Water quantity determination method and device, cooking equipment and storage medium

The water required for cooking tasks is predicted through the target prediction model, and the problem that water volume determination in the prior art depends on users' subjective judgment is solved, and more efficient and accurate water volume management is achieved, which improves cooking convenience and dish quality.

CN118897983APending Publication Date: 2024-11-05NINGBO FOTILE KITCHEN WARE CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202410809311.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-21
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

In the existing cooking technology, the determination of water volume depends on the user's subjective judgment, lacks convenience, and it is difficult to ensure whether the water volume is sufficient, which affects cooking efficiency and the quality of the dishes.

Method used

Using the target prediction model, by obtaining the expected and actual data of the cooking task, the model is trained to predict the required water volume, combine the benchmark water volume demand and the reference water volume demand, adjust the model parameters, generate water volume prediction results, and generate water addition instructions based on the prediction results.

Benefits of technology

It improves the efficiency and accuracy of water volume prediction, reduces the dependence on users' subjective judgments, and improves cooking convenience and dish quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118897983A_ABST
    Figure CN118897983A_ABST
Patent Text Reader

Abstract

The invention provides a water quantity determination method and device, cooking equipment and a storage medium. The method comprises the steps that target expected data and target actual data indicating a target cooking task are acquired, the target expected data comprise at least one of dish information, food material information and a cooking mode, and the dish information is used for describing a target dish expected to be obtained when the target cooking task is executed; the food material information is used for describing a target food material required for obtaining a target dish, the cooking mode is adopted to process the target food material, and the target actual data is used for describing a current state of a cooking utensil supporting a target cooking task; and outputting a corresponding water volume prediction result by using the target prediction model by taking the target expected data and the target actual data as input. The invention provides a more convenient and effective water volume determination scheme.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of kitchenware, and in particular to a method and device for determining water volume, cooking equipment and storage medium. Background Art

[0002] Water is often used as a heat transfer medium when cooking dishes with cooking utensils, such as steaming dishes. In the related art, whether the amount of water is sufficient depends on the user's inspection. If there is a lack of water, the user needs to add water manually, which lacks convenience. Therefore, it is necessary to provide a more convenient and effective water amount determination solution. Summary of the invention

[0003] In order to solve at least one of the technical problems mentioned above, the present application provides a method, device, cooking device and storage medium for determining water volume:

[0004] According to a first aspect of the present application, a method for determining water volume is provided, the method comprising:

[0005] Acquire target expected data and target actual data indicating a target cooking task, the target expected data including at least one of the following: dish information, ingredient information, and a cooking mode, the dish information being used to describe a target dish that the target cooking task is performed to obtain, the ingredient information being used to describe a target ingredient required to obtain the target dish, the cooking mode being used to process the target ingredient, and the target actual data being used to describe a current state of a cooking appliance that supports the target cooking task;

[0006] Taking the target expected data and the target actual data as input, using the target prediction model to output the corresponding water volume prediction result;

[0007] Among them, the target prediction model is obtained by training based on multiple sample groups and adjusting the parameters of the preset model during training, each of the sample groups corresponds to a preset cooking task, the sample group is composed of sample expected data and sample actual data indicating the preset cooking task, and the sample group is marked with a baseline required water volume for the preset cooking task, and the baseline required water volume is used to describe the demand for water used as a heat transfer medium to complete the preset cooking task.

[0008] According to a second aspect of the present application, a water quantity determination device is provided, the device comprising:

[0009] A data acquisition module: used to acquire target expected data and target actual data indicating a target cooking task, wherein the target expected data includes at least one of the following: dish information, ingredient information, and cooking mode, wherein the dish information is used to describe a target dish that is expected to be obtained by executing the target cooking task, the ingredient information is used to describe a target ingredient required to obtain the target dish, and the cooking mode is used to process the target ingredient, and the target actual data is used to describe a current state of a cooking appliance that supports the target cooking task;

[0010] Water volume prediction module: used to take the target expected data and the target actual data as input, and output the corresponding water volume prediction result using the target prediction model;

[0011] Among them, the target prediction model is obtained by training based on multiple sample groups and adjusting the parameters of the preset model during training, each of the sample groups corresponds to a preset cooking task, the sample group is composed of sample expected data and sample actual data indicating the preset cooking task, and the sample group is marked with a baseline required water volume for the preset cooking task, and the baseline required water volume is used to describe the demand for water used as a heat transfer medium to complete the preset cooking task.

