Control method, device and equipment of water heater and computer storage medium

The control method uses a large language model to simplify water heater operations by processing user interactions and executing prompts, addressing complexity and accidental touches, enhancing user experience.

CN119063259APending Publication Date: 2024-12-03QINGDAO ECONOMIC AND TECHNOLOGICAL DEVELOPMENT ZONE HAIER WATER HEATER CO LTD +1
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411223887.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional water heater control methods are cumbersome, requiring users to switch between multiple entrances and lack intelligent interaction, leading to accidental touches and misoperations.

Method used

A control method utilizing a large language model to process user interactions, output operation prompts, and execute operations based on confirmation, reducing the need for complex switching and enhancing user experience.

Benefits of technology

Simplifies operations, reduces accidental touches, and improves efficiency by providing intelligent and humanized interaction, eliminating the need to remember specific functional locations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119063259A_ABST
    Figure CN119063259A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of intelligent household appliances, and particularly relates to a control method, device and equipment of a water heater and a computer storage medium. According to the method, in response to interaction information which is sent by a client and used for operating the water heater, an operation prompt corresponding to the interaction information is output through a large language model, so that whether first operation corresponding to the interaction information is executed or not is determined; and when the confirmation information for the first operation sent by the client is responded, the first operation is triggered and executed, so that the operation process of the equipment is simplified. According to the method, intelligent interaction prompt is realized by utilizing the large language model, so that the operation complexity and the multi-entrance switching requirement are reduced, the operation efficiency of equipment is improved, the burden of a user for memorizing a specific function position is reduced, misoperation caused by the operation complexity of the user is avoided, and the use experience of the user is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of smart home appliances, and specifically relates to a control method, device, equipment and computer storage medium for a water heater. Background Art

[0002] Traditional water heater APPs have complicated functions. Users need to switch back and forth between multiple entrances when adjusting functions or checking usage, which makes the operation complicated. In addition, the interaction method of traditional water heater APPs mainly relies on menus and buttons. Users need to memorize the location of specific functions, and lack intelligent and humanized interactive experience.

[0003] Therefore, in the specific use of the water heater APP, the cumbersome switching operation makes it easy for users to make accidental touches or misoperations, which causes the water heater to execute wrong instructions and reduce the efficiency of the water heater. Summary of the invention

[0004] The present application provides a control method, device, equipment and computer storage medium for a water heater, which are used to solve the technical problem that the APP switching operation of a traditional water heater is complicated and prone to accidental touches or misoperations.

[0005] In a first aspect, the present application provides a method for controlling a water heater, comprising: in response to interaction information sent by a client for operating the water heater, outputting an operation prompt corresponding to the interaction information through a large language model, wherein the interaction information is interaction voice or interaction text, and the operation prompt is used to determine whether to perform a first operation corresponding to the interaction information;

[0006] In response to confirmation information for the first operation sent by the client, execution of the first operation is triggered.

[0007] Optionally, in response to the interactive information sent by the client for operating the water heater, outputting an operation prompt corresponding to the interactive information through the large language model includes:

[0008] In response to the interaction information sent by the client, outputting a question corresponding to the interaction through the large language model;

[0009] In response to the answer to the question sent by the client, an operation prompt corresponding to the interaction is output through the large language model.

[0010] Optionally, in response to confirmation information sent by the client for the first operation, triggering execution of the first operation includes:

[0011] If the first operation is a control operation, querying a control function parameter corresponding to the first operation from a preset control function table, and sending a control instruction corresponding to the first operation to the water heater based on the control function parameter;

[0012] If the first operation is a query operation, query parameters corresponding to the first operation are queried from a preset query function table, and query results corresponding to the first operation are queried from a preset database based on the query parameters, and the query results are sent to the client.

[0013] Optionally, in response to confirmation information sent by the client for the operation prompt, triggering execution of the first operation includes:

[0014] If the first operation is a control operation, querying in an operation trajectory prediction table whether there is a first operation trajectory starting with the first operation, wherein the operation trajectory prediction table includes one or more operation trajectories for the water heater, and the operation trajectory includes multiple operations executed in sequence;

[0015] If the first operation trajectory exists, sending a prediction prompt to the client, where the prediction prompt is used to determine whether to execute the first operation trajectory;

[0016] In response to confirmation information for the first operation trajectory sent by the client, execution of the first operation trajectory is triggered.

[0017] Optionally, the method further includes:

[0018] In response to the denial information regarding the first operation track sent by the client, the first operation is triggered to be executed, or if the first operation track does not exist, the first operation is triggered to be executed.

[0019] Optionally, the method further includes:

[0020] Get the knowledge question and answer corpus related to water heaters;

[0021] An initial large language model is obtained by training based on the knowledge question and answer corpus, and the initial large prediction model is used to output a corresponding answer based on the input information;

[0022] The output of the initial large language model is scored using a reward model, and parameters of the initial large language model are adjusted based on the score to obtain the large language model.

