Cooking utensil control method and device, equipment and storage medium

Abstract

The invention relates to the field of cooking control, in particular to a control method and device of a cooking utensil, equipment and a storage medium. The method comprises the following steps: identifying weight monitoring parameters of food materials in the cooking utensil, carrying out multi-time-point water content attenuation analysis on the food materials, and constructing a real-time water content attenuation curve; identifying real-time cooking environment state parameters, performing quantitative correlation analysis according to the real-time water content attenuation curve, and constructing a water content change quantitative correlation model; acquiring a monitoring image in the cooking utensil, and performing deep target visual identification and personalized taste preference mining to generate personalized taste features of a multi-food-material user; and performing real-time cooking state identification according to the real-time water content attenuation curve and the water content change quantitative correlation model, and performing rolling time window water attenuation prediction to obtain water attenuation prediction values of a plurality of time windows. The cooking utensil parameters are dynamically adjusted based on the moisture change in the cooking process, and therefore the cooking effect and quality are improved.

Description

Technical Field

[0001] The present invention relates to the field of cooking control, and in particular to a control method, device, equipment and storage medium for a cooking appliance. Background Art

[0002] As modern people's requirements for healthy diet and cooking experience continue to increase, innovation in cooking technology has gradually become an important direction for improving the quality of life. Traditional cooking methods rely on manual experience and simple mechanical control. Although these methods can meet daily cooking needs, they are often unable to accurately control the cooking process of ingredients, especially in the regulation of ingredient moisture, making it difficult to achieve the best taste and nutrient retention. With the rapid development of artificial intelligence, big data, and the Internet of Things, intelligent cooking has gradually become an important trend in future kitchens. Intelligent control methods for cooking appliances based on changes in ingredient moisture content have become an effective way to solve this problem with their precise moisture control and personalized cooking experience.

[0003] In traditional cooking, moisture is a key factor affecting the taste and nutritional value of ingredients. Different types of ingredients experience significant variations in the amount of moisture lost during cooking, which not only directly impacts the flavor but also determines the preservation of nutrients. Traditional cooking utensils are unable to precisely adjust to the moisture changes of different ingredients, often resulting in overcooking or excessive moisture loss, resulting in a loss of taste and nutrients. Therefore, accurately monitoring and regulating moisture changes in ingredients during cooking is crucial for improving cooking results and ensuring food quality and nutritional value.

[0004] Existing cooking equipment typically controls cooking based on simple parameters like temperature and time, lacking the ability to monitor and adjust moisture levels in ingredients in real time. While some high-end devices incorporate temperature and humidity sensors, their control mechanisms often rely on empirical rules or preset patterns, making it difficult to dynamically adjust to the varying needs of specific ingredients. This static control approach often fails to meet users' high standards for personalized taste and nutritional value, especially when cooking multiple ingredients simultaneously, where control accuracy and coordination pose significant challenges. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention proposes a cooking appliance control method, device, equipment and storage medium to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a method for controlling a cooking appliance, comprising the following steps: Step S1: Identify the weight monitoring parameters of the food in the cooking appliance, perform a multi-point moisture content decay analysis on the food, and construct a real-time moisture content decay curve; Step S2: Identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and construct a quantitative correlation model for moisture content changes; Step S3: Obtain monitoring images inside the cooking appliance, perform deep target visual recognition and personalized taste preference mining to generate personalized taste characteristics for multiple ingredients; Step S4: performing real-time cooking state identification based on the real-time moisture content decay curve and the moisture content change quantitative correlation model, and performing rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; Step S5: Calculating the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjusting cooking parameters based on the moisture attenuation prediction value, thereby generating dynamically adjusted cooking parameters; Step S6: Perform instant cooking control based on dynamically adjusted cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

[0007] The present invention further provides a cooking appliance control device for executing the cooking appliance control method described above, comprising: The moisture content decay curve module is used to identify the weight monitoring parameters of the ingredients in the cooking utensils, perform multi-point moisture content decay analysis on the ingredients, and construct a real-time moisture content decay curve; The quantitative correlation module is used to identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and build a quantitative correlation model for moisture content changes; An image recognition module is used to obtain monitoring images from cooking utensils, perform deep target visual recognition, and mine personalized taste preferences to generate personalized taste characteristics for multiple ingredients. The moisture decay prediction module is used to identify the real-time cooking state based on the real-time moisture decay curve and the quantitative correlation model of moisture content change, and to perform rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; A cooking parameter adjustment module, configured to calculate the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjust the cooking parameters based on the moisture decay prediction value, thereby generating dynamically adjusted cooking parameters; The cooking control module is used to perform instant cooking control based on dynamic adjustment of cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

[0008] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned cooking appliance control methods when executing the computer program.

[0009] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any one of the above-mentioned methods for controlling a cooking appliance.

[0010] The beneficial effects of the present invention include: By using high-precision sensors to monitor ingredient weight in real time, the system can accurately capture the rate of moisture loss in ingredients. Weight loss is positively correlated with water evaporation, and real-time data provides a reliable basis for subsequent moisture management. By collecting data at multiple time points, a dynamic moisture loss curve is constructed. This curve allows the system to promptly understand the changing trends in ingredient moisture, thereby preventing excessive moisture loss that can lead to dryness, hardness, or burnt texture. Using the real-time moisture loss curve, the system can tailor the moisture loss patterns of different ingredients to ensure that each ingredient maintains optimal moisture levels throughout the cooking process. By monitoring cooking environment parameters such as temperature, humidity, and heat, the system can identify the impact of environmental factors on changes in ingredient moisture content, thereby providing more precise control strategies. By combining the real-time moisture loss curve with environmental parameters, the system can quantify the relationship between various environmental factors and moisture changes. This quantitative correlation model enables the system to intelligently analyze and predict changes in ingredient moisture under different cooking environments, thereby optimizing cooking settings. Based on this model, the system can accurately predict changes in ingredient moisture under different cooking conditions and adjust cooking environment parameters such as temperature and humidity to achieve optimal moisture control. Using high-resolution image acquisition and deep vision recognition technology, the system can automatically identify the type, shape, and size of ingredients within cooking utensils. This enables the system to accurately classify and extract features for each ingredient, providing a basis for personalized control. Based on the user's cooking preferences, the system can perform personalized taste analysis based on the type of ingredient, cooking method, and user-defined texture requirements (such as "tender" or "crisp"). This extends cooking beyond temperature and humidity control to optimize the details of each dish based on user preferences. Leveraging user-specific taste characteristics, the system can more accurately predict and adjust the moisture content and texture of ingredients, ensuring each cooking session meets user expectations and providing a higher-quality, personalized cooking experience. By analyzing the real-time moisture decay curve of ingredients and the cooking environment, the system can automatically identify cooking stages (such as boiling, steaming, and stewing) and implement the most appropriate cooking strategies for each stage. Using a rolling window prediction algorithm, the system can predict moisture decay values over multiple future time windows. By predicting future trends, the system can proactively respond to prevent excessive moisture loss and maximize the preservation of the taste and nutrients of ingredients. This predictive function significantly improves moisture control accuracy during cooking, avoiding poor taste caused by premature or delayed adjustments, thereby improving overall cooking quality. By combining ingredient type, user taste preferences, and real-time moisture decay prediction, the system calculates the optimal moisture content for each ingredient. This ensures that each ingredient is cooked at the ideal moisture content, ensuring optimal taste.In multi-ingredient cooking scenarios, the system dynamically adjusts cooking parameters (such as heat, temperature, and humidity) based on the individual needs of each ingredient. This collaborative optimization mechanism ensures that the moisture content of each ingredient is properly controlled during the cooking process, preventing some ingredients from becoming overly dry or overly moist. Based on moisture decay prediction, the system can adjust cooking appliance parameters in real time to ensure precise control of the moisture content of each ingredient throughout the cooking process, preventing uneven heating due to human factors. By dynamically adjusting cooking parameters, the system can quickly respond to changes during the cooking process and automatically adjust cooking conditions to ensure optimal cooking of each ingredient at every moment. The system iteratively optimizes cooking parameters based on actual cooking results and the final moisture content of the ingredients. This intelligent feedback mechanism adjusts parameters based on user feedback and cooking deviations, thereby improving cooking stability and accuracy. Through continuous parameter iteration and optimization, the system accumulates more experience and data, gradually refining the intelligent cooking model and enhancing cooking results, making the system more intelligent and accurate over time. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic flow chart of the steps of a method for controlling a cooking appliance according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Schematic diagram of the detailed implementation steps of step S3. DETAILED DESCRIPTION

[0012] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0013] This application provides a cooking appliance control method, device, equipment, and storage medium. The execution entities of the cooking appliance control method, device, equipment, and storage medium include, but are not limited to, the following: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc., which can be considered as general computing nodes of this application. The data processing platform includes, but is not limited to, at least one of an audio and image management system, an information management system, and a cloud data management system.