[0012] According to a third aspect of the present application, a cooking device is provided, comprising a cooking utensil and a water quantity determination device as described in the second aspect, wherein the water quantity determination device determines whether to generate an instruction to add water to the cooking utensil based on a water quantity prediction result.

[0013] According to the fourth aspect of the present application, a computer-readable storage medium is provided, characterized in that at least one instruction or at least one program is stored in the computer-readable storage medium, and the at least one instruction or at least one program is loaded and executed by a processor to implement the water volume determination method as described in the first aspect.

[0014] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application.

[0015] Implementing this application has the following beneficial effects:

[0016] The present application provides a more convenient and effective water quantity determination solution. The water quantity is predicted by a target prediction model. With the help of the adaptability and reliability of the target prediction model obtained through training, the efficiency and accuracy of the determined water quantity prediction results can be improved. The water quantity prediction results are used to describe the demand for water used as a heat transfer medium to complete the target cooking task, which can reduce the reliance on the user's subjective judgment of whether the water quantity is sufficient, thereby improving the cooking convenience on the user side.

[0017] Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above-mentioned objectives, technical solutions and beneficial effects of the present invention can be clearly obtained through the following detailed description of specific embodiments that can implement the present invention, combined with the description of the accompanying drawings.

[0019] The same reference numerals and symbols in the drawings and the description are used to represent the same or equivalent elements.

[0020] Figure 1 It is a flow chart of a method for determining water volume provided in this application;

[0021] Figure 2 It is a flowchart of the training process of obtaining the target prediction model provided by this application;

[0022] Figure 3 It is a schematic diagram of the process of obtaining the benchmark water demand provided by this application;

[0023] Figure 4 This is a device block diagram of a water volume determination device provided in this application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application 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 data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0027] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0028] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0029] In addition, in order to better illustrate the present application, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present application can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present application.

[0030] Figure 1 A schematic diagram of a water volume determination method according to an embodiment of the present application is shown, as follows: Figure 1 As shown, the method includes:

[0031] S101: Acquire target expected data and target actual data indicating a target cooking task, wherein the target expected data includes at least one of the following: dish information, ingredient information, and a cooking mode, wherein the dish information is used to describe a target dish that the target cooking task is executed to obtain, the ingredient information is used to describe a target ingredient required to obtain the target dish, and the cooking mode is used to process the target ingredient, and the target actual data is used to describe a current state of a cooking appliance that supports the target cooking task;

[0032] In the embodiment of the present application, the execution subject of the water volume determination method provided in the embodiment of the present application can be the cooking utensil itself, or it can be a water volume determination device belonging to the same cooking device as the cooking utensil. The cooking utensil provides support for the cooking task. The cooking utensil can be a steamer, and the water used as the heat transfer medium is an indirect heat transfer medium, and the steam generated by it is a direct heat transfer medium; the cooking utensil can be a boiling pot or a stew pot, and the water used as the heat transfer medium itself is a direct heat transfer medium.

[0033] For a cooking task, it corresponds to a dish preparation. The target expectation data indicating the target cooking task describes the expectation for a dish preparation. The target expectation data includes at least one of the following: dish information, ingredient information, and cooking mode. Dish information is used to describe the target dish that the target cooking task is executed to obtain, such as the expected preparation of dish A. Ingredient information is used to describe the target ingredients required to obtain the target dish, such as meat ingredient a1 and vegetable ingredient a2 required to make dish A. Cooking mode is adopted to process the target ingredients. Cooking is a complex and regular process of converting ingredients into dishes. Cooking mode can be steaming mode, boiling mode, stewing mode, etc. If steaming mode, boiling mode, and stewing mode are large category modes. Then, there can be more specific sub-category modes under the large category mode.