[0023] Optionally, the method further includes:

[0024] Inputting the interaction information sample into the initial large language model to obtain an answer sample output by the initial large language model;

[0025] The interactive information sample, the answer sample, and the score annotated for the answer sample are used as training samples to train and obtain the reward model.

[0026] In a second aspect, the present application provides a control device for a water heater, comprising:

[0027] The response module is used to respond to the interactive information sent by the client for operating the water heater, wherein the interactive information is interactive voice or interactive text.

[0028] A processing module is used to output an operation prompt corresponding to the interaction information through a large language model, wherein the operation prompt is used to determine whether to perform a first operation corresponding to the interaction information.

[0029] The processing module is further configured to trigger execution of the first operation in response to confirmation information for the first operation sent by the client.

[0030] Optionally, the response module is further configured to respond to the interaction information sent by the client.

[0031] The processing module is further used to output questions corresponding to the interaction through the large language model.

[0032] The response module is further configured to respond to an answer to the question sent by the client.

[0033] The processing module is further configured to output an operation prompt corresponding to the interaction through the large language model.

[0034] Optionally, if the first operation is a control operation, the processing module is also used to query the control function parameters corresponding to the first operation from a preset control function table, and based on the control function parameters, send a control instruction corresponding to the first operation to the water heater.

[0035] If the first operation is a query operation, the processing module is further used to query the query parameters corresponding to the first operation from a preset query function table, and query the query results corresponding to the first operation from a preset database based on the query parameters, and send the query results to the client.

[0036] Optionally, if the first operation is a control operation, the processing module is also used to query in the operation trajectory prediction table whether there is a first operation trajectory starting with the first operation, wherein the operation trajectory prediction table includes one or more operation trajectories for the water heater, and the operation trajectory includes multiple operations performed in sequence.

[0037] If the first operation trajectory exists, the processing module is further used to send a prediction prompt to the client, and the prediction prompt is used to determine whether to execute the first operation trajectory.

[0038] The response module is further configured to respond to confirmation information sent by the client regarding the first operation track.

[0039] The processing module is further used to trigger the execution of the first operation trajectory.

[0040] Optionally, the response module is further configured to respond to a denial message regarding the first operation track sent by the client.

[0041] The processing module is further used to trigger the execution of the first operation, or if the first operation track does not exist, the processing module is further used to trigger the execution of the first operation.

[0042] Optionally, the control device of the water heater further includes: an acquisition module.

[0043] The acquisition module is used to obtain the knowledge question and answer corpus related to water heaters.

[0044] The processing module is also used to obtain an initial large language model based on the knowledge question and answer corpus training, and the initial large prediction model is used to output a corresponding answer based on the input information.

[0045] The processing module is further configured to score the output of the initial large language model using a reward model, and adjust parameters of the initial large language model based on the score to obtain the large language model.

[0046] Optionally, the processing module is further used to input the interaction information sample into the initial large language model to obtain an answer sample output by the initial large language model.

[0047] The processing module is further used to use the interaction information sample, the answer sample, and the score annotated for the answer sample as training samples to train and obtain the reward model.

[0048] In a third aspect, the present application provides a control device for a water heater, comprising:

[0049] Memory;

[0050] processor;

[0051] Wherein, the memory stores computer-executable instructions;

[0052] The processor executes the computer-executable instructions stored in the memory to implement the control method of the water heater as described in the first aspect and various possible implementation methods of the first aspect.

[0053] In a fourth aspect, the present application provides a computer storage medium having computer execution instructions stored thereon, wherein the computer execution instructions are executed by a processor to implement the water heater control method as described in the first aspect and various possible implementations of the first aspect.

[0054] In a fifth aspect, the present application provides a program product, including a computer program, which implements the control method of the water heater as described above when executed by a processor.

[0055] The control method of the water heater provided in the present application obtains a knowledge question and answer corpus related to the water heater, obtains an initial large language model based on the knowledge question and answer corpus training, then uses a reward model to score the output of the initial large language model, and adjusts the parameters of the initial large language model based on the score to obtain the large language model; in response to the interaction information sent by the client, the large language model outputs a question corresponding to the interaction, and in response to the answer to the question sent by the client, the large language model outputs an operation prompt corresponding to the interaction, so as to determine whether to execute the first operation corresponding to the interaction information; if the first operation is a control operation, the control function parameters corresponding to the first operation are queried from a preset control function table, and based on the control function parameters, a control instruction corresponding to the first operation is sent to the water heater; if the first operation is a query operation, the query parameters corresponding to the first operation are queried from a preset query function table, and based on the query parameters, a query result corresponding to the first operation is queried from a preset database, and the query result is sent to the client. This method uses a large language model to implement intelligent interactive prompts, reduces operational complexity and the need for multi-entry switching, improves the operating efficiency of the device, reduces the burden on users to remember specific functional locations, avoids user misoperations due to operational complexity, simplifies the device operation process, and improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0057] Figure 1 It is a schematic diagram of a scenario of a water heater control method provided by the present application;