[0014] See also Figures 1 to 4 The present invention provides a method for controlling a cooking appliance, the method comprising the following steps: Step S1: Identify the weight monitoring parameters of the food in the cooking appliance, perform a multi-point moisture content decay analysis on the food, and construct a real-time moisture content decay curve; Step S2: Identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and construct a quantitative correlation model for moisture content changes; Step S3: Obtain monitoring images inside the cooking appliance, perform deep target visual recognition and personalized taste preference mining to generate personalized taste characteristics for multiple ingredients; Step S4: performing real-time cooking state identification based on the real-time moisture content decay curve and the moisture content change quantitative correlation model, and performing rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; Step S5: Calculating the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjusting cooking parameters based on the moisture attenuation prediction value, thereby generating dynamically adjusted cooking parameters; Step S6: Perform instant cooking control based on dynamically adjusted cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

[0015] By using high-precision sensors to monitor ingredient weight in real time, the system can accurately capture the rate of moisture loss. Weight loss is positively correlated with water evaporation, and real-time data provides a reliable basis for subsequent moisture management. By collecting data at multiple time points, a dynamic moisture loss curve is constructed. This curve allows the system to promptly understand the changing trends of ingredient moisture, thereby preventing excessive moisture loss that can lead to dryness, hardness, or burnt texture. Using the real-time moisture loss curve, the system can tailor the moisture loss patterns of different ingredients to ensure that each ingredient maintains optimal moisture levels throughout the cooking process. By monitoring cooking environment parameters such as temperature, humidity, and heat, the system can identify the impact of environmental factors on changes in ingredient moisture content, thereby providing more precise control strategies. By combining the real-time moisture loss curve with environmental parameters, the system can quantify the relationship between various environmental factors and moisture changes. This quantitative correlation model enables the system to intelligently analyze and predict changes in ingredient moisture content under different cooking conditions, thereby optimizing cooking settings. Based on this model, the system can accurately predict changes in ingredient moisture content under different cooking conditions and adjust cooking environment parameters such as temperature and humidity to achieve optimal moisture control. Using high-resolution image acquisition and deep vision recognition technology, the system can automatically identify the type, shape, and size of ingredients within cooking utensils. This enables the system to accurately classify and extract features for each ingredient, providing a basis for personalized control. Based on the user's cooking preferences, the system can perform personalized taste analysis based on the type of ingredient, cooking method, and user-defined texture requirements (such as "tender" or "crisp"). This extends cooking beyond temperature and humidity control to optimize the details of each dish based on user preferences. Leveraging user-specific taste characteristics, the system can more accurately predict and adjust the moisture content and texture of ingredients, ensuring each cooking session meets user expectations and providing a higher-quality, personalized cooking experience. By analyzing the real-time moisture decay curve of ingredients and the cooking environment, the system can automatically identify cooking stages (such as boiling, steaming, and stewing) and implement the most appropriate cooking strategies for each stage. Using a rolling window prediction algorithm, the system can predict moisture decay values over multiple future time windows. By predicting future trends, the system can proactively respond to prevent excessive moisture loss and maximize the preservation of the taste and nutrients of ingredients. This predictive function significantly improves moisture control accuracy during cooking, avoiding poor taste caused by premature or delayed adjustments, thereby improving overall cooking quality. By combining ingredient type, user taste preferences, and real-time moisture decay prediction, the system calculates the optimal moisture content for each ingredient. This ensures that each ingredient is cooked at the ideal moisture content, ensuring optimal taste.In multi-ingredient cooking scenarios, the system dynamically adjusts cooking parameters (such as heat, temperature, and humidity) based on the individual needs of each ingredient. This collaborative optimization mechanism ensures that the moisture content of each ingredient is properly controlled during the cooking process, preventing some ingredients from becoming overly dry or overly moist. Based on moisture decay prediction, the system can adjust cooking appliance parameters in real time to ensure precise control of the moisture content of each ingredient throughout the cooking process, preventing uneven heating due to human factors. By dynamically adjusting cooking parameters, the system can quickly respond to changes during the cooking process and automatically adjust cooking conditions to ensure optimal cooking of each ingredient at every moment. The system iteratively optimizes cooking parameters based on actual cooking results and the final moisture content of the ingredients. This intelligent feedback mechanism adjusts parameters based on user feedback and cooking deviations, thereby improving cooking stability and accuracy. Through continuous parameter iteration and optimization, the system accumulates more experience and data, gradually refining the intelligent cooking model and enhancing cooking results, making the system more intelligent and accurate over time.

[0016] In the embodiment of the present invention, see Figure 1 , is a schematic flow chart of the steps of a cooking appliance control method of the present invention. In this example, the cooking appliance control method includes the following steps: Step S1: Identify the weight monitoring parameters of the food in the cooking appliance, perform a multi-point moisture content decay analysis on the food, and construct a real-time moisture content decay curve; In this embodiment, a high-precision weight sensor is installed in the cooking appliance to ensure accurate monitoring of ingredient weight changes. When selecting a sensor, consider its range, resolution, and stability. A sensor with an accuracy of 0.1 gram is generally recommended to accommodate the weight range of different ingredients. An electronic scale sensor with a maximum range of 5 kg and an accuracy of 0.1 gram is installed to monitor ingredient weight changes in real time during cooking. The sensor should be installed in a location that ensures it can directly support the weight of the ingredients and is not affected by external factors. A data acquisition system is configured to ensure the sensor can record weight data in real time. The data acquisition system should have high-frequency sampling capabilities, including sampling once per second, to capture subtle changes in ingredients during cooking. By connecting the sensor to a central processing unit or microcontroller, weight data is transmitted to a computer or mobile device in real time, ensuring continuous and accurate data collection. Before cooking begins, the initial weight of the ingredients is recorded. This data serves as a baseline for subsequent analysis to help determine moisture loss during cooking. For example, if 1000 grams of chicken is placed at the beginning of cooking, the system will record this initial weight as a baseline for subsequent calculation of moisture content changes. Collect ingredient weight data regularly during the cooking process. Recording weight every 5 minutes is recommended to track weight changes during cooking. For example, the recorded weight of chicken at 5 minutes is 950 grams, and at 10 minutes it is 900 grams. These data will be used for subsequent moisture decay analysis. Based on the recorded weight data, calculate the moisture content of the ingredient at each time point. The difference between the initial weight and the weight at each time point can be used to infer moisture loss. If the recorded weight at 5 minutes is 950 grams, the moisture loss is 100 grams (1000 grams - 950 grams). The moisture content at this point can be calculated as: Moisture content = Current weight / Initial weight × 100% = 950 grams / 1000 grams × 100% = 95%. Combine the moisture content data from all time points to create a database containing time, weight, and moisture content. Each record should include a timestamp, current weight, and calculated moisture content for subsequent analysis. The recording format is: Timestamp: 00:05, Current Weight: 950g, Moisture Content: 95%; Timestamp: 00:10, Current Weight: 900g, Moisture Content: 90%; and so on. Visualize the collected moisture content data to more intuitively demonstrate moisture decay trends. Use charting software or programming tools to generate a graph showing the relationship between time and moisture content. Use a scatter plot or line chart with time as the horizontal axis and moisture content as the vertical axis to clearly display the change in moisture content over time. Apply a curve fit to the visualized data to obtain a smoother moisture decay curve. This process helps identify decay trends and predict future moisture loss.Commonly used fitting methods include exponential regression or polynomial regression. The appropriate fitting function is selected according to the characteristics of the data to ensure that the fitting curve can accurately reflect the actual situation.