[0034] The target actual data indicating the target cooking task focuses on the cooking utensil, and is used to describe the current state of the cooking utensil supporting the target cooking task, such as the current temperature and the current humidity. For example, if the cooking utensil is a steamer, the target actual data may include the current temperature inside the steamer and the current humidity inside the steamer.

[0035] S102: Taking the target expected data and the target actual data as input, outputting corresponding water volume prediction results using a target prediction model;

[0036] In the embodiment of the present application, the target expected data and the target actual data are the input of the target prediction model, and the water volume prediction result is the output of the target prediction model. For the target prediction model, it is obtained by training based on multiple sample groups and adjusting the parameters of the preset model during the training. Each sample group corresponds to a preset cooking task, and the sample group is composed of sample expected data and sample actual data indicating the preset cooking task. The sample group is annotated with a reference required water volume for the preset cooking task, and the reference required water volume is used to describe the demand for water used as a heat transfer medium for the completion of the preset cooking task. For the preset cooking task here, reference can be made to the introduction of the cooking task and the target cooking task in the aforementioned S101, and for the sample expected data and sample actual data here, reference can be made to the introduction of the target expected data and the target actual data in the aforementioned S101, which will not be repeated. As for the reference required water volume annotated by the sample group, it can be understood that the completion of the preset cooking task requires the support of water used as a heat transfer medium. If there is insufficient water, it will affect the completion of the preset cooking task, that is, the dish preparation fails, such as some ingredients are not steamed. The reference required water volume gives the amount of water that can support the completion of the preset cooking task. It should be noted that the baseline required water volume focuses on whether the water volume can support the completion of the preset cooking task. Generally, the starting state provided by the samples participating in the preset cooking task is before using the cooking utensils to prepare the dishes.

[0037] The target prediction model obtained through training realizes the establishment of the correlation between the sample group and the benchmark water demand. During the training, the correlation between the sample group and the benchmark water demand is established through information mining and capture. The application of multiple sample groups also ensures the adaptability and reliability of the established correlation. Accordingly, when processing the target expected data and target actual data indicating the target cooking task, the water volume prediction result can be quickly obtained through the target prediction model. The water volume prediction result gives the water volume that can support the completion of the target cooking task. In this way, it can be free from the reliance on the user's subjective judgment of whether the water volume is sufficient.

[0038] The embodiments of the present application will be described in detail below.

[0039] As a possible implementation, Figure 2 As shown, the target prediction model is trained by the following steps:

[0040] S201: Taking the sample group as input, using the preset model to predict the corresponding reference water demand;

[0041] S202: Based on the difference between the reference demand water volume and the benchmark demand water volume, adjusting the parameters of the preset model to obtain the target prediction model.

[0042] During training, the parameters of the preset network are adjusted based on the preset differences, and the preset differences characterize the difference between the reference water demand and the benchmark water demand. By learning the preset differences during training, the size of the differences can be used to capture which data in the sample group provides support for more accurate water demand forecasts, and the size of the differences can be used to explore the contribution of each data in the sample group to more accurate water demand forecasts. Model training is ongoing, and the target prediction model can be the result of the previous training or the basis for the next training. The target prediction model is a model with high generalization ability obtained through training. Using the target prediction model for water demand forecasting can improve the adaptability and reliability of water demand forecasting.

[0043] Furthermore, the parameter adjustment of the preset model based on the difference between the reference water demand and the benchmark water demand may include the following steps: first, constructing a target loss function based on the difference between the reference water demand and the benchmark water demand; and then, adjusting the parameters of the preset model based on the target loss function to meet the model convergence conditions.

[0044] The reference water demand represents the actual value, and the benchmark water demand represents the ideal value. The target loss function is constructed based on the difference between the actual value and the ideal value, and then the parameters of the preset model are adjusted based on the target loss function. The loss function can be used to measure whether the model parameters fit the training data well and whether the model parameters are excellent. In this way, the model parameters can be better measured to determine whether to end the parameter adjustment together with the model convergence conditions. In the process of training to obtain the target prediction model, at least one model parameter adjustment may be involved, and the basis for each model parameter adjustment is the difference between the reference water demand and the benchmark water demand corresponding to each sample group in the corresponding batch sample group. A batch sample group can be a whole of multiple sample groups, or a part of a whole of multiple sample groups.