[0058] Figure 2 This is a schematic diagram of the process of the water heater control method provided by this application Figure 1 ;

[0059] Figure 3 This is a schematic diagram of the process of the water heater control method provided by this application Figure 2 ;

[0060] Figure 4 It is a structural schematic diagram of a control device for a water heater provided in the present application;

[0061] Figure 5 It is a structural schematic diagram of the control device of the water heater provided in this application.

[0062] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0064] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein, for example.

[0065] In the embodiments of the present application, the words "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0066] With the rapid development of smart home technology, more and more home appliances have begun to have intelligent control functions. As one of the common household appliances, the demand for intelligent water heaters is also increasing.

[0067] The traditional water heater control method mainly relies on physical buttons or simple mobile phone APP operations. However, the functions of the traditional water heater APP are cumbersome. Users need to switch back and forth between multiple entrances when adjusting functions or querying usage, which makes the operation complicated. Moreover, the interaction method of the traditional water heater APP mainly relies on menus and buttons. Users need to memorize the location of specific functions and lack intelligent and humanized interactive experience. Therefore, during the operation of the water heater APP, the water heater may execute wrong instructions due to accidental touch, which poses a safety hazard and inconvenience.

[0068] In view of the above problems, the present application provides a control method for a water heater.

[0069] First, the implementation scenarios involved in this application are described.

[0070] Figure 1 Schematic diagram of the water heater control method provided by this application. Figure 1 As shown, the intelligent interactive device 101 is connected to the server 102 in communication, and the server 102 is connected to the intelligent home appliance 103 in communication; the intelligent interactive device 101 can interact with the user, determine the user's usage demand at that time, and upload the usage demand to the server 102; the server 102 inputs the usage demand received at that time into the large language model set by itself, and interacts with the user again according to the result output by the large language model, wherein the result output by the large language model includes: control results and non-control results, and the control results need to control other intelligent home appliances to execute, and the non-control results can be directly output to the user by the large language model without performing additional operations; if the result output by the large language model at that time is a control result, then based on the control result, the intelligent home appliance 103 is controlled to perform the corresponding operation. The intelligent interactive device 101 involved in this application can be, for example, a smart mobile terminal provided with a water heater APP, the server 102 can be, for example, a cloud server, and the intelligent home appliance 103 can be, for example, an intelligent water heater or a hot water providing device.

[0071] The control method of the water heater provided in the present application obtains a knowledge question and answer corpus related to the water heater, trains an initial large language model based on the corpus, and performs scoring and parameter adjustment through a reward model, so that the trained large language model can better understand and respond to the user's natural language input; the interactive information (voice or text) sent by the client is processed by the large language model, and the operation prompt corresponding to the interactive information is directly output, and the operation is divided into control and query categories, and the corresponding parameters or results are respectively queried from a preset function table or database, and control instructions or query results are sent, so that the user does not need to switch between different function entrances, reducing the complexity of operation. The method uses a large language model to realize intelligent interactive prompts, reduces the complex operation of users switching between multiple entrances, improves the operating efficiency of the equipment, and uses a large language model to realize intelligent and humanized interactive experience, reduces the burden of users remembering specific function locations, and improves the user's experience.

[0072] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0073] Figure 2 Schematic diagram of the process of the water heater control method provided in the embodiment of the present application Figure 1 The execution subject of this embodiment may be, for example, a cloud server equipped with a large language model. Figure 2 As shown, the control method of the water heater provided in this embodiment includes:

[0074] S201: In response to interaction information for operating a water heater sent by a client, outputting an operation prompt corresponding to the interaction information through a large language model.

[0075] The interactive information is interactive voice or interactive text, and the operation prompt is used to determine whether to execute the first operation corresponding to the interactive information.

[0076] It can be understood that the first operation refers to the water heater operation corresponding to the interaction information received by the large language model; the executor of this step is the server, and the large language model is configured on the server. The large language model is trained based on historical usage data and basic common sense information of different smart home appliances. In this way, it can accurately identify the user's intention related to the water heater in the current usage request, and provide users with simpler and faster services.

[0077] After receiving the interaction information sent by the client for operating the water heater, the interaction information is input into the large language model. Based on the interaction information, the large language model determines the operation information corresponding to the execution of the interaction information, that is, the first operation, and generates an operation prompt for interacting with the user based on the obtained first operation.