[0017] Step S2: Identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and construct a quantitative correlation model for moisture content changes; In this embodiment, multiple environmental sensors are installed in the cooking appliance, including temperature, humidity, and air pressure sensors. These sensors should be selected based on their accuracy, response time, and adaptability. The temperature sensor should accurately measure temperatures between -10°C and 200°C, while the humidity sensor should be able to detect relative humidity between 0% and 100%. A digital temperature sensor with an accuracy of ±0.5°C is used to continuously monitor ambient temperature changes during cooking. Meanwhile, the humidity sensor should have an accuracy of ±2%RH to provide real-time feedback on the humidity level of the cooking environment. A data acquisition system is configured to record environmental parameters in real time. The acquisition frequency should be set to once per second to ensure that rapidly changing environmental conditions are captured. This data will be used for subsequent analysis and model building. The temperature sensor is set to record temperature data once per second during cooking and store it in a database. Assuming that the temperature is 80°C and the humidity is 60% at the 5th minute of cooking, this data will be recorded and timestamped. The collected environmental parameters are integrated with the ingredient weight and moisture content data to form a comprehensive dataset. Each record should include information such as time, temperature, humidity, air pressure, ingredient weight, and current moisture content for subsequent analysis. Select appropriate statistical analysis methods for quantitative correlation analysis. Common methods include the Pearson correlation coefficient, Spearman's rank correlation coefficient, and multiple regression analysis. When selecting a method, consider the distributional characteristics and correlation of the data. Use multiple regression analysis to evaluate the effects of temperature, humidity, and air pressure on ingredient moisture content. Set moisture content as the dependent variable and temperature, humidity, and air pressure as independent variables to construct a model and analyze the impact of each factor. Preprocess the integrated dataset to ensure data integrity and accuracy. This includes addressing missing values, outliers, and standardizing the data to facilitate subsequent analysis. Identify any missing records in the dataset. If humidity data is missing at a specific time, interpolation can be used to fill in the missing values to ensure data continuity and consistency. Analyze the processed data using the selected statistical analysis method to construct a quantitative correlation model for moisture content changes. Evaluate the model's performance using the goodness of fit (such as the R² value) to ensure that the model accurately reflects changes in moisture content. Through model evaluation, determine the impact of each factor. The model is validated using cross-validation or holdout methods to ensure its generalization ability. The dataset is divided into training and test sets, and the model's performance on unseen data is evaluated to verify its accuracy. Using 70% of the data as the training set and 30% as the test set, the model's effectiveness is evaluated by comparing the error between the predicted water content and the actual value, calculating metrics such as the root mean square error (RMSE). The results of the quantitative correlation model are recorded in a database, including model parameters, goodness of fit, and validation results. This record provides a basis for subsequent dynamic adjustment and intelligent control.Record the model form and parameters in a format such as: Model equation: Moisture content = 0.5 + 0.2 × temperature + 0.1 × humidity, R² = 0.85, RMSE = 2.5. Make sure to integrate all information for later reference and application.

[0018] Step S3: Obtain monitoring images inside the cooking appliance, perform deep target visual recognition and personalized taste preference mining to generate personalized taste characteristics for multiple ingredients; In this embodiment, a high-resolution camera is installed in the cooking appliance to capture images of the cooking process in real time. When selecting a camera, its resolution, frame rate and low-light performance should be considered to ensure that clear images can be provided under different cooking lighting conditions. A camera with 1080p resolution and 30fps is selected to ensure that clear images can be captured every second during the cooking process. The installation location should ensure that the camera can cover the entire cooking area to capture the status of all ingredients. An image acquisition system is set up to ensure that the camera can transmit image data to the processing unit in real time. The system should have a high-frequency acquisition capability, usually set to capture one frame of image per second to track the changes in ingredients. At the 5th minute of cooking, the system will automatically capture and store the current image, record the timestamp and related status information, for subsequent analysis and processing.

[0019] Store real-time image data in a database to ensure easy subsequent querying and analysis. Each image should be associated with information such as a timestamp, ingredient type, and current state to form a complete dataset. Recording formats can be: timestamp: 00:05, image file name: image_0001.jpg, ingredient state: uncooked; this ensures data integrity and traceability. Select an appropriate deep learning object detection model for visual identification of ingredients. Commonly used models include YOLO (You Only Look Once), Faster R-CNN, and Mask R-CNN, which can identify multiple objects in images in real time. Use the YOLO v4 model, which offers strong real-time performance and accuracy, enabling rapid detection and identification of ingredients in images. Before use, train the deep learning model using a well-labeled dataset of ingredient images to ensure accurate identification of different types of ingredients. Hyperparameter tuning is required during training to optimize model performance. Using a training set containing thousands of ingredient images, a learning rate of 0.001 and a batch size of 32 were set, and training was performed for 50 epochs to improve model recognition accuracy. Real-time cooking images were fed into the trained deep learning model to perform object detection. The model identified each ingredient in the image and generated a bounding box and corresponding label for each ingredient. For example, in an image taken at the 10th minute, the model successfully identified ingredients such as chicken, carrots, and onions, generating corresponding bounding boxes and labels. This provides a foundation for subsequent personalized texture preference mining. Data is collected based on users' historical cooking records and preferences to understand their desired textures for different ingredients. This can be obtained through in-app questionnaires and user feedback. Users can indicate their preferences for chicken in the app, such as "tender," "juicy," or "charred." The system records these preferences for subsequent analysis. By combining the ingredient information identified by the deep learning model with the user's texture preference data, personalized texture characteristics for each ingredient are extracted. This process is achieved by analyzing the relationship between user feedback and actual cooking results. If a user's preference for chicken is "tender and smooth," the system will extract relevant cooking parameters, such as slow cooking at low temperatures and maintaining appropriate moisture, to create a personalized taste profile for the chicken. This extracted personalized taste profile is then incorporated into the user's taste profile for future reference during cooking. Each ingredient's taste profile should include information such as the optimal cooking method, temperature range, and time. The resulting profile would be: Ingredient: Chicken, Taste Profile: Tender and Smooth, Optimal Cooking Method: Slow Cooking at Low Temperatures, Recommended Moisture: 65%. This ensures the system can provide more personalized recommendations for subsequent cooking.

[0020] Step S4: performing real-time cooking state identification based on the real-time moisture content decay curve and the moisture content change quantitative correlation model, and performing rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; In this embodiment, during the cooking process, the system compares the real-time collected ingredient weight, temperature, humidity, and other environmental parameters with a previously constructed moisture content decay curve. Using this data, the system can identify the ingredient's current cooking state in real time. For example, at the 10th minute, the temperature is 80°C, the humidity is 55%, and the ingredient weighs 900 grams. The system compares these parameters with the previous decay curve to determine the current state. Based on a quantitative correlation model for moisture content variation, the system calculates the current moisture content of the ingredient and compares it with the optimal moisture content. If the current moisture content is lower than the optimal value, the ingredient's cooking state is determined to be "overdry," while if it is lower, the cooking state is determined to be "suitable" or "overwet." If the model calculates the current moisture content to be 50% and the optimal moisture content is 65%, the system identifies the chicken's state as "overdry" and prepares to dynamically adjust the state. The real-time identified cooking state is recorded in a database for subsequent analysis and adjustment. The record should include a timestamp, ingredient type, current moisture content, identified state, and recommended adjustment measures. A rolling window is defined, for example, with a 5-minute window. Based on current moisture content and environmental parameters, the system collects and organizes past moisture content data for prediction. At the 10th minute, the system analyzes data from the previous 5 minutes (e.g., moisture content and weight at the 5th minute) to predict moisture change over the next 5 minutes. Using a previously constructed quantitative correlation model for moisture content change, combined with real-time monitoring data, the system predicts future moisture decay. Model inputs include current temperature, humidity, the initial moisture content of the ingredients, and previous moisture decay curve data. The system predicts moisture loss over the next 5 minutes based on a current temperature of 80°C and a humidity of 55%. The model outputs are used to calculate moisture decay predictions for multiple time windows. Based on the current state and the prediction model, the system estimates moisture loss in each time window to formulate appropriate cooking adjustments. For example, if the model predicts a moisture decay of 10 grams over the next 5 minutes, the system will record a predicted moisture content of 40% for the next time window (the 15th minute) (assuming an initial moisture content of 50%). Record the predicted moisture decay values for each time window in the system to ensure effective adjustments and analysis later. The record should include the time window, predicted moisture decay value, and related parameters. The record format is: Time window: 00:10-00:15, predicted moisture decay: 10 grams, expected moisture content: 40%. This data will help the system dynamically adjust cooking parameters to optimize the cooking results.