[0045] In addition, considering the type of data in the sample group, more targeted processing can be provided for different types of data, thereby improving the effectiveness and accuracy of data participating in the demand water volume forecast. For the aforementioned step S201, taking the sample group as input and using the preset model to predict the corresponding reference demand water volume may include the following steps:

[0046] 1) for each text class data in the sample group, using the pre-trained language model to output a text vector corresponding to the text class data to obtain a text vector set;

[0047] 2) for each numerical data in the sample group, using the preset normalization algorithm to process the numerical data to obtain a corresponding normalization result, so as to obtain a normalization result set;

[0048] 3) Taking the text vector set and the normalized result set as input, the reference required water volume is output using the preset prediction model.

[0049] The preset model includes a pre-trained language model, a preset normalization algorithm, and a preset prediction model. The pre-trained language model is responsible for processing text data, that is, converting it into vectors. The preset normalization algorithm is responsible for processing numerical data, that is, normalization. The preset prediction model is responsible for predicting the demand water volume. The text vector set and the normalized result set are the inputs of the preset prediction model, and the reference demand water volume is the output of the preset prediction model.

[0050] Exemplarily, the pre-trained language model adopts the BERT model (a language representation model). The sample dish information (such as recipe nouns) and the sample cooking mode in the sample group are all text data. The sample dish information is input into the BERT model to obtain the corresponding first text vector. The sample cooking mode is input into the BERT model to obtain the corresponding second text vector. The dimensions of the first text vector and the second text vector can both be 1*768. The first text vector can be represented by x1-x768, and the second text vector can be represented by x769-x1536.

[0051] The preset normalization algorithm can be expressed as a i represents the i-th numerical data among N numerical data, and the value range of i is 1-N. μ represents the mean of N numerical data, and σ represents the standard deviation of N numerical data. The sample food information (such as food weight), sample temperature, and sample humidity in the sample group can constitute the N numerical data here. Through the preset normalization algorithm, the first normalized result x1537 corresponding to the sample food information, the first normalized result x1538 corresponding to the sample temperature, and the third normalized result x1539 corresponding to the sample humidity can be obtained.

[0052] On this basis, considering the relationship between numerical data and the executed cooking task in the time dimension, the introduction of a preset time decay algorithm can improve the contribution of such data to the demand water volume prediction. It can be understood that the executed cooking task can be in a state before the cooking pot is used to make dishes, or in a state during the cooking pot is used to make dishes. For the same cooking task, the different stages of using cooking pots to make dishes are related to time. Through the preset time decay algorithm, the current stage can be determined from the time dimension, and then the expression of such data participating in the demand water volume prediction can be updated to improve the accuracy of the demand water volume prediction. Therefore, regardless of whether the current state of the cooking task is before the cooking pot is used to make dishes, or during the cooking pot is used to make dishes, the target prediction model can be used to accurately predict the demand water volume. Of course, at this time, the sample group also needs to add data in the time dimension, such as the difference between the current time and the start time mentioned later. The benchmark demand water volume marked by the sample group focuses on whether the water volume can support the completion of the preset cooking task, and does not limit the starting state provided by the preset cooking task participating sample. The starting state provided by the preset cooking task participation sample can be before using the cooking pot to prepare the dish, or can be during the use of the cooking pot to prepare the dish.

[0053] For the aforementioned step 2), for each numerical data in the sample group, after using the preset normalization algorithm to process the numerical data to obtain the corresponding normalized result to obtain a normalized result set, the method may further include the following steps: for each normalized result, determining the corresponding time decay coefficient based on the data item to which the normalized result belongs, and processing the normalized result based on the preset time decay algorithm and the time decay coefficient to obtain the target result to obtain a target result set. Accordingly, taking the text vector set and the normalized result set as input and using the preset prediction model to output the reference required water volume may include the following steps: taking the text vector set and the target result set as input and using the preset prediction model to output the reference required water volume.