[0078] For example, a user interacts with a smart mobile terminal through a water heater APP set on the smart mobile terminal, and sends the interaction information to a cloud server; if the interaction information received by the cloud server is: "I want to turn off the electric water heater in the kitchen", the interaction information is input into the large language model, and the output operation prompt is: "Please confirm to turn off the electric water heater in the kitchen", and the first operation corresponding to the operation prompt is: "Turn off the electric water heater in the kitchen".

[0079] S202: In response to confirmation information for the first operation sent by the client, trigger execution of the first operation.

[0080] The confirmation information is used to instruct the user to give feedback on the operation prompt and to clarify whether the first operation is executed.

[0081] It can be understood that after the large language model generates an operation prompt, the server sends the operation prompt to the client; based on the operation prompt, the client interacts with the user, and the user confirms whether the operation prompt recognized by the large language model is correct and whether to perform the corresponding first operation.

[0082] If the confirmation information for the first operation sent by the client is received, it indicates that the user confirms that the operation prompt output by the large language model is correct, and the first operation corresponds to the operation intention of the user. Figure 1 At this time, the first operation is triggered.

[0083] Since the server where the large language model is located is also equipped with various function tables, including: a device control function table and a device statistics function table; after triggering the execution of the first operation, the operation type corresponding to the first operation is determined, that is, it is determined whether the operation to be executed at that time belongs to the operation that can be implemented by the device control function table or the operation that can be implemented by the device statistics function table; after determining the corresponding operation type, the first operation is matched with the function table of the corresponding operation type, from which the target operation to be executed is determined, and the relevant equipment is controlled to execute.

[0084] For example, after the large language model generates an operation prompt, the cloud server feeds back the operation prompt to the smart mobile terminal, which interacts with the user until the user confirms the operation information to be performed this time. Specifically, the smart mobile terminal outputs a confirmation window based on the operation information, that is, a "Please confirm to turn off the electric water heater in the kitchen" window pops up, and the user determines whether to perform the operation; if the user chooses to perform the operation, the confirmation information received by the smart mobile terminal may be: "Turn off the electric water heater in the kitchen", and the confirmation information is sent to the cloud server; the cloud server receives the confirmation information for the first operation sent by the smart mobile terminal, which indicates that the user confirms that the operation prompt output by the large language model corresponds to the operation intention corresponding to the user's request. Figure 1 At this time, the first operation is triggered and the operation type of the first operation is confirmed; if the first operation is: "Turn off the electric water heater in the kitchen", it corresponds to the operation that can be realized in the device control function table. At this time, the first operation is matched with the device control function table to determine the specific operation information as: "Shutdown", and the control object is: "Kitchen electric water heater", and the specific operation information is sent to the control object to realize the control of the water heater, avoiding the cumbersome switching entrance of the water heater APP and simplifying the operation process of the device. This application does not impose special restrictions on the type and number of interactions between users and smart interactive devices.

[0085] The water heater control method provided in this embodiment responds to the interactive information sent by the client for operating the water heater, outputs the operation prompt corresponding to the interactive information through the large language model, thereby determining whether to execute the first operation corresponding to the interactive information; when responding to the confirmation information for the first operation sent by the client, the execution of the first operation is triggered, thereby simplifying the operation process of the device. The method uses the large language model to realize intelligent interactive prompts, reduces the complex operation of users switching between multiple entrances, improves the operation efficiency of the device, and improves the user experience.

[0086] Figure 3 Schematic diagram of the process of the water heater control method provided in the embodiment of the present application Figure 2 .like Figure 3 As shown, in this embodiment Figure 2 Based on the embodiment, the control method of the water heater is described in detail. The control method of the water heater shown in this embodiment includes:

[0087] S301: Acquire a knowledge question and answer corpus related to water heaters.

[0088] S302: Obtain an initial large language model based on the knowledge question and answer corpus training.

[0089] S303: Use the reward model to score the output of the initial large language model, and adjust the parameters of the initial large language model based on the score to obtain a large language model.

[0090] Among them, the knowledge question and answer corpus is used to indicate a database containing a large number of question and answer pairs related to water heaters, the reward model is used to measure the output quality of the large language model, the initial large language model is used to indicate the preliminary model trained based on the knowledge question and answer corpus, and the large language model is used to indicate the language model after multiple rounds of training and optimization.

[0091] It is understandable that the knowledge question and answer corpus contains a large number of question and answer pairs related to water heaters and basic common sense information related to water heaters, and the question and answer pairs stored in the corpus cover various functions of water heaters, usage methods, common problems and their solutions, etc.; the instruction question and answer corpus is used to perform preliminary training of the large language model, so that it can understand and answer questions related to water heaters, as well as learn the operating logic of water heaters and common user needs, thereby improving the accuracy and practicality of the large language model. This application does not impose any special restrictions on the scope of data information stored in the knowledge question and answer corpus.