[0021] Step S5: Calculating the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjusting cooking parameters based on the moisture attenuation prediction value, thereby generating dynamically adjusted cooking parameters; In this example, previously collected user feedback and cooking records are used to extract personalized taste characteristics for each ingredient. These characteristics should include user preferences for ingredients (such as tender, juicy, and charred), as well as their optimal cooking conditions (such as temperature, time, and moisture content). For example, if a user prefers "tender" chicken, the system will record this preference and associate it with the corresponding optimal moisture content. By analyzing historical data, the optimal moisture content for chicken is determined to be 65%. The personalized taste characteristics of different ingredients are integrated into a database to facilitate subsequent calculations and dynamic adjustments. This database should include information such as each ingredient's name, optimal moisture content, and optimal cooking method and time. The characteristics of ingredients such as chicken (optimal moisture content 65%), carrots (optimal moisture content 75%), and onions (optimal moisture content 70%) are integrated to form a comprehensive taste characteristic database. The integrated personalized taste characteristics are recorded in the system database to ensure data accuracy and traceability. User feedback can be used to verify that the recorded characteristics meet the user's actual taste requirements. The record format is: Ingredient: Chicken, Optimal Moisture Content: 65%, Optimal Cooking Method: Low-Slow Cooking. Each record is timestamped and includes user feedback for easy verification. Based on the individual taste characteristics of each ingredient, a previously established quantitative correlation model for moisture content variation is applied to calculate the optimal moisture content for each ingredient under specific cooking conditions. The model inputs the temperature and humidity data of the current cooking environment and calculates the optimal moisture content of the chicken under these conditions. Assuming the current temperature is 80°C and the humidity is 55%, the model outputs the optimal moisture content for chicken as 65%. Considering that different ingredients may require different amounts of moisture during cooking, the system dynamically adjusts the optimal moisture content for each ingredient based on the actual cooking state (such as the current moisture content and user preferences). If the current moisture content of the chicken is 50%, the system calculates the amount of moisture that needs to be added to ensure the optimal moisture content by the end of cooking and recommends appropriate adjustments. The calculated optimal moisture content and its adjustment recommendations are recorded in the system for subsequent operations and user reference. Each record should include the ingredient type, current moisture content, calculated optimal moisture content, and adjustment recommendations. Incorporate previous moisture decay predictions into the dynamic adjustment decision-making process. These predictions will help the system assess how the moisture content of ingredients will change over time and adjust cooking parameters accordingly. Suppose it is predicted that the moisture content of the chicken will decrease by 10 grams in the next 5 minutes. The system needs to take this into account when adjusting cooking parameters. Design a dynamic adjustment strategy based on the calculated optimal moisture content and moisture decay predictions. If the current moisture content of the chicken is lower than the optimal value and is predicted to continue to decrease in a short period of time, the system will recommend lowering the cooking temperature or increasing the humidity. Suppose the system recommends lowering the cooking temperature of the chicken from 80°C to 75°C and increasing steam to retain moisture, thereby ensuring that the chicken reaches the optimal moisture content of 65% by the end of cooking.During the cooking process, the system will adjust cooking parameters in real time based on dynamic adjustment strategies and continuously monitor the state of the ingredients. Each parameter change should be recorded for subsequent analysis and optimization. The record format is: Timestamp: 00:15, Ingredient: Chicken, Adjusted Temperature: 75°C, Adjusted Humidity: Increased Steam. Ensure that all adjustments are well-founded to optimize the subsequent cooking process.

[0022] Step S6: Perform instant cooking control based on dynamically adjusted cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

[0023] In this embodiment, during the cooking process, the system continuously monitors the state of the ingredients, including key parameters such as weight, temperature, and moisture content. Using installed sensors, the system acquires this data in real time and adjusts based on the set optimal parameters. For example, suppose the current temperature of the chicken is monitored to be 75°C and the moisture content is 60%. The system will provide real-time feedback and make adjustments based on the previously calculated optimal moisture content (65%) and user preferences. Based on this real-time monitoring data, the system automatically adjusts cooking parameters such as temperature, time, and humidity. If the moisture content of the ingredients is detected to be below the optimal value, the system will increase steam or lower the temperature to prevent excessive evaporation. If the current moisture content of the chicken is 60%, the system will recommend adjusting the cooking temperature from 75°C to 72°C and increasing steam to ensure that the moisture content remains within the ideal range. After implementing these dynamic adjustments, the system will continuously monitor the state of the ingredients to ensure the effectiveness of the adjustments. The status data after each adjustment will be recorded for subsequent analysis and optimization. The recorded adjusted state is: Timestamp: 00:15, Ingredient: Chicken, Adjusted Temperature: 72°C, Current Moisture Content: 62%. This allows the system to respond to changes in ingredients in real time. During the cooking process, the system collects data after each adjustment, including the current state of the ingredient and user feedback. This data is used for subsequent parameter optimization and model iteration. The collected data includes the temperature, humidity, and moisture content after each adjustment, as well as user feedback on taste. Suppose, after cooking, a user reports that the chicken is not tender enough. The system will record this information for analysis. Using this real-time data and user feedback, the system will analyze the impact of different parameters on the taste of the ingredient and identify the optimal cooking conditions. By building a regression model or machine learning algorithm, the system can learn the patterns of ingredient changes under different conditions. By analyzing data from past cooking cycles, the system may discover that chicken retains more moisture at 72°C and has a taste that is closer to the user's preference, prompting the system to prioritize this parameter in future cooking cycles. Ultimately, the system will build an intelligent cooking parameter control model based on this analysis. This model will comprehensively consider factors such as ingredient type, user preferences, and real-time environmental parameters to achieve intelligent cooking control. The intelligent control model might set a rule: During the chicken cooking process, as long as the moisture content falls below 65%, the temperature will be automatically adjusted to 70°C and the steam will be activated. This ensures that the ingredients are kept in optimal condition under all circumstances. All adjustments and feedback will be recorded in detail in the system database, forming a complete cooking process profile. This record will include information such as time, ingredient type, all adjustment parameters, and user feedback. The record format is: Timestamp: 00:20, Ingredient: Chicken, Final Moisture Content: 65%, User Feedback: Tender and Smooth. This data will help the system make more precise adjustments during future cooking processes.The system reviews data after each cooking session, analyzing the effectiveness of various parameters and user satisfaction for continuous iterative optimization. By continuously updating the model, the system can gradually improve cooking results and ensure that users' taste preferences are met. If a user reports that chicken is not tender enough when cooked at 80°C, the system will adjust the model accordingly and lower the recommended cooking temperature in the future to improve user satisfaction.

[0024] In this embodiment, refer to Figure 2 , is a flowchart of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Identify food weight monitoring parameters in cooking utensils based on high-precision weight sensors; Calculating the initial food weight based on the food weight monitoring parameters; Calculate the weight change of the food weight monitoring parameters during the cooking process based on the initial food weight, and mark it as the food moisture content change value; Performing a multi-time-point moisture content decay analysis on the moisture content change value of the food to generate a moisture content decay rate and decay amplitude; The water content decay rate and decay amplitude are subjected to multi-time point decay fitting to construct a real-time water content decay curve.