[0054] Exemplarily, the preset time decay algorithm can be expressed as x j represents the normalized result, such as the first normalized result x1537, the second normalized result x1538, and the third normalized result x1539. λ represents the time decay coefficient. The time decay coefficient corresponding to the first normalized result x1537 can be 1, the time decay coefficient corresponding to the second normalized result x1538 can be 3, and the time decay coefficient corresponding to the third normalized result x1539 can also be 3. T1 represents the current time, and T0 represents the start time. The start time can be understood as the time when the cooking pot is used to start cooking.

[0055] In practical applications, in addition to the above-mentioned sample dish information, sample cooking mode, and sample ingredient information, the sample expected data may also include sample cooking time (the expected time for using cooking utensils to prepare dishes) and sample cooking temperature information (the expected temperature for using cooking utensils to prepare dishes). The above-mentioned preset normalization algorithm can be used to process the sample cooking time and the sample cooking temperature information respectively to obtain the fourth normalization result x1540 corresponding to the sample cooking time and the fifth normalization result x1541 corresponding to the sample cooking temperature information. Correspondingly, the target result set input into the preset prediction model also includes the fourth normalization result x1540 and the fifth normalization result x1541.

[0056] The processing of the target result set by the preset prediction model can be reflected by the following formula:

[0057] H(w,b)=wX+b=w1x1+w2x2+w3x3+w4x4...+w1541x1541+b

[0058] Among them, w1-1541 represents the training parameters of the model, and b represents the bias parameters of the model.

[0059] The loss function constructed during training can be expressed as follows:

[0060]

[0061] Among them, n represents the total number of sample groups, y i It represents the benchmark water demand of the i-th sample group among the n sample groups.

[0062] The partial derivative of w can be obtained through the minimum loss function to obtain the downhill direction of the loss function:

[0063]

[0064] Among them, x i Represents the i-th sample group among n sample groups.

[0065] Then control the step size through the learning rate α to update the training parameters w of the model:

[0066]

[0067] Repeat the step of "controlling the step size by the learning rate α to update the training parameter w of the model" until the specified number of iterations or the set loss function threshold.

[0068] Of course, the above-mentioned preset time decay algorithm can also be used to continue processing the fourth normalized result x1540 and the fifth normalized result x1541. The time decay coefficient corresponding to the fourth normalized result x1540 can be 1.5, and the time decay coefficient corresponding to the fifth normalized result x1541 can also be 1.5. Correspondingly, the target result set input into the preset prediction model also includes the results of the fourth normalized result x1540 and the fifth normalized result x1541 processed by the above-mentioned preset time decay algorithm.

[0069] It should be noted that when applying the target prediction model, the respective connotations of the target expected data and the target actual data of the input model should generally be consistent with the respective connotations of the training sample expected data and the sample actual data.

[0070] As a possible implementation, Figure 3 As shown, the method also includes:

[0071] S301: Acquire a first water volume, where the first water volume describes a prepared state of preset water used as a heat transfer medium to complete the preset cooking task;

[0072] S302: Acquire a second water volume, where the second water volume describes the remaining amount of the preset water after the preset cooking task is completed;

[0073] S303: Determine the difference between the first water volume and the second water volume to obtain the reference required water volume.

[0074] For the first water volume, it describes the preparation of the preset water used as a heat transfer medium for the completion of the preset cooking task. It can be understood that before the introduction of the preset time decay algorithm, the water preparation here focuses on the application of cooking pots before the preparation of dishes. Before the introduction of the preset time decay algorithm, the water preparation here can focus on the application of cooking pots before the preparation of dishes, and the application of cooking pots during the preparation of dishes (such as a stage where the ingredients have been steamed but not yet cooked). For the second water volume, it describes the remaining situation of the preset water after the preset cooking task is completed. The difference between the first water volume and the second water volume can describe the water consumption starting from before the application of cooking pots to prepare dishes and ending at the completion of the preset cooking task; it can also describe the water consumption starting from the application of cooking pots to prepare dishes and ending at the completion of the preset cooking task. Such water consumption as the baseline demand water volume is conducive to ensuring that the demand water volume predicted by the model is more accurate. If the predicted demand water volume is excessive, the user can be guided to add excess water, which is wasteful and affects the cooking efficiency.