[0092] Specifically, the construction of the reward model includes:

[0093] The interactive information samples are input into the initial large language model to obtain the answer samples output by the initial large language model; the interactive information samples, the answer samples, and the scores annotated for the answer samples are used as training samples to train a reward model.

[0094] The interactive information samples are used to indicate the instructions or questions input by the user through voice or text, and the answer samples are used to indicate the answers generated by the initial large language model based on the interactive information samples.

[0095] It can be understood that the interactive information sample is various operations or question-and-answer information related to the water heater. The interactive information sample is input into the initial large language model to obtain the response information output by the initial large language model, and the response information is determined as the answer sample, and the answer sample has a unique corresponding interactive information sample; the answer sample is labeled to obtain the score of the answer sample, and the interactive information sample, the answer sample, and the score of the annotated answer sample are used as training samples to train the reward model; wherein the score of the answer sample can be, for example, a degree score corresponding to a, b, c, d, or a score out of ten. This application does not impose any special restrictions on the labeling method and scoring type of the answer sample.

[0096] For example, historical usage data and basic equipment knowledge information related to water heaters are obtained to construct a knowledge question and answer corpus; a supervised learning model is used, and the knowledge question and answer corpus is input into the supervised learning model to train an initial large language model; a large number of dialogue exercises are conducted through the initial large language model, sampling an input and corresponding multiple answer outputs, and manually marking all the pros and cons, such as a>b>c>d (a indicates excellent answer quality, b indicates good answer quality, c indicates passing answer quality, and d indicates failing answer quality), to generate a reward model, wherein multiple annotations can be performed when performing scoring annotation, and the average of multiple annotations is calculated as the score of the corresponding answer; the reward model is used to score the output of the initial large language model, and the parameters of the initial large language model are adjusted based on the score to obtain a large language model.

[0097] S304: In response to the interaction information sent by the client, output questions corresponding to the interaction through the large language model.

[0098] S305: In response to the answer to the question sent by the client, output an operation prompt corresponding to the interaction through the large language model.

[0099] After receiving the interactive information for operating the water heater sent by the client, the interactive information is input into the large language model, the large language model parses and understands the user's intention, and based on the parsed information, generates a question corresponding to the user interaction, and sends the question to the intelligent interactive device, which interacts with the user again to further determine the user's operation intention; when receiving the answer to the question sent by the client, the answer is input into the large language model, and the large language model is used to further parse and understand the user's specific needs, so as to determine the operation information corresponding to the answer, that is, the first operation, and generate an operation prompt for interacting with the user based on the obtained first operation.

[0100] It is understandable that the main reason for multiple interactions is that the user's interaction information is not accurate or specific enough, and the large language model needs to clarify the user's specific needs through further dialogue; when the large language model outputs the corresponding question, the model determines that the user's operation intention is unclear, which may easily lead to the system being unable to directly determine the user's specific needs. Therefore, it is necessary to interact with the user again based on the question determined at that time.

[0101] For example, if the cloud server receives the user's interactive information as "I want to increase the temperature of the water heater", the interactive information is input into the large language model, and the large language model parses the interactive information to generate a question: "How many degrees do you want to adjust the temperature to?"; after receiving the question, the smart mobile terminal continues to interact with the user. If the user's answer obtained this time is: "45 degrees", the large language model is used again to parse the answer and generate an operation prompt: "Do you want to adjust the water heater temperature to 45 degrees?".

[0102] S306: If the first operation is a control operation, query a control function parameter corresponding to the first operation from a preset control function table, and send a control instruction corresponding to the first operation to the water heater based on the control function parameter.

[0103] S307: If the first operation is a query operation, query a query parameter corresponding to the first operation from a preset query function table, query a query result corresponding to the first operation from a preset database based on the query parameter, and send the query result to the client.

[0104] Among them, control operations are used to indicate that the user wants to perform some specific control operations on the water heater, such as adjusting the temperature, turning the machine on and off, etc.; query operations are used to indicate that the user wants to obtain certain status information or data of the water heater, such as the electricity consumption of the month, the number of temperature adjustments of the month, etc.; control function parameters are used to assist the server in generating specific control instructions, and query parameters are used to assist the server in generating specific query results; the preset database is used to store queryable data information related to the water heater.

[0105] It can be understood that in the smart water heater system, user operations can be divided into two categories: control operations and query operations; the server processes these two types of operations respectively through the preset control function table and query function table to simplify the operation process and improve operation accuracy and user experience.