[0025] In this embodiment, a suitable high-precision weight sensor is selected to ensure it can accurately measure the weight of the ingredients within the cooking appliance. The sensor must have sufficient resolution and accuracy, typically with an accuracy of less than 0.1 gram, to meet the requirements for monitoring weight changes during the cooking process. The selected sensor should be able to accurately measure within the range of 0-5 kilograms, ensuring accurate data for the weight changes of different ingredients. The weight sensor should be properly installed within the cooking appliance to ensure good contact with the ingredients and the ability to independently record weight changes. After installation, the sensor needs to be calibrated to ensure that it accurately displays zero at zero time to eliminate device errors. During calibration, standard weights can be used to test the sensor to ensure that the sensor accurately reflects the actual weight under different loads. Calibration parameters should also be recorded for subsequent data analysis. At the beginning of cooking, real-time data from the weight sensor is collected, and the initial weight of the ingredients is recorded. This process requires continuous monitoring to capture the weight changes of the ingredients at every moment during the cooking process. An initial weight of 1000 grams is recorded at the beginning of cooking, and the weight is recorded every 5 minutes thereafter to obtain detailed data on the weight changes of the ingredients. Based on the data recorded in the previous step, calculate and determine the initial ingredient weight. This weight will serve as a baseline for subsequent analysis, allowing tracking of changes in the ingredients during the cooking process. The initial ingredient weight must be accurately determined. For example, if the recorded initial weight is 1000 grams, this value will serve as the basis for subsequent calculations. The collected weight data is cleaned to remove outliers and noise to ensure accurate calculations. A reasonable threshold range can be set to filter out valid data. If, during monitoring, a weight of 950 grams is recorded and shows a significant change compared to the previous and subsequent data, it should be considered an outlier and removed. The initial ingredient weight and related parameters are recorded in a data storage system to ensure that this baseline data can be easily referenced in subsequent calculations and analyses. The initial ingredient weight of 1000 grams, along with information such as the recording time and sensor status, is stored in a database for easy access. During the cooking process, the weight change of the ingredients is continuously monitored and compared with the initial ingredient weight to calculate the change in moisture content. If the recorded weight at a certain point during the cooking process is 800 grams, the water loss can be calculated as: initial weight 1000 grams - current weight 800 grams = 200 grams. The calculated weight change provides the change in moisture content of the ingredient. This change can be used to assess the dehydration of the ingredient during the cooking process. For example, if the recorded weight change is 200 grams throughout the cooking process, the moisture content of the ingredient can be inferred to have changed by 20% (based on the ratio of 200 grams of water to the initial 1000 grams). This moisture change is recorded and correlated with the weight data at each monitoring time point, allowing subsequent analysis to accurately reflect moisture content changes at different points in time.At the 15th minute of cooking, the moisture content of the ingredient changes to 200 grams. This data is stored along with the corresponding time, current weight, and other information. During the cooking process, the moisture content change values are continuously recorded at different time points to generate time series data. This process requires ensuring the accuracy and completeness of the data for subsequent analysis. The moisture content change is recorded every 5 minutes during the cooking process, resulting in a series of data points, such as: 200 grams at 0 minutes, 180 grams at 5 minutes, and 160 grams at 10 minutes. An appropriate mathematical model is selected to analyze the decay of the collected moisture content change data. Common methods include exponential decay models or linear regression analysis to describe the decay trend of the moisture content of the ingredient over time. Using the exponential decay model, the data can be fitted to the form: Y = Y0 * e^(-kt), where Y0 is the initial moisture content, k is the decay rate, and t is time. Based on the selected model, the decay rate and magnitude of the moisture content of the ingredient are calculated. This process requires fitting the data points to determine the model parameters to accurately reflect the actual situation. Assuming that the analysis indicates a decay rate k of 0.1, we can infer the moisture content trend over a certain period of time, such as a 10% moisture loss per minute. Based on the decay rate and amplitude calculated in the previous step, a moisture content decay curve is generated. This curve visually reflects the moisture changes in the ingredients during the cooking process. By plotting the relationship between time and moisture content change, the moisture loss of the ingredients during cooking can be clearly displayed. The generated moisture content decay curve is applied to a real-time monitoring system. By dynamically monitoring the moisture content changes of the ingredients, cooking parameters (such as temperature and time) can be adjusted in a timely manner to ensure optimal cooking results. If real-time monitoring indicates that the moisture content of the ingredients is decaying too quickly, adjustments can be made by lowering the temperature or shortening the cooking time to maintain the taste and nutritional value of the ingredients. Finally, the generated moisture content decay curve and its related parameters are recorded in a database for subsequent analysis and reference. This data will provide valuable reference for future cooking processes. The final decay curve is recorded, along with relevant parameters such as initial moisture content, final moisture content, and decay rate, to facilitate subsequent optimization of cooking strategies.

[0026] In this embodiment, refer to Figure 3 , is a flowchart of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Identify the real-time cooking environment status parameters of the current cooking appliance; Calculating the cooking temperature, humidity, and heat of the real-time cooking environment state parameters to generate a multi-dimensional cooking environment state feature; Perform time-series cooking state analysis on the multi-dimensional cooking environment state characteristics to generate cooking environment state characteristics at different time points; Based on the real-time moisture content decay curve and the cooking environment status characteristics at different time points, timestamp matching is performed, and quantitative correlation analysis is performed to generate a quantitative change relationship between cooking status and moisture content decay; The quantitative change relationship is dynamically modeled to construct a quantitative correlation model for water content change.

[0027] In this embodiment, the cooking appliance is equipped with multiple environmental sensors, including temperature sensors, humidity sensors, and a power monitoring device. These sensors must possess high precision and rapid response capabilities to ensure real-time monitoring of the cooking environment. A temperature sensor should accurately measure within the range of -10°C to 200°C. The humidity sensor should provide accurate readings within the relative humidity range of 0% to 100%. The power monitoring device should provide real-time feedback on flame intensity or electromagnetic wave power. During the cooking process, sensor data is collected in real time and the cooking environment parameters are recorded. This data includes the current cooking temperature, humidity, and power level. Collection is performed at regular intervals (e.g., 1 second) to ensure data timeliness and accuracy. For example, at the first minute of cooking, the temperature is 150°C, the humidity is 60%, and the power level is 80%. This data will provide a basis for subsequent analysis. By processing the sensor data, the real-time cooking environment parameters are calculated. The temperature, humidity, and power level values can be standardized to facilitate subsequent feature generation and time series analysis. The normalization formulas for temperature, humidity, and power level are: , the real-time monitored cooking state parameters are integrated into a multi-dimensional feature vector. This feature vector should contain the standardized values of temperature, humidity, and firepower, which can be expressed as Features=[Temperature,Humidity,FirePower].

[0028] If the current temperature is standardized to 0.75, the humidity is 0.6, and the heat level is 0.8, then the feature vector is Features = [0.75, 0.6, 0.8]. During the cooking process, the feature vectors are recorded at each time point to form time series data. This time series data will be used for subsequent state analysis to help understand changes in the cooking environment at different time points. During the cooking process, feature vectors are recorded every 5 minutes. Suppose the feature vectors recorded at the 5th and 10th minutes are [0.78, 0.65, 0.85] and [0.80, 0.60, 0.90], respectively. The recorded time series features are stored in the database with timestamps to ensure easy query and analysis. Each record should include a timestamp, feature vector, and related monitoring status. The stored record format is: Timestamp: 00:05, Feature Vector: [0.78, 0.65, 0.85]. Select an appropriate time series analysis method and use statistical analysis tools (such as autoregressive models and moving average models) to analyze the multi-dimensional features. This analysis helps identify patterns in how cooking environment conditions change over time. The recorded feature data is modeled using an autoregressive integrated moving average (ARIMA) model to capture the dynamic changes in temperature, humidity, and power over time. Based on the results of the time series analysis, trends in the cooking environment state characteristics at different time points are identified. This trend can be used to assess the impact of environmental changes on the moisture content of ingredients during the cooking process. For example, if the temperature steadily rises while the humidity gradually decreases over the first 15 minutes, this trend will provide important information for subsequent analysis. The analysis results are compiled into a report, ensuring that the cooking environment state characteristics at each time point are clearly recorded. This report provides a basis for analyzing the relationship between cooking conditions and moisture content decay. The state characteristics at the 10th minute are recorded as follows: temperature 0.80, humidity 0.60, and power 0.90. The real-time moisture content decay curve is timestamped with the cooking environment state characteristics at different time points to ensure accurate correspondence between the available data. By comparing timestamps, each feature vector is matched to its corresponding moisture content change data. If the moisture content decay recorded at the 10th minute is 200 grams, ensure that this data correctly corresponds to the cooking environment state characteristics at the 10th minute. Select an appropriate statistical analysis method (such as correlation analysis, regression analysis, etc.) to perform quantitative analysis on the matched data. This analysis will help identify the relationship between cooking status and moisture content decay.

