operation on the control mode selection interface, the air conditioning unit is controlled to operate according to the user-selected control mode. The control mode can include an operation intent control mode and an intelligent control mode.

[0078] If the air conditioning unit is in operation intent control mode, after receiving a voice command from the user, the controller extracts the operation intent information from the voice command and controls the operation of the air conditioning unit according to the operation intent information. It no longer acquires environmental information or extracts voiceprint features; the operation of the air conditioning unit is determined by the operation intent information in the voice command.

[0079] If the air conditioning unit is in intelligent control mode, after receiving a voice command from the user, the controller acquires environmental information and extracts voiceprint features and operational intent information from the voice command; it identifies the user's target identity category based on the voiceprint features; and it controls the operation of the air conditioning unit based on the target identity category, operational intent information, and environmental information. The operation of the air conditioning unit is jointly determined by the target identity category, operational intent information, and environmental information.

[0080] In some feasible embodiments, users can select different control modes for different identity categories. For example, a smart control mode can be configured to enable users identified as "children." In this mode, when a child attempts to adjust parameters such as temperature and fan speed to preset extreme or unhealthy ranges due to accidental touch or a desire for stimulation, the air conditioning device will not directly execute their voice commands. Instead, it will combine the constraints corresponding to the target identity category and environmental information to constrain and dynamically adjust the operational intent, obtaining operating parameters that meet health and comfort requirements, thereby controlling the operation of the air conditioning device.

[0081] In some feasible embodiments, when the air conditioning device operates according to a combination of parameter values ​​that is not exactly the same as the combination of parameter values ​​in the operation intention information, a prompt message can be displayed through a speaker to inform the user of the control logic and reason.

[0082] In some feasible embodiments, when displaying prompts through a speaker, corresponding titles can be added to the broadcast message according to the user's identity category, such as little friend, little master, little cutie, little treasure, lady, gentleman, master, mistress, young lady, etc.

[0083] In some feasible embodiments, when the target identity category is identified as a child, a cute voice and frequency reduction processing can be enabled when displaying prompt information through a speaker, breaking through the interaction bottleneck of the mechanical response of traditional voice systems and improving the user interaction experience.

[0084] In this embodiment, after receiving a user's voice command, voiceprint features and operation intent information are extracted simultaneously, and external environmental data is collected. Then, the user's target identity category is accurately identified through voiceprint features. The target identity category, operation intent information, and external environmental data are then integrated for comprehensive analysis to capture the user's current actual needs. The system automatically generates and executes a combination of operating parameters that matches the user's current needs, realizing a shift from passively executing user input to actively adapting to actual needs. When the same user issues power-on or simple adjustment commands in different scenarios, the air conditioning device can automatically match a suitable combination of parameters such as temperature, mode, and fan speed based on the user's comfort needs corresponding to their identity category, current specific intent, and real-time environmental status. This significantly reduces the tedious manual adjustment required by users due to historical parameter mismatches. This not only reduces interaction complexity but also allows users to obtain a personalized comfort environment that matches their identity category and real-time environment through simpler voice interaction, thus significantly improving the ease of use and smoothness of operation of the air conditioning device.

[0085] In some embodiments, such as Figure 7 As shown, after extracting voiceprint features from voice commands, the controller is further configured to perform the following steps:

[0086] Step 702: Query the user's target historical preference information based on voiceprint features;

[0087] In the process of controlling the operation of the air conditioning equipment based on the target identity category, operational intent information, and environmental information, the controller is further configured as follows:

[0088] Step 704: Control the operation of the air conditioning equipment based on the target identity category, operation intention information, environmental information, and target historical preference information.

[0089] It should be noted that even individuals within the same identity category may have different comfort preferences. For example, some people are naturally inclined to feel hot but not cold, while others are naturally inclined to feel cold. Their actual needs for temperature environments are clearly different. Adjusting based solely on identity category makes it difficult to capture and adapt to individualized, long-term evolving preferences, resulting in insufficient control precision and low user satisfaction.

[0090] Historical preference information can refer to information used to characterize a user's long-term usage habits and comfort preferences.

[0091] As an example, historical preference information can be a set of data learned or recorded by different users during the historical operation of air conditioning equipment, reflecting their preferences for air conditioning operating parameters. For example, for adult male user A, historical preference information may indicate that he "prefers a temperature 0.5°C higher than the standard adult male model setting."

[0092] In some feasible embodiments, during the operation of the air conditioning equipment, the context information and final execution parameters after each successful voice control interaction can be continuously recorded, and this data can be aggregated on a user-by-user basis. Subsequently, by performing statistical analysis on the interaction data of the same user in similar environmental scenarios, such as calculating the average, mode, distribution range of the set parameters, or establishing simple regression relationships, summary information representing the user's long-term preference pattern can be generated. This summary information constitutes the user's historical preference information.

[0093] As another example, historical preference information can be a collection of historical usage data associated with an individual user. For example, the temperature, wind speed, and wind direction values ​​that user A last set when the outdoor temperature was 35°C.

[0094] In some embodiments, after extracting voiceprint features, the voiceprint features can be compared with a second voiceprint feature model library pre-established and stored within the controller. By calculating the similarity or distance between the voiceprint features and each voiceprint feature model in the second voiceprint feature model library, the target identity identifier that best matches is determined. Subsequently, using the target identity identifier as a query condition, the target historical preference information corresponding to the target identity identifier is queried from the storage path of historical preference information in the memory. Then, the controller takes the target identity category, operation intention information, environmental information, and target historical preference information as input parameters, generates an operation control command according to the preset control logic, and sends the operation control command to the corresponding actuator. The actuator drives its associated physical components to perform actions based on the received operation control command, providing a temperature environment that matches the user's needs.

[0095] The second voiceprint feature model library stores at least one voiceprint feature model corresponding to an individual user. Each individual user's voiceprint feature model can be pre-constructed by collecting the user's voice samples through a voiceprint registration process. For example, a user can use the voiceprint registration function of an application or air conditioner to read aloud a specified text or phrase as prompted. The voice acquisition module collects the user's voice and extracts voiceprint features from it. These voiceprint features are then bound to an identity identifier set by the user or assigned by the system, completing the registration and storage of an independent voiceprint feature model.

[0096] In this embodiment, by querying and integrating users' historical preference information, personalized temperature environments can be provided for different users based on their individual needs through adaptive regulation. For example, even among the same "elderly" category, a slightly higher base temperature can be maintained for users who prefer warmer temperatures, while a wider temperature adjustment range can be provided for users with stronger adaptability. This significantly improves the precision and personalization of control, allowing users to obtain a more comfortable experience that aligns with their long-term habits and reducing the frequency of manual adjustments due to individual differences.