[0106] Specifically, the control function table and the query function table are both preset databases or data structures; the control function table contains all possible control operations of smart home appliances and their corresponding control function parameters, and is used to store and query parameter information related to control operations to help the system generate corresponding control instructions; the control function table may contain items such as "set temperature", "turn on", "turn off", etc., and each item corresponds to a specific control function parameter. Help the system generate accurate control instructions to execute the user's control operations; Table 1 is the control function table provided by this application, as shown below:

[0107] Table 1

[0108] The query function table contains all possible query operations and their corresponding query parameters. The query function table is used to store and query parameter information related to query operations to help the system generate corresponding query results; the query function table may contain entries such as "query the electricity consumption of the month", "query the number of temperature adjustment times", "query the electricity consumption of the day", etc., and each entry corresponds to a specific query parameter; Table 2 is the query function table provided by this application, which is as follows:

[0109] Table 2

[0110] In addition, the above-mentioned control function table 1 and query function table 2 are only illustrative of part of the storage content in the corresponding table. This application does not impose any special restrictions on the specific content stored in the control function table 1 and query function table 2.

[0111] According to the operation prompt output by the large language model at that time, the corresponding first operation and the operation type of the first operation are determined. If the first operation is a control operation, the control function parameters corresponding to the first operation are queried from the preset control function table; based on the control function parameters, the corresponding control instructions are generated, and the control instructions corresponding to the first operation are sent to the water heater; if the first operation is a query operation, the query parameters corresponding to the first operation are queried from the preset query function table; based on the query parameters, the query results corresponding to the first operation are queried from the preset database, and the query results are sent to the client to respond to the user's query request.

[0112] For example, if the first operation determined this time is: "Adjust the temperature of the electric water heater described by a to 45 degrees", and the operation type of the first operation is a control operation, at this time, the server calls the device control function table, that is, the control function table, and queries the control function parameters of the "set temperature" operation from the control function table, that is, the function field "onOffStatus" in Table 1, and generates a control instruction based on the function field: "Set the temperature to 45 degrees", and sends the control instruction to the electric water heater described by a that performs the operation, and the electric water heater described by a performs the temperature adjustment operation; if the first operation determined this time is: "Query the current electricity consumption of water heater a", and the operation type of the first operation is a query operation, at this time, the server calls the device query function table, that is, the query function table, and queries the query parameters of the "query current electricity consumption" operation from the query function table, that is, the indicator field "onOffStatus" in Table 2, and based on the indicator field, queries the query result corresponding to the first operation from the preset database, and sends the query result to the smart mobile terminal to respond to the user's query request.

[0113] Preferably, in response to confirmation information sent by the client for the operation prompt, triggering execution of the first operation includes:

[0114] If the first operation is a control operation, query whether there is a first operation trajectory with the first operation as the starting operation in the operation trajectory prediction table; if the first operation trajectory exists, send a prediction prompt to the client; in response to the confirmation information sent by the client for the first operation trajectory, trigger the execution of the first operation trajectory.

[0115] The operation trajectory prediction table includes one or more operation trajectories for the water heater, the operation trajectory includes multiple operations performed in sequence, and the prediction prompt is used to determine whether to execute the first operation trajectory.

[0116] It can be understood that the intelligent interactive device can also provide users with intelligent trajectory prediction functions. Through the operation trajectory prediction table, the server can predict the user's operation needs in advance, simplify the operation process, improve operation efficiency, and enable users to control the water heater more conveniently and efficiently.

[0117] After determining the first operation, first determine whether the user's operation is a control operation; if the first operation is a control operation, it indicates that the current operation involves at least one smart home appliance. At this time, query in the operation trajectory prediction table whether there is a first operation trajectory with the first operation as the starting operation; if the first operation trajectory exists, send a prediction prompt to the client, and prompt the user of possible subsequent operations based on the user's usage habits at the time; when the client receives confirmation information for the first operation trajectory, it sends the confirmation information to the server, and the server determines that the first operation trajectory needs to be executed.

[0118] For example, the operation trajectory prediction table is shown in Table 3 below:

[0119] Table 3

[0120] If the first operation received this time is: "Adjust the temperature of the water heater to 45 degrees", and the first operation is a control operation; call the operation trajectory prediction table, and query in the operation trajectory prediction table whether there is an operation trajectory with "set temperature" as the starting operation, that is, the operation trajectory corresponding to electric water heater a: "Adjust the temperature to 30→antibacterial wash", at this time, generate a prediction prompt based on the operation trajectory corresponding to the electric water heater a, and send the prediction prompt to the smart mobile terminal, "Do you need to turn on smart trajectory control?", and receive the confirmation information sent by the smart mobile terminal. If the user confirms "Yes, turn on smart trajectory control", the operation trajectory corresponding to the electric water heater a is executed, and according to the time node corresponding to the operation trajectory, the electric water heater a is controlled to perform the corresponding operation.

[0121] Preferably, in response to the denial information for the first operation track sent by the client, the first operation is triggered to be executed, or, if the first operation track does not exist, the first operation is triggered to be executed.