[0029] The Pearson correlation coefficient was used to evaluate the linear relationship between cooking environment characteristics (temperature, humidity, and heat) and moisture content changes. The results of the quantitative correlation analysis were compiled into a report, ensuring that each analysis result was clearly presented. This report will provide an important basis for subsequent modeling. The correlation coefficient between temperature and moisture content change was found to be 0.85, indicating a significant positive correlation between the two. Based on the quantitative relationship obtained in the previous analysis, appropriate modeling methods (such as multiple linear regression and nonlinear regression) were selected to construct a quantitative correlation model for moisture content changes. This model will be used to predict moisture content trends under different cooking environment conditions. The constructed model was trained using the previously collected data and its accuracy was evaluated through methods such as cross-validation. This ensured that the model could effectively predict moisture content changes and had good generalization ability. If the prediction error of the trained model on the validation set was less than 5%, the model was considered valid. The final model and its parameters were recorded in a database to ensure easy query and application. This model will provide a scientific basis for intelligent control of cooking appliances. The coefficients of the model and their corresponding parameter values are recorded so that they can be used to dynamically adjust the cooking environment during the actual cooking process.

[0030] In this embodiment, refer to Figure 4 , is a flowchart of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Obtain monitoring images inside cooking appliances and user-preset cooking instructions; Perform multi-image region segmentation on the monitoring image inside the cooking appliance and calculate the local contrast; performing global adaptive differential sharpening according to the local contrast to construct an adaptively sharpened monitoring image; Perform deep target visual recognition on adaptively sharpened monitoring images and mark ingredients in multiple utensils; Perform visual type analysis on ingredients in multiple utensils and generate the type of each ingredient; Based on the user's preset cooking instructions, personalized taste preferences are mined for each type of ingredient to generate personalized taste characteristics for multiple ingredients.

[0031] In this embodiment, a high-resolution camera is installed within the cooking appliance to capture real-time images of the cooking process. When selecting a camera, consider its resolution, frame rate, and low-light performance to ensure clear images in various lighting conditions. A 1080p resolution camera with a 30 fps frame rate and low-light imaging support is selected to ensure a clear view of the ingredients during the cooking process. A user-friendly input interface is designed to allow users to easily preset cooking instructions. This can be achieved through a touchscreen, mobile app, or voice recognition system, allowing users to quickly enter their preferred cooking method, temperature, and time. Users can enter instructions through the app, such as "cook chicken to 70°C for 30 minutes." The system records these instructions for subsequent intelligent control and analysis. The captured monitoring images and user-entered cooking instructions are integrated and stored in a system database for subsequent analysis and processing. Ensure that each image capture and instruction input is accurately recorded and timestamped for easy traceability and retrieval. Recording formats can include timestamps, image file names, and user instruction content to form a complete cooking process dataset. The collected monitoring images undergo preprocessing, including noise removal, contrast enhancement, and brightness adjustment. This step aims to improve image quality for subsequent region segmentation and analysis. A Gaussian filter is used to denoise the image, followed by histogram equalization to increase image contrast, making the different ingredients more discernible within the image. Image segmentation algorithms (such as K-means clustering, region growing, or image watershed) are then applied to segment the monitoring images to identify and separate the individual ingredient regions. The chosen segmentation algorithm should be tailored to the image's characteristics and complexity. The K-means clustering algorithm is used to divide the image into multiple regions. By setting an appropriate K value (such as 3 or 4), different ingredient regions can be effectively distinguished. Local contrast is calculated for each segmented region. This is measured by the relationship between the standard deviation and mean of the pixel intensities within the region. Regions with high local contrast typically indicate clear edges and rich details. For example, if the standard deviation of the pixel intensities for a particular ingredient region is 50 and the mean is 150, the local contrast can be calculated as the ratio of the standard deviation to the mean, helping to determine the level of detail in that region. Select an appropriate global adaptive sharpening algorithm to enhance detail-rich areas of the image while avoiding noise enhancement. Sharpening can be performed based on local contrast using techniques such as the Laplacian operator or high-pass filtering. A threshold is set, and when the local contrast exceeds this threshold, a sharpening filter is applied to highlight important food details. Adaptive sharpening is performed on the monitored image based on the calculated local contrast. The sharpening intensity is dynamically adjusted based on the contrast of different areas, ensuring that image detail is preserved without introducing excessive noise.A higher sharpening factor is applied to areas with a local contrast of 0.3, while the sharpening intensity is reduced for areas with a contrast of 0.1 to ensure image quality. The processed, adaptively sharpened image is exported and saved as a new file for subsequent object visual recognition and analysis. The image should be saved in a lossless format to preserve processed details. The sharpened image file is saved as "sharpened_image.jpg" and the parameter settings used during the processing are recorded for later reference and adjustment. A deep learning model (such as YOLO, Mask R-CNN, or Faster R-CNN) is used for object visual recognition to identify multiple ingredients in the image. The selected model should have high accuracy and real-time processing capabilities to adapt to the dynamic cooking environment. A trained YOLO v4 model is used to quickly identify ingredients and utensils in the image, ensuring real-time feedback during the cooking process. The sharpened monitoring image is input into the selected deep learning model for object detection. The model will identify each ingredient in the image and generate a bounding box and corresponding label for each ingredient. Ingredients such as chicken, carrots, and onions are detected in the image, and corresponding bounding boxes and labels are generated to ensure accurate recognition. The identified ingredients and their types are recorded in the system database to ensure accurate recording and tracking of each ingredient. Recognition results should include information such as ingredient type, location coordinates, and confidence level. The record format might be: ingredient type: chicken, location coordinates: [(x1, y1), (x2, y2)], confidence level: 0.95. Based on the recognition results, visual type parsing is performed on each ingredient to ensure accurate classification and feature extraction. This analysis can be based on visual features such as color, texture, and shape. The color and texture of chicken are analyzed to ensure that different types of meat, such as chicken, beef, and pork, can be distinguished. Based on the user's preset cooking instructions, the personalized taste characteristics of each ingredient are discovered. This can be achieved by analyzing the user's historical preferences, ingredient combinations, and cooking methods. If the user's historical data indicates a preference for "tender and smooth chicken," a corresponding taste profile is generated for the identified chicken, such as "suitable for slow cooking at low temperatures to maintain tenderness." The discovered taste characteristics are organized into a personalized taste profile for the user, ensuring personalized recommendations and adjustments for future cooking. This profile should include information such as the optimal cooking method, taste preferences, and suitable seasonings for each ingredient. For example, a profile might record: Chicken - Taste Characteristics: Tender and Smooth, Best Cooking Method: Low-Cooking Slow, Recommended Seasoning: Olive Oil, Salt, and Pepper.

[0032] In this embodiment, step S4 includes the following steps: Recognize the real-time cooking status of ingredients on the adaptively sharpened monitoring image to obtain the real-time cooking status of ingredients; Analyze the current cooking stage of the food in real time based on the real-time moisture content decay curve to generate the current cooking stage of the food; performing moisture decay trend evolution on the current cooking stage according to a moisture content change quantitative correlation model to obtain a moisture decay trend of the current stage; A rolling time window moisture attenuation prediction is performed according to the moisture attenuation trend to obtain moisture attenuation prediction values for multiple time windows.