[0097] In some embodiments, such as Figure 8 As shown, the operation intent information includes a set of operation intent parameters; the target historical preference information includes a set of historical preference parameters; during the process of controlling the operation of the air conditioning equipment based on the target identity category, operation intent information, environmental information, and target historical preference information, the controller is further configured to perform the following steps:

[0098] Step 802: Query the target recommended parameter set and target recommended parameter range corresponding to the target identity category, and query the environmental parameter compensation value corresponding to the environmental information;

[0099] Step 804: Filter the set of operation intent parameters according to the target recommended parameter range to obtain the filtered parameter set;

[0100] Step 806: Merge the filter parameter set and the target recommendation parameter set to obtain the fused parameter set;

[0101] Step 808: Aggregate and merge the parameter set, historical preference parameter set, and environmental parameter compensation value to obtain the target parameter set;

[0102] Step 810: Control the air conditioning equipment to operate according to the target parameter set.

[0103] It should be noted that the operation commands issued by users based on instantaneous sensations often have an inherent contradiction with the gradual and stable environment required for the human body to obtain long-term comfort and health. This contradiction can easily lead to frequent secondary adjustments and poor user experience.

[0104] In some scenarios, child users may set excessively low temperatures and strong direct airflow in pursuit of rapid sensory stimulation. If the air conditioner operates directly on these settings, although it may provide temporary sensory stimulation, it is detrimental to health and increases energy consumption.

[0105] In some scenarios, when users enter an indoor environment from a high-temperature outdoor environment, they may set target parameters that deviate too much from their steady-state comfort requirements in pursuit of rapid compensation. As a result, the air conditioner may run at full power, causing discomfort to the user before reaching the set value, thus requiring readjustment.

[0106] In some scenarios, accidental touches or unintentional operations by users can cause unexpected changes in parameters. The air conditioner may switch to an unintended state without recognizing the user's true intentions, and the issue is usually only noticed and corrected after the user feels uncomfortable.

[0107] Traditional control mechanisms execute commands in a one-way, rigid manner, failing to effectively reconcile the immediacy and subjectivity of user commands with the stability and objectivity expected in a comfortable environment. This results in cumbersome operation, discontinuous experience, and a decline in overall satisfaction.

[0108] The operation intent parameter set refers to the set of specific parameter values ​​that the user wants the air conditioning device to be set, which can be directly parsed from the user's current voice command. For example, the {temperature: 26℃, fan speed: strong wind} that can be parsed from the voice command "turn on to 26 degrees with strong wind" is the operation intent parameter set.

[0109] Historical preference parameter set refers to a set of parameter values ​​or parameter tendencies extracted from the historical operating data of a target user, reflecting their long-term usage habits. For example, user A's historical preference parameter set at 30-40℃ could be {temperature: 24℃, wind speed: medium wind}, or {temperature tendency: +0.5℃, wind speed tendency: +1 level}.

[0110] The recommended parameter set can refer to the set of recommended values ​​for each parameter corresponding to an identity category. For example, the recommended parameter set for the "child" identity category could be {recommended temperature: 26℃, recommended wind speed: light breeze}.

[0111] The recommended parameter range can refer to the upper and lower limits of the parameter value that are allowed to fluctuate safely, corresponding to the identity category. For example, the recommended parameter range for temperature corresponding to the "child" identity category can be [24℃, 28℃].

[0112] In some feasible embodiments, a configuration interface for recommended parameter ranges can be set to suit the actual needs of users. This configuration interface can be displayed to the user through the air conditioning device's display interface, speaker or other output modules, or through application software on the user terminal associated with the air conditioning device. In response to the user's modification operation on the configuration interface for the recommended parameter ranges, the recommended parameter ranges corresponding to any one or more identity categories can be modified.

[0113] Environmental parameter compensation values ​​refer to offsets or adjustment coefficients calculated based on current environmental information to fine-tune baseline parameters. For example, when the ambient humidity is higher than 70%, the humidity compensation value might be {temperature compensation value: -0.5℃}, meaning the perceived temperature will be too high, and the set temperature needs to be slightly lowered to maintain the same level of comfort. The environmental parameter compensation values ​​for each parameter can be the same or different. For example, a +5% offset could be set for each parameter; or a +0.5℃ offset could be set for temperature, and a +1 level offset for fan speed; etc.

[0114] In some embodiments, after determining the target identity category, the controller queries the set of recommended parameters corresponding to the target identity category from the preset mapping relationship between identity categories and recommended parameters, using this as the target recommended parameter set and the target recommended parameter range. After obtaining environmental information, the controller queries the compensation value corresponding to the currently obtained environmental information from the preset mapping relationship between environmental information and compensation value, using this as the environmental parameter compensation value for this control process. Subsequently, the controller uses the target recommended parameter range as a health and comfort constraint to filter each operation intent parameter value in the operation intent parameter set parsed from the voice command. After filtering each operation intent parameter value in the operation intent parameter set, a filtered parameter set is obtained in which all parameter values ​​meet the health and comfort requirements of the target identity category. Then, according to the preset fusion rules, the parameter values ​​of each parameter in the filtered parameter set and the target recommended parameter set are fused, and the fused and retained parameter values ​​form a fused parameter set. For each operating parameter of the air conditioning equipment whose value needs to be determined, the controller calculates its fused parameter value in the fused parameter set, its historical preference parameter value in the historical preference parameter set, and its corresponding environmental parameter compensation value according to a preset aggregation algorithm to obtain the final result of the parameter. The final results of all operating parameters are then combined into a target parameter set. Subsequently, the controller generates operating control commands based on the target parameter set and sends these commands to the corresponding actuators. The actuators, based on the received operating control commands, drive their associated physical components to perform actions, providing a temperature environment that matches the target parameter set.

[0115] In some feasible embodiments, the mapping relationship between the preset identity category and the recommendation parameters can be in the form of a data table or a calculation model, etc., and this embodiment does not limit it.

[0116] In some feasible embodiments, the mapping relationship between preset environmental information and compensation values ​​can be in the form of data tables or calculation models, and this embodiment does not limit this.