[0075] As a possible implementation, for the aforementioned step S202, after the target expected data and the target actual data are used as inputs and the corresponding water volume prediction result is outputted using the target prediction model, the method may also include the following steps: when the water volume prediction result is greater than the current water volume, a water adding instruction is generated, the water adding instruction is used to instruct the addition of water to the cooking utensil, and the current water volume describes the current situation of the water used as the heat transfer medium for the completion of the target cooking task. Whether to add water is determined based on the water volume prediction result. When the water volume prediction result is greater than the current water volume, a water adding instruction is generated. When the water volume prediction result is less than or equal to the current water volume, no water adding instruction is generated. By quickly obtaining a water volume prediction result with high accuracy, the timeliness and accuracy of determining whether to add water can be improved, which is beneficial to improving the quality of the dishes produced, and thus improving the user's cooking experience.

[0076] It can be seen from the technical solutions provided by the above embodiments of the present application that the embodiments of the present application provide a more convenient and effective solution for determining the amount of water. The water volume is predicted by the target prediction model, and with the help of the adaptability and reliability of the target prediction model obtained through training, the efficiency and accuracy of the determined water volume prediction results can be improved. The water volume prediction results are used to describe the demand for water used as a heat transfer medium to complete the target cooking task, which can reduce the reliance on the user's subjective judgment of whether the amount of water is sufficient, thereby improving the cooking convenience on the user side.

[0077] The present application also provides a water volume determination device, such as Figure 4As shown, the water volume determining device 40 comprises:

[0078] Data acquisition module 401: used to acquire target expected data and target actual data indicating a target cooking task, wherein the target expected data includes at least one of the following: dish information, ingredient information, and cooking mode, wherein the dish information is used to describe the target dish that the target cooking task is executed to obtain, the ingredient information is used to describe the target ingredients required to obtain the target dish, and the cooking mode is used to process the target ingredients, and the target actual data is used to describe the current state of the cooking appliance supporting the target cooking task;

[0079] The water volume prediction module 402 is used to take the target expected data and the target actual data as input and output the corresponding water volume prediction result using the target prediction model;

[0080] Among them, the target prediction model is obtained by training based on multiple sample groups and adjusting the parameters of the preset model during training, each of the sample groups corresponds to a preset cooking task, the sample group is composed of sample expected data and sample actual data indicating the preset cooking task, and the sample group is marked with a baseline required water volume for the preset cooking task, and the baseline required water volume is used to describe the demand for water used as a heat transfer medium to complete the preset cooking task.

[0081] In one embodiment, the target prediction model is trained by the following steps: taking the sample group as input, using the preset model to predict the corresponding reference demand water volume; based on the difference between the reference demand water volume and the benchmark demand water volume, adjusting the parameters of the preset model to obtain the target prediction model.

[0082] In one embodiment, the preset model includes a pre-trained language model, a preset normalization algorithm and a preset prediction model, and the sample group is taken as input and the corresponding reference required water volume is predicted using the preset model, including: for each text class data in the sample group, using the pre-trained language model to output the text vector corresponding to the text class data to obtain a text vector set; for each numerical class data in the sample group, using the preset normalization algorithm to process the numerical class data to obtain a corresponding normalized result to obtain a normalized result set; taking the text vector set and the normalized result set as input, using the preset prediction model to output the reference required water volume.

[0083] In one embodiment, the preset model further includes a preset time decay algorithm, and after the preset normalization algorithm is used to process each numerical data in the sample group to obtain a corresponding normalized result, so as to obtain a normalized result set, the device further includes:

[0084] A target result set acquisition module: for determining, for each normalized result, a corresponding time decay coefficient based on the data item to which the normalized result belongs, and processing the normalized result based on the preset time decay algorithm and the time decay coefficient to obtain a target result, so as to obtain a target result set;

[0085] The step of taking the text vector set and the normalized result set as input and using the preset prediction model to output the reference required water volume includes: taking the text vector set and the target result set as input and using the preset prediction model to output the reference required water volume.