[0122] It can be understood that in the smart water heater system, when the user's operation is identified as a control operation, the system will predict possible subsequent operations through the operation trajectory prediction table and send prediction prompts to the user; in addition, the user can confirm or reject the system's prediction through confirmation information or denial information, and the confirmation information refers to the user's confirmation to execute the first operation trajectory, and the denial information refers to the user's refusal to execute the first operation trajectory.

[0123] For example, if the first operation received this time is: "Adjust the temperature of the water heater to 45 degrees", and the first operation is a control operation; when a denial message sent by the smart mobile terminal is received, if the user confirms that "there is no need to turn on smart trajectory control", the operation trajectory corresponding to the electric water heater a will not be executed; if the operation trajectory prediction table cannot match an operation trajectory with the first operation as the starting operation, the first operation can be executed directly.

[0124] The control method of the water heater provided in the present embodiment obtains a knowledge question and answer corpus related to the water heater, obtains an initial large language model based on the knowledge question and answer corpus training, then uses a reward model to score the output of the initial large language model, and adjusts the parameters of the initial large language model based on the score to obtain the large language model; in response to the interaction information sent by the client, the large language model outputs a question corresponding to the interaction, and in response to the answer to the question sent by the client, the large language model outputs an operation prompt corresponding to the interaction, so as to determine whether to execute the first operation corresponding to the interaction information; if the first operation is a control operation, the control function parameters corresponding to the first operation are queried from a preset control function table, and based on the control function parameters, a control instruction corresponding to the first operation is sent to the water heater; if the first operation is a query operation, the query parameters corresponding to the first operation are queried from a preset query function table, and based on the query parameters, a query result corresponding to the first operation is queried from a preset database, and the query result is sent to the client. This method uses a large language model to implement intelligent interactive prompts, reduces operational complexity and the need for multi-entry switching, improves the operating efficiency of the device, reduces the burden on users to remember specific functional locations, avoids user misoperations due to operational complexity, simplifies the device operation process, and improves the user experience.

[0125] Figure 4 This is a schematic diagram of the structure of the control device of the water heater provided in this application. Figure 4 As shown, the present application provides a control device for a water heater, and the control device 400 for the water heater includes:

[0126] The response module 401 is used to respond to the interactive information sent by the client for operating the water heater, wherein the interactive information is interactive voice or interactive text.

[0127] The processing module 402 is configured to output an operation prompt corresponding to the interaction information through a large language model, wherein the operation prompt is used to determine whether to execute a first operation corresponding to the interaction information.

[0128] The processing module 402 is further configured to trigger execution of the first operation in response to confirmation information sent by the client for the first operation.

[0129] Optionally, the response module 401 is further configured to respond to the interaction information sent by the client.

[0130] The processing module 402 is further configured to output a question corresponding to the interaction through the large language model.

[0131] The response module 401 is further configured to respond to the answer to the question sent by the client.

[0132] The processing module 402 is further configured to output an operation prompt corresponding to the interaction through the large language model.

[0133] Optionally, if the first operation is a control operation, the processing module 402 is further used to query the control function parameters corresponding to the first operation from a preset control function table, and based on the control function parameters, send a control instruction corresponding to the first operation to the water heater.

[0134] If the first operation is a query operation, the processing module 402 is also used to query the query parameters corresponding to the first operation from the preset query function table, and query the query results corresponding to the first operation from the preset database based on the query parameters, and send the query results to the client.

[0135] Optionally, if the first operation is a control operation, the processing module 402 is also used to query in the operation trajectory prediction table whether there is a first operation trajectory starting with the first operation, wherein the operation trajectory prediction table includes one or more operation trajectories for the water heater, and the operation trajectory includes multiple operations performed in sequence.

[0136] If the first operation track exists, the processing module 402 is further configured to send a prediction prompt to the client, where the prediction prompt is used to determine whether to execute the first operation track.

[0137] The response module 401 is further configured to respond to confirmation information regarding the first operation track sent by the client.

[0138] The processing module 402 is further configured to trigger execution of the first operation trajectory.

[0139] Optionally, the response module 401 is further configured to respond to a denial message regarding the first operation track sent by the client.

[0140] The processing module 402 is further used to trigger the execution of the first operation, or if the first operation track does not exist, the processing module 402 is further used to trigger the execution of the first operation.

[0141] Optionally, the control device of the water heater further includes: an acquisition module 403.

[0142] The acquisition module 403 is used to acquire a knowledge question and answer corpus related to water heaters.

[0143] The processing module 402 is also used to train an initial large language model based on the knowledge question and answer corpus, and the initial large prediction model is used to output a corresponding answer based on the input information.

[0144] The processing module 402 is further configured to score the output of the initial large language model using a reward model, and adjust parameters of the initial large language model based on the score to obtain the large language model.

[0145] Optionally, the processing module 402 is further configured to input the interaction information sample into the initial large language model to obtain an answer sample output by the initial large language model.