[0117] In some feasible embodiments, the specific filtering method may include: for each operation intent parameter value in the operation intent parameter set, first determine whether it falls within the corresponding target recommended parameter range. If it is within the corresponding target recommended parameter range, then retain the operation intent parameter value; if it exceeds the corresponding target recommended parameter range, then adjust the operation intent parameter value to the range boundary value that is closest to the operation intent parameter value within the target recommended parameter range. For example, assuming the target recommended parameter range is [24℃, 28℃], if the operation intent parameter value corresponding to the temperature parameter is 24℃, then keep the operation intent parameter value of 24℃ unchanged; if the operation intent parameter value corresponding to the temperature parameter is 20℃, then adjust the operation intent parameter value of 24℃.

[0118] In other feasible embodiments, the filtering method may include: for each operation intent parameter value in the operation intent parameter set, first determine whether it falls within the corresponding target recommended parameter range. If it is within the corresponding target recommended parameter range, the operation intent parameter value is retained; if it exceeds the corresponding target recommended parameter range, the operation intent parameter value is cleared. The operating parameters corresponding to the cleared operation intent parameter values ​​are determined based on the recommended parameter values ​​in the subsequent fusion process with the recommended parameter values ​​in the target recommended parameter set. For example, assuming the target recommended parameter range is [24℃, 28℃], if the operation intent parameter value corresponding to the temperature parameter is 24℃, then the operation intent parameter value of 24℃ is kept unchanged, and the fusion result is determined jointly based on the operation intent parameter value and the corresponding recommended parameter value in the subsequent fusion process; if the operation intent parameter value corresponding to the temperature parameter is 20℃, then the operation intent parameter value is deleted, and the fusion result is determined separately based on the recommended parameter value corresponding to the parameter in the subsequent fusion process.

[0119] In some feasible embodiments, the preset fusion rules may include: preferentially using the filter parameter values ​​from the filter parameter set, where the filter parameter values ​​are the filtered operational intent parameter values, which directly reflect the user's current needs; if a parameter is missing from the filter parameter set, then the recommended parameter value from the target recommended parameter set is used to fill it in. This continues until the fusion parameter values ​​in the fusion parameter set cover every operating parameter of the air conditioning equipment.

[0120] In other feasible embodiments, the preset fusion rules may include: using the target recommended parameter set as a fusion parameter set template, and using each recommended parameter value in the target recommended parameter set as the default value of each parameter in the fusion parameter set template; then querying the filter parameter value corresponding to each slot in the modifiable state from the filter parameter set, and overwriting the corresponding default value with the queried filter parameter value; after the filter parameter value query and overwriting are completed for each slot in the modifiable state, the fusion parameter set is obtained. It is understood that some slots in the fusion parameter set template may be in a prohibited modification state. For these slots in the prohibited modification state, regardless of whether a corresponding filter parameter value exists in the filter parameter set, their default values ​​are not modified, and the recommended parameter values ​​remain unchanged.

[0121] In some feasible embodiments, the aggregation algorithm may include at least one of summation, averaging, weighted summation, and weighted averaging.

[0122] In this embodiment, after identifying the user's target identity category, the operation intention parameter values ​​are first filtered for health and comfort based on the preset target recommendation parameter range for that target identity category. Extreme or unhealthy operation intention parameter values ​​are corrected to a healthy and comfortable range, thereby preventing irrational parameter settings caused by children's missetting, users' momentary impulses, or accidental touches from being directly executed. Next, the filtered operation intention parameter values ​​are fused with the recommended parameter values ​​based on the target identity category to obtain a fused parameter set that reflects both the user's current intention and the group's health and comfort needs. Furthermore, the system further integrates the user's historical preference parameter values ​​and environmental parameter compensation values ​​calculated based on the real-time environment to obtain a target parameter set that incorporates multiple factors such as safety constraints, group recommendations, individual habits, and environmental adaptation while respecting the user's immediate operation intention. This realizes the transformation of control logic from unidirectional execution to intelligent harmonization. When a child issues a low-temperature, high-wind command, the air conditioning equipment can constrain it within a healthy range; when a user enters the room from a high-temperature outdoor environment and issues an excessively low-temperature command, the air conditioning equipment, combining environmental compensation and historical preferences, may provide a more gradual temperature environment that is closer to their steady-state comfort needs. Therefore, in most cases, the parameter set executed by the air conditioner on the first run is close to the user's real and rational long-term comfort needs, which greatly reduces the probability that the user needs to manually adjust it again due to unreasonable parameters or discomfort. This significantly reduces unnecessary repeated adjustment operations and improves the smoothness and satisfaction of the user experience.

[0123] In some embodiments, the filter parameter set includes filter parameter values; such as Figure 9 As shown, during the process of fusing the filter parameter set and the target recommendation parameter set, the controller is further configured to perform the following steps:

[0124] Step 902: Use the target recommended parameter set as the fusion parameter set template, and use the recommended parameter values ​​in the target recommended parameter set as the default values ​​of each parameter in the fusion parameter set template;

[0125] Step 904: If there is a target default value among the default values ​​that has the same parameter type as the filter parameter value, and the target default value is in a modifiable state, then the filter parameter value is used to overwrite the target default value.

[0126] Step 906: If there is a target default value among the default values ​​that has the same parameter type as the filter parameter value, and the target default value is in a state where modification is prohibited, then keep the target default value unchanged.

[0127] It should be noted that when integrating the intended operation parameter values ​​with the recommended parameter values, if a simple weighted average or unconditional replacement strategy is adopted, some key and restrictive recommended parameters (such as parameters related to safety, energy efficiency, or core operating modes) may be easily overwritten, thereby undermining the system's preset safety baseline or energy efficiency strategy and affecting the rationality and reliability of control.

[0128] The fusion parameter set template refers to the set of parameters used as the foundation and framework during the fusion process. During initialization, the fusion parameter set template contains slots for all parameters to be decided and their initial values. For example, the fusion parameter set template could be a data structure containing operational parameter fields such as "temperature," "wind speed," and "mode," with each operational parameter field already assigned a corresponding recommended parameter value as a default value.

[0129] Default values ​​refer to the initial values ​​set for each running parameter in the fusion parameter set template. The default values ​​in the fusion parameter set template are equal to the recommended parameter values ​​in the target recommended parameter set. For example, assuming the target recommended parameter set is {temperature: 24℃, wind speed: low}, the default value for the "temperature" field in the fusion parameter set template could be 24℃, and the default value for the "wind speed" field could be low.

[0130] The parameter type can refer to the category or name to which the parameter value belongs. For example, "temperature", "wind speed", and "operating mode" are different parameter types.