[0033] In this embodiment, the previously generated adaptively sharpened monitoring image is first input into a deep learning model to identify the real-time cooking status of ingredients. This image is sharpened to highlight edges and details, improving recognition accuracy. A convolutional neural network (CNN) model is used, which requires proper training to identify different food states, such as raw, cooked, or burnt. A large number of annotated images are used as a training dataset to ensure model accuracy and robustness. The model performs real-time recognition, analyzing the food features in the image and determining the cooking status based on the trained weights. The recognition process minimizes latency to achieve real-time monitoring. At the 10th minute of cooking, the model predicts the chicken's state as "medium-rare" and records the corresponding confidence level (e.g., 0.92) to ensure the reliability of the result. The real-time cooking status of the identified ingredients is recorded in a database to ensure that each recognition is reliable. This record should include information such as timestamp, food type, cooking status, and confidence level for subsequent analysis and backtracking. The record format is: Timestamp: 00:10, Ingredient: Chicken, State: Rare, Confidence: 0.92. The previously generated real-time moisture content decay curve is used to analyze the current cooking stage of the ingredient. This curve reflects the changes in moisture content during the cooking process, helping to determine the ingredient's current cooking stage. The current moisture content change of the ingredient is compared with the decay curve to find the moisture value corresponding to the current moment, thereby inferring the ingredient's cooking stage. The current cooking stage of the ingredient is determined based on the current moisture content value (e.g., 800 grams) and the data in the decay curve (e.g., 300 grams is "cooked"). Different thresholds can be set to define cooking stages, such as "raw," "half-cooked," "cooked," and "overcooked." If the current moisture content is 800 grams and the curve indicates that the moisture value is between "half-cooked" and "cooked," the current stage is determined to be "half-cooked." The analysis results of the current cooking stage are recorded in the database to ensure that the system can track the cooking progress of the ingredient. The record should include the timestamp, ingredient type, current stage, and the corresponding judgment basis. The record format is: Timestamp: 00:10, Ingredient: Chicken, Current Stage: Rare, Judgment Basis: Moisture Content 800g. Using the previously established quantitative correlation model for moisture content change, analyze the moisture decay trend for the current cooking stage. The model inputs the state parameters of the current stage and outputs the expected moisture decay trend. The model may consider the impact of multiple variables (such as temperature, humidity, and heat) on moisture decay to ensure accurate prediction of moisture changes. Based on the model output, analyze the moisture decay trend for the current stage. This includes calculating the moisture change over a future period, assessing the duration of the stage, and its impact on the food's taste. If the model outputs a moisture decay rate of 5g / minute for the current stage, it can be inferred that the moisture content will decrease by 50g over the next 10 minutes. The moisture decay trend analysis results are recorded in a database to facilitate monitoring and adjustment of the cooking process.Records should include the current moisture decay rate, trend, and subsequent impact. The record format is: Timestamp: 00:10, Ingredient: Chicken, Moisture Decay Trend: 5g / minute, Estimated Duration: 10 minutes. Set a rolling window (e.g., every 5 minutes) and evaluate the current moisture decay at the end of each window. This step aims to optimize prediction accuracy through continuous data updates. Moisture decay predictions are set every 5 minutes to allow for timely adjustments to cooking parameters. At the end of each window, a moisture decay prediction for the next window is calculated based on the current moisture decay trend and the moisture content variation model. This prediction is dynamically updated based on the moisture value and decay rate of the previous window. For example, if the current moisture value is 800g at the 10th minute and the decay rate is 5g / minute, the predicted moisture value at the 15th minute is 750g. The moisture decay prediction for each window is recorded in a database for subsequent analysis and adjustment. Records should include the timestamp, the current predicted moisture value, and the reference decay trend.

[0034] In this embodiment, the specific steps of step S5 are: Analyze the optimal demand state based on the personalized taste characteristics of multiple ingredients to obtain the optimal demand state for each ingredient; Calculating the optimal moisture content according to the optimal demand state to generate the optimal moisture content state of each food ingredient; Performing multi-objective collaborative cooking parameter planning according to the optimal moisture content state to obtain multi-objective collaborative cooking parameters; Extracting the latest water content of the food based on the food weight monitoring parameters; The cooking parameters are dynamically adjusted for the latest moisture content of the food according to the multi-objective collaborative cooking parameters and the moisture attenuation prediction value, thereby generating dynamically adjusted cooking parameters.

[0035] In this embodiment, the demand state of each ingredient is integrated based on the personalized taste characteristics previously entered by the user. The user may prefer "tender and smooth" for chicken and "crisp" for carrots. These characteristics will serve as the basis for optimized analysis. User historical feedback and preference data are collected to ensure that the user's expectations for each ingredient are fully reflected. This information can be obtained through questionnaires, application feedback, etc. to ensure the accuracy and comprehensiveness of the data. Based on the user's personalized taste characteristics, a demand state model is constructed to identify the optimal state of each ingredient at different cooking stages. Chicken needs to maintain a certain moisture and temperature during the cooking process to ensure a "tender and smooth" taste. Multivariate analysis methods, such as cluster analysis, are used to group similar taste characteristics to facilitate the identification of the optimal demand state. The optimal cooking temperature for chicken is determined to be 75°C, and the moisture content range is 60%-70%. Based on the constructed model, the optimal demand state is extracted for each ingredient. This process needs to combine user feedback and experimental data to ensure that the obtained state can truly reflect user expectations. The optimal requirements for chicken are determined to be "temperature 75°C, moisture content 65%," while those for carrots are "temperature 85°C, moisture content 75%." Based on the optimal requirements for each ingredient, a model for calculating optimal moisture content is developed. This model should take into account the characteristics of the ingredient, cooking method, and user preferences. Linear regression or other appropriate statistical models are used to correlate the ingredient characteristics with their corresponding optimal moisture content to ensure model accuracy. Using this model, the optimal moisture content for each ingredient is calculated. Based on user preferences and the physical properties of the ingredients, the required moisture content for each ingredient for optimal taste is determined. Based on the model calculations, the optimal moisture content for chicken is 65%, and for carrots is 75%. This ensures that these calculations reflect actual cooking results. The optimal moisture content for each ingredient is recorded in a database for subsequent analysis and adjustment. This record should include the ingredient type, optimal moisture content, and the basis for its calculation. The record format is: Ingredient: Chicken, Optimal Moisture Content: 65%. Based on the optimal requirements and moisture content of each ingredient, a multi-objective collaborative cooking parameter planning model is constructed. The model should comprehensively consider parameters such as cooking time, temperature, and heat for each ingredient. Using linear programming or an optimization algorithm, set the objective function to minimize cooking time or maximize the taste of the ingredients while simultaneously meeting the optimal state requirements for each ingredient. Use the optimization algorithm to generate collaborative cooking parameters for each ingredient. When cooking chicken and carrots simultaneously, set the temperature and time to ensure that both reach their optimal moisture content. Suppose the planned parameters are: chicken at 75°C for 30 minutes; carrots at 85°C for 20 minutes. Ensure that these parameters work together during the cooking process. Record the resulting multi-objective collaborative cooking parameters in the system to ensure that they are adjusted accordingly for each cooking session. The record should include information such as ingredient type, cooking temperature, time, and heat.The record format is: Ingredient: Chicken, Temperature: 75°C, Time: 30 minutes; Ingredient: Carrot, Temperature: 85°C, Time: 20 minutes. A high-precision weight sensor installed in the cooking appliance monitors the weight changes of the ingredients in real time to extract the latest moisture content. The sensor must have high resolution and be able to accurately capture weight changes within a short period of time. The initial weight of the chicken is monitored to be 1000 grams, and during the cooking process, the weight decreases to 800 grams. This change is recorded to calculate the moisture content. The current moisture content of the ingredient is calculated based on the change between the latest and initial weights. The current weight can be compared with the initial weight to determine moisture loss. If the chicken loses 200 grams of moisture during cooking, the current moisture content can be calculated as: Current weight = Initial weight - Moisture loss. The latest moisture content of the ingredient is recorded in a database for subsequent analysis and dynamic parameter adjustment. The record should include the ingredient type, current weight, and moisture content status. A dynamic adjustment model is constructed based on multi-objective collaborative cooking parameters and the latest moisture content of the ingredient. The model analyzes the difference between the moisture decay prediction and the current state in real time to adjust parameters. A feedback control system is used to compare real-time monitoring data with preset targets and dynamically adjust cooking parameters. Parameters such as cooking temperature, time, and heat are adjusted in real time based on the current moisture content of the ingredients and the moisture decay prediction. If the current moisture content is lower than the optimal value, increase the cooking time or lower the temperature to prevent over-drying. Assuming the current moisture content of the chicken is 60% and the optimal value is 65%, the cooking temperature can be adjusted to 70°C and the cooking time extended to 35 minutes to ensure the desired taste. The dynamically adjusted cooking parameters are recorded in a database to ensure that changes during each cooking process can be tracked. The records should include information such as the adjusted temperature, time, and heat.