[0146] The processing module 402 is further configured to use the interaction information sample, the answer sample, and the score annotated for the answer sample as training samples to train and obtain the reward model.

[0147] Figure 5 This is a schematic diagram of the structure of the water heater control device provided in this application. Figure 5 As shown, the present application provides a control device for a water heater, and the control device 500 for the water heater includes: a receiver 501 , a transmitter 502 , a processor 503 and a memory 504 .

[0148] Receiver 501, used for receiving instructions and data;

[0149] A transmitter 502, used for sending instructions and data;

[0150] Memory 504, used to store computer-executable instructions;

[0151] The processor 503 is used to execute the computer-executable instructions stored in the memory 504 to implement the various steps executed by the water heater control method in the above embodiment. For details, please refer to the relevant description in the above water heater control method embodiment.

[0152] Optionally, the memory 504 may be independent or integrated with the processor 503 .

[0153] When the memory 504 is independently provided, the electronic device further includes a bus for connecting the memory 504 and the processor 503 .

[0154] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, a water heater control method as executed by the water heater control device described above is implemented.

[0155] An embodiment of the present application further provides a computer program product, including a computer program, which implements the method described in any of the above embodiments when executed by a processor.

[0156] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0157] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments, and the above embodiments are only used to illustrate the technical solution of the present application rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for controlling a water heater, characterized in that: The method comprises: In response to interaction information sent by the client for operating the water heater, outputting an operation prompt corresponding to the interaction information through a large language model, wherein the interaction information is interaction voice or interaction text, and the operation prompt is used to determine whether to perform a first operation corresponding to the interaction information; In response to confirmation information for the first operation sent by the client, execution of the first operation is triggered.

2. The method according to claim 1, characterized in that: The step of responding to the interactive information sent by the client for operating the water heater and outputting an operation prompt corresponding to the interactive information through the large language model includes: In response to the interaction information sent by the client, outputting a question corresponding to the interaction through the large language model; In response to the answer to the question sent by the client, an operation prompt corresponding to the interaction is output through the large language model.

3. The method according to claim 1, characterized in that The triggering, in response to the confirmation information sent by the client for the first operation, to execute the first operation comprises: If the first operation is a control operation, querying a control function parameter corresponding to the first operation from a preset control function table, and sending a control instruction corresponding to the first operation to the water heater based on the control function parameter; If the first operation is a query operation, query parameters corresponding to the first operation are queried from a preset query function table, and query results corresponding to the first operation are queried from a preset database based on the query parameters, and the query results are sent to the client.

4. The method according to claim 3, characterized in that: The triggering, in response to the confirmation information sent by the client for the operation prompt, to execute the first operation includes: If the first operation is a control operation, querying in an operation trajectory prediction table whether there is a first operation trajectory starting with the first operation, wherein the operation trajectory prediction table includes one or more operation trajectories for the water heater, and the operation trajectory includes multiple operations executed in sequence; If the first operation trajectory exists, sending a prediction prompt to the client, where the prediction prompt is used to determine whether to execute the first operation trajectory; In response to confirmation information for the first operation trajectory sent by the client, execution of the first operation trajectory is triggered.

5. The method according to claim 4, characterized in that Also includes: In response to the denial information regarding the first operation track sent by the client, the first operation is triggered to be executed, or if the first operation track does not exist, the first operation is triggered to be executed.

6. The method according to any one of claims 1 to 5, characterized in that Also includes: Get the knowledge question and answer corpus related to water heaters; An initial large language model is obtained by training based on the knowledge question and answer corpus, and the initial large prediction model is used to output a corresponding answer based on the input information; The output of the initial large language model is scored using a reward model, and parameters of the initial large language model are adjusted based on the score to obtain the large language model.

7. The method according to claim 6, characterized in that Also includes: Inputting the interaction information sample into the initial large language model to obtain an answer sample output by the initial large language model; The interactive information sample, the answer sample, and the score annotated for the answer sample are used as training samples to train and obtain the reward model.

8. A control device for a water heater, characterized in that: include: A response module, used to respond to the interactive information sent by the client for operating the water heater, wherein the interactive information is interactive voice or interactive text; a processing module, configured to output an operation prompt corresponding to the interaction information through a large language model, wherein the operation prompt is used to determine whether to perform a first operation corresponding to the interaction information; The processing module is further configured to trigger execution of the first operation in response to confirmation information for the first operation sent by the client.

9. A water heater control device, characterized in that: include: Memory; processor; Wherein, the memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the control method of the water heater according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer execution instructions, and when the computer execution instructions are executed by the processor, they are used to implement the control method of the water heater according to any one of claims 1 to 7.

Citation Information

Cited By

  • Large model-based man-machine interaction method and device and man-machine interaction equipment

    CN119646125A

  • Method and device for constructing compliance index system based on large language model

    CN119990923A