[0131] The target default value can refer to the default value in the fusion parameter set template that is the same as the parameter type of the currently processed filter parameter value. For example, assuming the currently processed filter parameter value is "24℃" and the parameter type is "temperature", then the default value of the "temperature" field in the fusion parameter set template is the target default value corresponding to the currently processed filter parameter value.

[0132] The modification status can refer to an attribute flag pre-set for each parameter or parameter type in the fusion parameter set template, used to indicate whether the parameter value is allowed to be overridden by filtered parameter values ​​during the fusion process. The modification status includes at least a modifiable status and a prohibited modification status. The modifiable status indicates that the parameter is allowed to be overridden by filtered parameter values, and the prohibited modification status indicates that the parameter is not allowed to be overridden by filtered parameter values.

[0133] In some feasible embodiments, the modification status of various parameters can be determined based on the current operating mode and preset safety, energy efficiency, or equipment protection constraints. For example, the temperature parameter cannot be modified in air supply mode; for safety reasons, the maximum operating frequency of the compressor cannot be modified; and so on.

[0134] In some embodiments, after obtaining the target recommended parameter set, the controller makes a complete copy of the target recommended parameter set as the starting framework for subsequent fusion operations, i.e., a fusion parameter set template. The default value of each parameter in the fusion parameter set template is initialized to the corresponding recommended parameter value in the target recommended parameter set. Subsequently, the controller iterates through each filter parameter value in the filter parameter set. For the currently processed filter parameter value, the controller first searches for a default value in the fusion parameter set template that has the same parameter type.

[0135] If a corresponding default value is found, it is used as the target default value, and the modification status of the target default value is further confirmed. If the target default value is marked as modifiable, the controller writes the filter parameter value into the corresponding parameter item in the fusion parameter set template, replacing the original target default value.

[0136] If the target default value is marked as prohibited from modification, the controller will not perform the overwrite operation and will keep the original target default value of the parameter unchanged.

[0137] If no corresponding default value is found, it means that the user does not intend to adjust the value of this parameter. Therefore, the default value, i.e., the recommended parameter value, can be used.

[0138] In this embodiment, a recommended parameter set serves as the baseline template for health and comfort. Based on this, only parameters explicitly instructed by the user's intent and permitted to be modified are updated. For critical parameters related to safety, energy efficiency, or equipment protection, or those contradicting the current operating mode, default values ​​are maintained even if the filtered parameter set provides intended values. This ensures that the fusion process respects the user's explicit operational intent while strictly guaranteeing the system's safety, rationality, and energy efficiency limits, achieving a better balance between meeting user comfort needs and ensuring long-term system stability.

[0139] In some embodiments, such as Figure 10 As shown, in the process of aggregating and fusing the parameter set, the historical preference parameter set, and the environmental parameter compensation value to obtain the target parameter set, the controller is further configured to perform the following steps:

[0140] Step 1002: Perform weighted aggregation of the fusion parameter set and the historical preference parameter set according to the preset weight matrix to obtain the aggregated parameter set;

[0141] Step 1004: Perform parameter compensation on the aggregated parameter set based on the environmental parameter compensation value to obtain the target parameter set.

[0142] It should be noted that if the air conditioning equipment operates solely based on historical preference parameter values, although it can reflect personalized habits, it is easy to become disconnected from the user's current actual operating intentions. Moreover, since the influence of the environment on temperature regulation is ignored, the control results are likely to be inconsistent with expectations. On the other hand, if the air conditioning equipment operates solely based on fused parameter values, it is difficult to fully meet the long-term personalized comfort preferences of different users. Again, since the influence of the environment on temperature regulation is ignored, the control results are likely to be inconsistent with expectations.

[0143] The preset weight matrix can be a predefined data structure used to specify the relative importance of each parameter in the fusion parameter set and the historical preference parameter set during aggregation calculation. As an example, the preset weight matrix can be a two-dimensional table or a mapping table, where its rows or keys can correspond to different parameter types, such as "temperature", "wind speed", "mode", etc., and the columns or values ​​contain the weight coefficients of the fusion parameter value and the historical preference parameter value for that type of parameter.

[0144] In some embodiments, after obtaining the fusion parameter set and the historical preference parameter set, for any parameter that exists in both sets, the corresponding fusion parameter value can be read from the fusion parameter set, the corresponding historical preference parameter value can be read from the historical preference parameter set, and at least one of the corresponding fusion weight value and preference weight value can be read from the weight matrix pre-stored in memory. Then, the read fusion parameter value and historical preference parameter value are weighted and summed according to the fusion weight value and preference weight value, and the weighted sum is used as the aggregated parameter value. After weighted summation calculations are performed on all parameters in both the fusion parameter set and the historical preference parameter set, the resulting aggregated parameter values ​​form the aggregated parameter set. For parameters that exist only in the fusion parameter set or the historical preference parameter set, their parameter values ​​can be directly used as aggregated parameter values.

[0145] Subsequently, parameter compensation calculations are performed on each aggregated parameter value in the aggregated parameter set based on the environmental parameter compensation values. The result of the parameter compensation calculation is used as the target parameter value. After all aggregated parameter values ​​in the aggregated parameter set have completed the parameter compensation calculation, the obtained target parameter values ​​are combined to form the target parameter set.

[0146] In some feasible embodiments, parameter compensation calculations may include addition or multiplication. When the environmental parameter compensation value is an incremental value, the parameter compensation calculation may be addition; when the environmental parameter compensation value is a proportional value or a proportional coefficient, the parameter compensation calculation may be multiplication.

[0147] In some feasible embodiments, the environmental parameter compensation value ΔE env It can be calculated using the following formula:

[0148]

[0149]

[0150]

[0151]

[0152]

[0153] in, This represents the temperature gradient compensation term; This represents the dynamic humidity compensation term; α and β are compensation coefficients, which can be determined in advance based on actual conditions or test results, and this embodiment does not impose any restrictions on them. For real-time ambient temperature, Set the temperature for the user, which is the temperature parameter value in the operation intention information; Real-time ambient humidity; This is the relative humidity deviation coefficient.

[0154] In this embodiment, a weighted aggregation of the fusion parameter set and the historical preference parameter set is performed using a preset weight matrix, achieving a flexible balance between the user's current explicit operational intention and their long-term comfort habits. By introducing environmental parameter compensation values ​​to correct the aggregation result in real time, the final target parameter set can dynamically adapt to environmental changes. In this way, through the organic integration and dynamic optimization of the three dimensions of the user's current intention, personalized preferences, and real-time environmental conditions, the matching degree between the control result and the user's actual needs can be effectively improved, reducing the possibility of expectation discrepancies or repeated adjustments caused by ignoring any dimension, and achieving more accurate, stable, and personalized comfort control.