[0086] In one embodiment, the parameters of the preset model are adjusted based on the difference between the reference water demand and the benchmark water demand, including: constructing a target loss function based on the difference between the reference water demand and the benchmark water demand; adjusting the parameters of the preset model based on the target loss function to meet the model convergence conditions.

[0087] In one embodiment, the apparatus further comprises:

[0088] A first water volume acquisition module: used to acquire a first water volume, wherein the first water volume describes the preparedness of preset water used as a heat transfer medium to complete the preset cooking task;

[0089] A second water volume acquisition module: used to acquire a second water volume, where the second water volume describes the remaining amount of the preset water after the preset cooking task is completed;

[0090] A reference required water volume acquisition module is used to determine the difference between the first water volume and the second water volume to obtain the reference required water volume.

[0091] In one embodiment, after taking the target expected data and the target actual data as input and outputting the corresponding water volume prediction result using the target prediction model, the device further comprises:

[0092] An instruction generation module: used to generate a water adding instruction when the water volume prediction result is greater than the current water volume, wherein the water adding instruction is used to instruct the addition of water to the cooking appliance, and the current water volume describes the current situation of water used as a heat transfer medium to complete the target cooking task.

[0093] It should be noted that the device in the device embodiment and the method embodiment are based on the same inventive concept.

[0094] An embodiment of the present application further provides a cooking device, which includes a cooking utensil and the above-mentioned water quantity determination device, and the water quantity determination device determines whether to generate an instruction to add water to the cooking utensil based on a water quantity prediction result.

[0095] In practical applications, when the water volume prediction result is greater than the current water volume, a water adding instruction can be generated. The water adding instruction is used to instruct the addition of water to the cooking appliance (such as adding water to the water tank of a steamer). The current water volume describes the current situation of water used as a heat transfer medium to complete the target cooking task.

[0096] The water volume determination device may be in the form of a server, such as a cloud. The user may choose to upload target expected data indicating a target cooking task to the server through a client running on the terminal. The server also obtains target actual data indicating a target cooking task through detection of sensors in the cooking utensil. The server may also obtain the current water volume through detection of sensors in the cooking utensil. After the server applies the target prediction model to generate a water volume prediction result, if it is determined that the water volume prediction result is greater than the current water volume, a water addition instruction may be sent to the client.

[0097] The embodiment of the present application also provides a computer-readable storage medium, wherein at least one instruction or at least one program is stored in the computer-readable storage medium, and the at least one instruction or at least one program is loaded and executed by a processor to implement the above method. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0098] An embodiment of the present application also provides a computer program product, which includes at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the above method.

[0099] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A method for determining water volume, characterized in that: The method comprises: Acquire target expected data and target actual data indicating a target cooking task, the target expected data including at least one of the following: dish information, ingredient information, and a cooking mode, the dish information being used to describe a target dish that the target cooking task is performed to obtain, the ingredient information being used to describe a target ingredient required to obtain the target dish, the cooking mode being used to process the target ingredient, and the target actual data being used to describe a current state of a cooking appliance that supports the target cooking task; Taking the target expected data and the target actual data as input, using the target prediction model to output the corresponding water volume prediction result; Among them, the target prediction model is obtained by training based on multiple sample groups and adjusting the parameters of the preset model during training, each of the sample groups corresponds to a preset cooking task, the sample group is composed of sample expected data and sample actual data indicating the preset cooking task, and the sample group is marked with a baseline required water volume for the preset cooking task, and the baseline required water volume is used to describe the demand for water used as a heat transfer medium to complete the preset cooking task.

2. The method according to claim 1, characterized in that The target prediction model is trained by the following steps: Taking the sample group as input, using the preset model to predict the corresponding reference water demand; Based on the difference between the reference demand water volume and the benchmark demand water volume, the parameters of the preset model are adjusted to obtain the target prediction model.