[0155] In some feasible embodiments, the parameter set can be represented in vector form. For example, assuming a parameter set includes temperature T and wind speed V, the parameter set can be represented as a parameter vector of (T, V); the target parameter vector P can be calculated using the following formula:

[0156]

[0157] Where W represents the identity category U id The corresponding preset weight matrix; ⊙ represents the Hadamard product, indicating element-wise multiplication; For identity category U id The corresponding target recommendation parameter vector; For identity category U id The corresponding historical preference parameter vector; For slot filters set based on the target recommended parameter range; These are compensation values ​​for environmental parameters; The slot key-value pair consists of the operation intent parameter value and the operation intent parameter type; This represents the strategy fusion operator.

[0158] The fusion logic of the policy fusion operator is shown in Table 1:

[0159] Table 1 Fusion Logic Table

[0160]

[0161] In some embodiments, such as Figure 11 As shown, after controlling the air conditioning equipment to operate according to the target parameter set, the controller is further configured to perform the following steps:

[0162] Step 1102: Obtain user feedback information;

[0163] Step 1104: Update the preset weight matrix based on user feedback information.

[0164] It should be noted that although the parameters and strategies used in the control process are based on extensive historical data and test verification, covering most typical scenarios, they are essentially static, group-oriented experience summaries. When users' actual comfort needs deviate from the general model due to individual differences or changes in habits, existing solutions lack a closed-loop mechanism that allows users to provide feedback to directly guide the system to make personalized corrections. Users can only passively accept unsatisfactory operating results or completely abandon automatic functions and revert to manual adjustment, which will weaken users' trust in intelligent functions and their willingness to use them in the long term.

[0165] User feedback information can refer to information monitored by the air conditioning equipment after it has been operating according to the target parameter set, which reflects the user's satisfaction with the current operating status.

[0166] As an example, user feedback can be explicit, such as user comments like "It's too cold" or "Satisfied" given via voice or application software.

[0167] As another example, user feedback can also be implicit feedback, such as indirect behavioral data like a user manually adjusting the air conditioner parameters again within a certain period of time or a user not operating the system for a long time while in a comfortable temperature.

[0168] In some embodiments, after the air conditioning unit operates according to the target parameter set, the controller can actively or passively acquire explicit feedback information from the user through user interaction modules integrated on or associated with the air conditioning unit, such as microphones or application software. Simultaneously, the controller can continuously record user behavior data after the air conditioning unit operates according to the target parameter set through an internal status monitoring module, such as whether the air conditioning parameters were manually modified within a preset time window, which parameters were modified, and the direction of the modification. After summarizing and structuring the collected explicit feedback information and behavioral data, user feedback information reflecting the satisfaction with the control effect can be generated. The controller then adaptively adjusts the preset weight matrix based on the acquired user feedback information.

[0169] In some feasible embodiments, the adaptive adjustment of the preset weight matrix based on the obtained user feedback information includes: if the user feedback information indicates that the user is satisfied with the temperature control result, for example, no further action is taken or a positive evaluation is given, the preset weight matrix can be kept unchanged; conversely, if the user feedback information indicates that the user is dissatisfied with the temperature control result, at least some weight values ​​in the preset weight matrix can be fine-tuned in a direction that satisfies the user, for example, if the user quickly and manually increases the temperature, it indicates that the temperature given by the system may be too low, and the weight values ​​corresponding to the temperature parameter can be further fine-tuned in a direction that reduces the temperature. The updated weight matrix will be stored for subsequent aggregation decisions.

[0170] Parameter fine-tuning can be done by increasing or decreasing the step size by a fixed amount, or by using gradient descent algorithms.

[0171] In this embodiment, an adaptive closed loop that continuously optimizes based on the user's personalized experience is constructed by introducing a weight matrix update mechanism based on user feedback. When a user is persistently dissatisfied with the operating results, the air conditioning equipment can automatically identify the direction of the deviation through feedback information and make targeted fine-tuning of the parameter weights in the weight matrix that cause the deviation. This allows the preset general model to be progressively and directionally optimized based on the actual experience of individual users, gradually reducing the systematic deviation between the system output and the user's actual comfort needs. This transforms a static, general control strategy into a dynamic, personalized comfort solution. This not only improves the long-term adaptability of the control, but more importantly, it enables user feedback to effectively influence and improve the subsequent experience, effectively enhancing the user's sense of control, trust, and stickiness with the intelligent functions, ensuring the long-term effectiveness of the function and user satisfaction.

[0172] In some embodiments, user feedback information includes feedback adjustment operation information for any parameter in the target parameter set within a preset feedback time range; such as... Figure 12As shown, during the process of updating the preset weight matrix based on user feedback, the controller is further configured to perform the following steps:

[0173] Step 1202: Extract the adjustment amplitude value from the feedback adjustment operation information;

[0174] Step 1204: Update the preset weight matrix based on the adjustment amplitude value using an online learning algorithm.

[0175] It should be noted that without quantitative analysis of feedback behavior and gradient backpropagation, it is difficult to accurately assess the contribution of different parameter sources to the current decision bias, resulting in a coarse weight update process, which usually only allows for equal-step or heuristic adjustments. This inefficient update method makes the air conditioning equipment converge slowly in the process of adapting to users' personalized preferences, requiring multiple interactions to approximate the user's actual needs, thus affecting the immediate improvement of user experience and the perceived effectiveness of the function.

[0176] The preset feedback time range refers to a specific period of time during which the air conditioning equipment continuously monitors user adjustments after operating according to the target parameter set. For example, the preset feedback time range can be set to "within 10 minutes after operating according to the target parameter set." Only adjustments occurring within this time range are considered feedback to the current control.

[0177] Feedback adjustment operation information refers to the information recorded when a user manually changes the air conditioning parameters within a preset feedback time range. This feedback adjustment operation information may include, but is not limited to, the type of parameter being adjusted, the direction of adjustment, and the values ​​before and after the adjustment.

[0178] The adjustment range value can refer to a quantified value that represents the actual amount of adjustment by the user, which can be read from or calculated from the feedback adjustment operation information. For example, if the user manually adjusts the temperature from the system setting of 24℃ to 26℃, the adjustment range value can be quantified as +2℃ (absolute difference) or +8.3% (relative rate of change).

[0179] Online learning algorithms are machine learning algorithms that can update a preset weight matrix in real time and incrementally by adjusting the magnitude value during system operation. These include stochastic gradient descent, online gradient descent, or their variants.

[0180] In some embodiments, after receiving feedback adjustment operation information, the controller first determines which parameter in the target parameter set the feedback operation targets. Then, the controller reads the original setpoint of the parameter under the feedback condition in the target parameter set, and the new setpoint after user feedback adjustment, calculates the difference or relative change ratio between the two, and uses the calculation result as the adjustment amplitude value. The sign of the adjustment amplitude value can indicate the user's adjustment direction, and its magnitude can indicate the degree of user dissatisfaction with the current control. Subsequently, the controller uses the adjustment amplitude value and its associated parameter type as input to drive a preset online learning algorithm, treating the adjustment amplitude value as a loss signal or gradient signal, and fine-tuning the weight values ​​in the preset weight matrix corresponding to the parameter type.

[0181] In some feasible implementations, the weight value of a user with identity category k is fine-tuned. It can be represented as:

[0182]

[0183] in, For users with identity category k, the weight values ​​before and after fine-tuning are given. The loss gradient at step t is represented by , which is determined based on the adjustment magnitude, which is based on the feedback adjustment information. and the corresponding target parameter values Sure; This represents the adaptive learning rate parameter corresponding to user with identity category k. These are the L1 and L2 regularization parameters, respectively.

[0184] In this embodiment, by quantifying the adjustment magnitude value of user feedback and driving the online learning algorithm, the algorithm uses the adjustment magnitude value to more accurately evaluate the shortcomings of the current weight allocation and perform targeted and step-wise fine-tuning updates. This makes the adjustment of the weight matrix no longer a blind trial and error, but a targeted gradient optimization. This effectively improves the efficiency and accuracy of the air conditioning equipment's adaptive learning, allowing users to experience effective improvements with fewer interactions, accelerating the personalized adaptation process, and thus enhancing the user experience and trust in the intelligent functions.

[0185] In some embodiments, voice commands include power-on commands; such as Figure 13 As shown, during the process of querying a user's target historical preference information based on voiceprint features, the controller is further configured to perform the following steps:

[0186] Step 1302: If the smart power-on function is enabled, query the user's target historical preference information based on voiceprint characteristics;

[0187] In the process of controlling the operation of the air conditioning equipment based on the target identity category, operation intent information, and environmental information, the controller is further configured to perform the following steps:

[0188] Step 1304: If the target's historical preference information that matches the voiceprint features is found, control the operation of the air conditioning equipment according to the target's identity category, operation intention information, environmental information and target's historical preference information;

[0189] Step 1306: If no target historical preference information matching the voiceprint feature is found, control the operation of the air conditioning equipment based on the target identity category, operation intention information and environmental information.

[0190] It's important to note that in traditional technology, adjusting air conditioning parameters relies heavily on user input. If no new settings are entered upon startup, the air conditioner will use the saved operating parameter combination from the last shutdown. However, when usage scenarios change, users often need to reset multiple parameters such as temperature, operating mode, fan speed, and airflow direction to achieve a suitable indoor environment. This frequent and complex operation significantly reduces the ease of use of air conditioning equipment.

[0191] The power-on command can refer to the voice command used to start the air conditioning equipment.

[0192] As an example, a power-on command can be used simply to indicate that the device is powered on, such as "turn on the air conditioner" or "air conditioner is on".

[0193] As another example, the power-on command can be a multi-intent command, that is, in addition to containing information indicating power-on, it can also contain specific parameter values, such as "turn on the air conditioner, 20℃", or directly say "run at 20℃" when the air conditioner is off.

[0194] The intelligent start-up function refers to the operating mode of an air conditioning unit when responding to a start-up command, which integrates information such as user identity, operating intent, and environmental information. This can be achieved through the air conditioning unit's display interface, speakers, or other output modules, or through application software on a user terminal associated with the air conditioning unit. The system displays the on / off elements of the intelligent start-up function to the user, responding to the user's triggering of these elements, and controls the air conditioning unit to turn the intelligent start-up function on or off according to the user's selection.

[0195] In some embodiments, when the controller detects a power-on command, it first checks the on / off status of the smart power-on function in the air conditioning unit. If the smart power-on function is enabled, the controller initiates the subsequent smart control process, executing the step of querying the user's target historical preference information based on voiceprint characteristics. If the smart power-on function is disabled, the controller does not initiate the subsequent smart control process and may run according to the parameters recorded during the last power-off, or according to the parameters input by the user, etc. This embodiment does not impose any limitations on this.

[0196] After querying the user's target historical preference information based on voiceprint features, for users who have not powered on the device for the first time or have not entered voice commands for the first time, the controller can query the target historical preference information that matches the voiceprint features. Therefore, it can execute the steps of controlling the operation of the air conditioning equipment based on the target identity category, operation intention information, environmental information, and target historical preference information. For users who are powering on the device for the first time or entering voice commands for the first time, the controller cannot query the target historical preference information that matches the voiceprint features. Therefore, it can execute the steps of controlling the operation of the air conditioning equipment based on the target identity category, operation intention information, and environmental information.

[0197] In this embodiment, when a user issues only the "power on" voice command, the air conditioning device can first identify the user's target identity category and obtain environmental information through voiceprint features. For users who have not powered on for the first time or entered a voice command for the first time, the air conditioning device will further query the target historical preference information corresponding to their target identity category, and automatically generate a set of startup parameters that integrates health and comfort constraints, personalized habits, and real-time environmental adaptability by combining the identity category, environmental information, and historical preferences. For users who are powering on for the first time or entering a voice command for the first time, since the corresponding target historical preference information cannot be found, the air conditioning device generates a set of startup parameters that conforms to the group's comfort consensus based on their identity category and environmental information. In this way, when the air conditioning device is powered on, it can automatically avoid using parameter values ​​that may no longer be applicable when it was last powered off, and instead provide a more comfortable starting point that fits the current user identity, environmental conditions, and personal preferences. Users no longer need to manually reset multiple parameters one by one due to changes in season, scenario, or time. This significantly simplifies the power-on operation and effectively improves the ease of use of the air conditioning device.