3. The method according to claim 2, characterized in that The preset model includes a pre-trained language model, a preset normalization algorithm and a preset prediction model. The method of taking the sample group as input and using the preset model to predict the corresponding reference water demand includes: For each text class data in the sample group, using the pre-trained language model to output a text vector corresponding to the text class data to obtain a text vector set; For each numerical data in the sample group, the numerical data is processed by the preset normalization algorithm to obtain a corresponding normalization result, so as to obtain a normalization result set; The text vector set and the normalized result set are used as input, and the reference required water volume is output using the preset prediction model.

4. The method according to claim 3, characterized in that The preset model further includes a preset time decay algorithm. For each numerical data in the sample group, the preset normalization algorithm is used to process the numerical data to obtain a corresponding normalized result, so as to obtain a normalized result set. The method further includes: For each of the normalized results, determining a corresponding time decay coefficient based on the data item to which the normalized result belongs, and processing the normalized result based on the preset time decay algorithm and the time decay coefficient to obtain a target result, so as to obtain a target result set; The step of taking the text vector set and the normalized result set as input and using the preset prediction model to output the reference required water volume includes: The text vector set and the target result set are used as input, and the reference required water volume is output using the preset prediction model.

5. The method according to any one of claims 2 to 4, characterized in that: The step of adjusting parameters of the preset model based on the difference between the reference water demand and the benchmark water demand includes: constructing a target loss function based on the difference between the reference water demand and the benchmark water demand; Based on the target loss function, the parameters of the preset model are adjusted to meet the model convergence conditions.

6. The method according to claim 1, characterized in that The method further comprises: acquiring a first water amount, wherein the first water amount describes a preparedness of preset water used as a heat transfer medium to complete the preset cooking task; Acquire a second water volume, where the second water volume describes the remaining amount of the preset water after the preset cooking task is completed; A difference between the first water volume and the second water volume is determined to obtain the reference required water volume.

7. The method according to claim 1, characterized in that After taking the target expected data and the target actual data as input and outputting the corresponding water volume prediction result using the target prediction model, the method further includes: When the water amount prediction result is greater than the current water amount, a water adding instruction is generated, wherein the water adding instruction is used to instruct the addition of water to the cooking appliance, and the current water amount describes the current situation of water used as a heat transfer medium to complete the target cooking task.

8. A water quantity determination device, characterized in that: The device comprises: A data acquisition module: used to acquire target expected data and target actual data indicating a target cooking task, wherein the target expected data includes at least one of the following: dish information, ingredient information, and cooking mode, wherein the dish information is used to describe a target dish that is expected to be obtained by executing the target cooking task, the ingredient information is used to describe a target ingredient required to obtain the target dish, and the cooking mode is used to process the target ingredient, and the target actual data is used to describe a current state of a cooking appliance that supports the target cooking task; Water volume prediction module: used to take the target expected data and the target actual data as input, and output the corresponding water volume prediction result using the target prediction model; Among them, the target prediction model is obtained by training based on multiple sample groups and adjusting the parameters of the preset model during training, each of the sample groups corresponds to a preset cooking task, the sample group is composed of sample expected data and sample actual data indicating the preset cooking task, and the sample group is marked with a baseline required water volume for the preset cooking task, and the baseline required water volume is used to describe the demand for water used as a heat transfer medium to complete the preset cooking task.

9. A cooking device, characterized in that: The invention comprises a cooking appliance and the water amount determination device according to claim 8, wherein the water amount determination device determines whether to generate an instruction to add water to the cooking appliance based on a water amount prediction result.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the water volume determination method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Steam box, control method and system thereof and medium

    CN115227111A

  • Method and device for predicting food weight according to cooking utensil and related equipment

    CN115271155A

  • Water supply control method and device, electronic equipment and computer readable medium

    CN116562600A

  • Cooking water tank water level processing method and device, electronic equipment and storage medium

    CN116602550A

  • Control method for cooking electrical appliance, cooking electrical appliance, and readable storage medium

    WO2021218083A1