[0198] In some embodiments, such as Figure 14 As shown, in the process of identifying the target identity category of a user based on voiceprint features, the controller is further configured to perform the following steps:

[0199] Step 1402: Identify the user's multi-dimensional physiological characteristics based on voiceprint features;

[0200] Step 1404: Match the corresponding target identity category based on each physiological characteristic.

[0201] It should be noted that different user groups may have significant physiological differences in their thermal comfort needs. If these physiological characteristics cannot be identified from the voice, it is difficult to build a refined comfort model that fits the user's physiological basis by relying solely on the uniqueness or broad classification of voiceprints. As a result, the accuracy of personalized control is still limited to the level of individual user identification, and fails to delve into the intrinsic dimension of their physiological comfort needs.

[0202] Physiological characteristics can refer to inherent or relatively stable physical attributes and states that can be indirectly reflected from the user's voice signals and are related to their thermal comfort needs, including but not limited to gender, age, and health status.

[0203] In some embodiments, after the voiceprint features are extracted, the controller can input the voiceprint features into multiple trained physiological feature recognition models. Then, the recognition results are used as a set of input conditions to query a pre-stored identity category mapping table or rule engine to obtain the target identity category that matches the input conditions.

[0204] In some feasible embodiments, the physiological feature recognition model may include a gender recognition model, an age estimation model, and a health status recognition model. The gender recognition model analyzes gender-related acoustic parameters in the voiceprint features, such as average fundamental frequency and spectral centroid, and outputs a probability or category label for "male" or "female." The age estimation model analyzes age-related acoustic parameters in the voiceprint features, such as perturbations and harmonic noise ratio, and outputs an estimated age (e.g., 12 years old) or an age group label (e.g., "child," "youth," "elderly," etc.). The health status recognition model analyzes acoustic parameters related to physical condition in the voiceprint features, such as speech rate, cough characteristics, and uniformity of energy distribution, and outputs a health status label, such as "normal," "fatigued," or "cold."

[0205] In some feasible embodiments, the user's identity category U id It can be represented as:

[0206] U id =sex * 1 + age * 2

[0207] In this context, sex represents gender, with 0 indicating male and 1 indicating female; age represents age, with 0 indicating child, 1 indicating adult, and 2 indicating old age.

[0208] If U id =0, then it represents a boy; if U id =1, then it represents a girl; if U id =2, then it represents an adult male; if U id =3, then it represents an adult female; if U id=4, then it represents an elderly male; if U id =5 indicates an elderly woman.

[0209] In this embodiment, testing and research revealed that a user's gender, age, and health status are key physiological variables determining their thermal comfort needs. Based on this, this embodiment pre-constructs an identity category system based on these three physiological attributes, directly linking the category division to the physiological causes of thermal comfort. In practical applications, by extracting and identifying these three physiological attributes from voiceprint features and matching them to the target identity category, a deep characterization of the user's physiological comfort needs can be achieved. For example, these three physiological attributes can differentiate between "young men" and "elderly women," or "healthy adults" and "infirm individuals," allowing for the adaptation of control strategies that match their respective physiological metabolic levels, thermoregulation capabilities, and comfort sensitivities. This deepens personalized control from superficial user identification to physiological need modeling, significantly improving the scientific rigor and accuracy of personalized control, and providing users with a more physiologically aligned and personalized health and comfort experience.

[0210] In some embodiments, a control method for an air conditioning device is provided, such as Figure 15 As shown, the method includes:

[0211] Step 1502: In response to the user's voice command, obtain environmental information and extract voiceprint features and operation intent information from the voice command;

[0212] Step 1504: Identify the user's target identity category based on voiceprint features;

[0213] Step 1506: Control the operation of the air conditioning equipment based on the target identity category, operation intention information, and environmental information.

[0214] In some embodiments, after extracting voiceprint features from voice commands, the method further includes: querying the user's target historical preference information based on the voiceprint features; and controlling the operation of the air conditioning equipment based on the target identity category, operation intent information, and environmental information, including: controlling the operation of the air conditioning equipment based on the target identity category, operation intent information, environmental information, and target historical preference information.

[0215] In some embodiments, the operation intent information includes an operation intent parameter set; the target historical preference information includes a historical preference parameter set; controlling the operation of the air conditioning equipment according to the target identity category, operation intent information, environmental information, and target historical preference information includes: querying the target recommended parameter set and target recommended parameter range corresponding to the target identity category, and querying the environmental parameter compensation value corresponding to the environmental information; filtering the operation intent parameter set according to the target recommended parameter range to obtain a filtered parameter set; fusing the filtered parameter set and the target recommended parameter set to obtain a fused parameter set; aggregating the fused parameter set, the historical preference parameter set, and the environmental parameter compensation value to obtain a target parameter set; and controlling the air conditioning equipment to operate according to the target parameter set.

[0216] In some embodiments, the filter parameter set includes filter parameter values; fusing the filter parameter set and the target recommended parameter set includes: using the target recommended parameter set as a fusion parameter set template, and using each recommended parameter value in the target recommended parameter set as the default value of each parameter in the fusion parameter set template; if there is a target default value among the default values ​​that has the same parameter type as the filter parameter value, and the target default value is in a modifiable state, then the filter parameter value is used to overwrite the target default value; if there is a target default value among the default values ​​that has the same parameter type as the filter parameter value, and the target default value is in a non-modifiable state, then the target default value is kept unchanged.

[0217] In some embodiments, aggregating the fusion parameter set, the historical preference parameter set, and the environmental parameter compensation value to obtain the target parameter set includes: weighting and aggregating the fusion parameter set and the historical preference parameter set according to a preset weight matrix to obtain an aggregated parameter set; and performing parameter compensation on the aggregated parameter set according to the environmental parameter compensation value to obtain the target parameter set.

[0218] In some embodiments, after controlling the air conditioning equipment to operate according to the target parameter set, the method further includes: obtaining user feedback information; and updating the preset weight matrix based on the user feedback information.

[0219] In some embodiments, user feedback information includes feedback adjustment operation information for any parameter in the target parameter set within a preset feedback time range; updating the preset weight matrix based on user feedback information includes: extracting adjustment amplitude values ​​from the feedback adjustment operation information; and updating the preset weight matrix based on the adjustment amplitude values ​​using an online learning algorithm.

[0220] In some embodiments, the voice command includes a power-on command; querying the user's target historical preference information based on voiceprint features includes: querying the user's target historical preference information based on voiceprint features when the smart power-on function is enabled; during the process of controlling the operation of the air conditioning equipment based on target identity category, operation intent information, and environmental information, the controller is further configured to: control the operation of the air conditioning equipment based on the target identity category, operation intent information, environmental information, and target historical preference information when target historical preference information matching the voiceprint features is found; and control the operation of the air conditioning equipment based on the target identity category, operation intent information, and environmental information when no target historical preference information matching the voiceprint features is found.

[0221] In some embodiments, identifying a user's target identity category based on voiceprint features includes: identifying multi-dimensional physiological characteristics of the user based on voiceprint features; and matching the corresponding target identity category based on each physiological characteristic.

[0222] In this embodiment, after receiving a user's voice command, voiceprint features and operation intent information are extracted simultaneously, and external environmental data is collected. Then, the user's target identity category is accurately identified through voiceprint features. The target identity category, operation intent information, and external environmental data are then integrated for comprehensive analysis to capture the user's current actual needs. The system automatically generates and executes a combination of operating parameters that matches the user's current needs, realizing a shift from passively executing user input to actively adapting to actual needs. When the same user issues power-on or simple adjustment commands in different scenarios, the air conditioning device can automatically match a suitable combination of parameters such as temperature, mode, and fan speed based on the user's comfort needs corresponding to their identity category, current specific intent, and real-time environmental status. This significantly reduces the tedious manual adjustment required by users due to historical parameter mismatches. This not only reduces interaction complexity but also allows users to obtain a personalized comfort environment that matches their identity category and real-time environment through simpler voice interaction, thus significantly improving the ease of use and smoothness of operation of the air conditioning device.

[0223] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the methods of the above embodiments.

[0224] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the methods of the above embodiments.

[0225] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods of the above embodiments.

[0226] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0227] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0228] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0229] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An air conditioning device, characterized in that, include: The refrigerant circulation loop, including the compressor, condenser, expansion valve and evaporator, is used to realize the refrigeration or heating cycle of the refrigerant; The sound acquisition module is configured to receive voice commands; The environmental information collection module is configured to collect environmental information; The controller is configured as follows: After receiving a voice command input from the user, environmental information is obtained, and voiceprint features and operation intent information are extracted from the voice command. The user's target identity category is identified based on the voiceprint features; The air conditioning equipment is controlled to operate based on the target identity category, the operation intent information, and the environmental information.

2. The air conditioning equipment according to claim 1, characterized in that, After extracting voiceprint features from the voice command, the controller is further configured to: Based on the voiceprint features, query the user's target historical preference information; In the process of controlling the operation of the air conditioning equipment according to the target identity category, the operation intent information, and the environmental information, the controller is further configured to: The operation of the air conditioning equipment is controlled based on the target identity category, the operation intent information, the environmental information, and the target historical preference information.

3. The air conditioning equipment according to claim 2, characterized in that, The operation intent information includes an operation intent parameter set; the target historical preference information includes a historical preference parameter set; during the process of controlling the operation of the air conditioning equipment according to the target identity category, the operation intent information, the environmental information, and the target historical preference information, the controller is further configured to: Query the target recommendation parameter set and target recommendation parameter range corresponding to the target identity category, and query the environmental parameter compensation value corresponding to the environmental information; The set of operation intent parameters is filtered according to the target recommended parameter range to obtain a filtered parameter set; The filtering parameter set and the target recommendation parameter set are merged to obtain the fused parameter set; By aggregating the fusion parameter set, the historical preference parameter set, and the environmental parameter compensation value, a target parameter set is obtained; Control the air conditioning equipment to operate according to the target parameter set.

4. The air conditioning equipment according to claim 3, characterized in that, The filtering parameter set includes filtering parameter values; during the process of fusing the filtering parameter set and the target recommendation parameter set, the controller is further configured to: The target recommendation parameter set is used as the fusion parameter set template, and the recommendation parameter values ​​in the target recommendation parameter set are used as the default values ​​of each parameter in the fusion parameter set template. If there is a target default value among the default values ​​that has the same parameter type as the filter parameter value, and the target default value is in a modifiable state, then the filter parameter value is used to overwrite the target default value; If any of the default values ​​has a target default value of the same type as the filter parameter value, and the target default value is in a state where modification is prohibited, then the target default value remains unchanged.

5. The air conditioning equipment according to claim 3, characterized in that, In the process of aggregating the fusion parameter set, the historical preference parameter set, and the environmental parameter compensation value to obtain the target parameter set, the controller is further configured to: The fusion parameter set and the historical preference parameter set are weighted and aggregated according to a preset weight matrix to obtain an aggregated parameter set; The aggregated parameter set is compensated based on the environmental parameter compensation values ​​to obtain the target parameter set.

6. The air conditioning equipment according to claim 5, characterized in that, After the air conditioning equipment is controlled to operate according to the target parameter set, the controller is further configured to: Obtain user feedback information; The preset weight matrix is ​​updated based on the user feedback information.

7. The air conditioning equipment according to claim 6, characterized in that, The user feedback information includes feedback adjustment operation information for any parameter in the target parameter set within a preset feedback time range; during the process of updating the preset weight matrix based on the user feedback information, the controller is further configured to: Extract the adjustment amplitude value from the feedback adjustment operation information; The preset weight matrix is ​​updated based on the adjustment amplitude value using an online learning algorithm.

8. The air conditioning equipment according to claim 2, characterized in that, The voice commands include a power-on command; during the process of querying the user's target historical preference information based on the voiceprint features, the controller is further configured to: With the smart power-on function enabled, the user's target historical preference information is queried based on the voiceprint characteristics; In the process of controlling the operation of the air conditioning equipment according to the target identity category, the operation intent information, and the environmental information, the controller is further configured to: If target historical preference information matching the voiceprint features is found, the air conditioning equipment is controlled to operate based on the target identity category, the operation intention information, the environmental information, and the target historical preference information. If no target historical preference information matching the voiceprint features is found, the air conditioning equipment is controlled to operate based on the target identity category, the operation intention information, and the environmental information.

9. The air conditioning equipment according to any one of claims 1 to 8, characterized in that, In the process of identifying the user's target identity category based on the voiceprint features, the controller is further configured to: The user's multi-dimensional physiological characteristics are identified based on the voiceprint features; Match the corresponding target identity category based on each of the stated physiological characteristics.

10. A control method for an air conditioning device, characterized in that, The method includes: In response to a user's voice command, environmental information is obtained, and voiceprint features and operation intent information are extracted from the voice command. The user's target identity category is identified based on the voiceprint features; The air conditioning equipment is controlled to operate based on the target identity category, the operation intent information, and the environmental information.

Images