[0036] In this embodiment, step S6 includes the following steps: Instant cooking control based on dynamic adjustment of cooking parameters and moisture content parameters of ingredients after cooking; Performing a final cooking deviation calculation on the moisture content parameters of each ingredient based on the optimal moisture content state of the ingredient, thereby obtaining cooking deviation feedback information; Iterative parameter regulation and optimization are performed on the cooking deviation feedback information to build an intelligent cooking parameter control model.

[0037] In this embodiment, during the cooking process, cooking parameters (such as temperature, time, and heat) are dynamically adjusted based on the real-time monitoring of the ingredient moisture content and the predicted moisture decay value. This process relies on real-time data feedback to ensure that the ingredients are cooked in optimal conditions. If the moisture content of the chicken is detected to be decreasing too rapidly during cooking, the system will automatically lower the cooking temperature or shorten the cooking time to prevent over-drying. For example, if the initial setting is 75°C, if the moisture content is detected to be decreasing too rapidly, the temperature is adjusted to 70°C and the cooking time is set to 35 minutes. Sensors within the cooking appliance continuously monitor the state of the ingredients. At regular intervals (e.g., every minute), the system obtains the latest weight and moisture content data and compares them with target values. If, at the 15th minute, the system records the current weight of the chicken as 750 grams, compared to the initial weight of 1000 grams, the current moisture content is calculated to be 60%. If the target moisture content is 65% at this time, the system will issue an adjustment instruction. At the end of cooking, the final moisture content parameters of the ingredients are recorded. This data will be used for subsequent deviation calculations and feedback analysis. After cooking, the final weight of the chicken is measured at 700 grams, and the calculated final moisture content is 55%. The record should include information such as the timestamp, ingredient type, and final moisture content. The final moisture content parameters for each ingredient are compared with their optimal moisture content to calculate the cooking deviation. The optimal moisture content is typically determined based on user preferences and experimental data. Assuming the optimal moisture content for chicken is 65%, and the actual moisture content after cooking is 55%, the deviation is calculated as: Deviation = Optimal Moisture Content - Actual Moisture Content = 65% - 55% = 10%. Based on the calculated deviation, cooking deviation feedback is generated. This feedback is used for subsequent parameter adjustment and optimization. The record format may include: ingredient type, optimal moisture content, actual moisture content, deviation value, and corresponding feedback suggestions. A record such as: Ingredient: Chicken, Optimal Moisture Content: 65%, Actual Moisture Content: 55%, Deviation: 10% is recorded in the system database for subsequent analysis and adjustment. This record will provide the data foundation for building an intelligent cooking parameter control model. Deviation information for all ingredients, including chicken and carrots, is recorded to form a data set for analysis. Based on the collected cooking deviation feedback, the causes of the deviation are analyzed. These may include factors such as insufficient cooking time, excessive heat, and insufficient initial moisture content of the ingredients. Through this analysis, a cooking parameter control model is constructed. Statistical analysis methods (such as regression analysis) are used to determine the extent to which each parameter affects the final moisture content, providing a basis for the model. Based on the model's output, new control strategies are designed to optimize cooking parameters. The goal is to reduce deviations and improve the final taste and customer satisfaction of the ingredients. If the deviation analysis shows that high temperature causes rapid moisture evaporation, the strategy can be adjusted to reduce the heat and extend the cooking time. At the same time, new parameter ranges are set to ensure closer to optimal results with each cooking.After applying the new parameter control strategy, evaluate and optimize it. Collect new data through actual cooking tests to evaluate the moisture content and taste of the adjusted ingredients. Compare these results with previous results to assess the effectiveness of the model. Record the cooking results after each adjustment to determine the success rate of the new strategy. If the final moisture content of the chicken after adjustment reaches 60%, closer to the optimal target, the model can be considered effectively optimized.

[0038] The present invention further provides a cooking appliance control device for executing the cooking appliance control method described above, comprising: The moisture content decay curve module is used to identify the weight monitoring parameters of the ingredients in the cooking utensils, perform multi-point moisture content decay analysis on the ingredients, and construct a real-time moisture content decay curve; The quantitative correlation module is used to identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and build a quantitative correlation model for moisture content changes; An image recognition module is used to obtain monitoring images from cooking utensils, perform deep target visual recognition, and mine personalized taste preferences to generate personalized taste characteristics for multiple ingredients. The moisture decay prediction module is used to identify the real-time cooking state based on the real-time moisture decay curve and the quantitative correlation model of moisture content change, and to perform rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; A cooking parameter adjustment module, configured to calculate the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjust the cooking parameters based on the moisture decay prediction value, thereby generating dynamically adjusted cooking parameters; The cooking control module is used to perform instant cooking control based on dynamic adjustment of cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

[0039] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned cooking appliance control methods when executing the computer program.

[0040] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any one of the above-mentioned methods for controlling a cooking appliance.

[0041] Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0042] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution is embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0043] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0044] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling a cooking appliance, characterized in that: The following steps are involved: Step S1: Identify the weight monitoring parameters of the food in the cooking appliance, perform a multi-point moisture content decay analysis on the food, and construct a real-time moisture content decay curve; Step S2: Identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and construct a quantitative correlation model for moisture content changes; Step S3: Obtain monitoring images inside the cooking appliance, perform deep target visual recognition and personalized taste preference mining to generate personalized taste characteristics for multiple ingredients; Step S4: performing real-time cooking state identification based on the real-time moisture content decay curve and the moisture content change quantitative correlation model, and performing rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; Step S5: Calculating the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjusting cooking parameters based on the moisture attenuation prediction value, thereby generating dynamically adjusted cooking parameters; Step S6: Perform instant cooking control based on dynamically adjusted cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

2. The cooking appliance control method according to claim 1, wherein: The specific steps of step S1 are: Identify food weight monitoring parameters in cooking utensils based on high-precision weight sensors; Calculating the initial food weight based on the food weight monitoring parameters; Calculate the weight change of the food weight monitoring parameters during the cooking process based on the initial food weight, and mark it as the food moisture content change value; Performing a multi-time-point moisture content decay analysis on the moisture content change value of the food to generate a moisture content decay rate and decay amplitude; The water content decay rate and decay amplitude are subjected to multi-time point decay fitting to construct a real-time water content decay curve.

3. The cooking appliance control method according to claim 1, wherein: The specific steps of step S2 are: Identify the real-time cooking environment status parameters of the current cooking appliance; Calculating the cooking temperature, humidity, and heat of the real-time cooking environment state parameters to generate a multi-dimensional cooking environment state feature; Perform time-series cooking state analysis on the multi-dimensional cooking environment state characteristics to generate cooking environment state characteristics at different time points; Based on the real-time moisture content decay curve and the cooking environment status characteristics at different time points, timestamp matching is performed, and quantitative correlation analysis is performed to generate a quantitative change relationship between cooking status and moisture content decay; The quantitative change relationship is dynamically modeled to construct a quantitative correlation model for water content change.

4. The cooking appliance control method according to claim 1, wherein: The specific steps of step S3 are: Obtain monitoring images inside cooking appliances and user-preset cooking instructions; Perform multi-image region segmentation on the monitoring image inside the cooking appliance and calculate the local contrast; performing global adaptive differential sharpening according to the local contrast to construct an adaptively sharpened monitoring image; Perform deep target visual recognition on adaptively sharpened monitoring images and mark ingredients in multiple utensils; Perform visual type analysis on ingredients in multiple utensils and generate the type of each ingredient; Based on the user's preset cooking instructions, personalized taste preferences are mined for each type of ingredient to generate personalized taste characteristics for multiple ingredients.

5. The cooking appliance control method according to claim 1, wherein: The specific steps of step S4 are: Recognize the real-time cooking status of ingredients on the adaptively sharpened monitoring image to obtain the real-time cooking status of ingredients; Analyze the current cooking stage of the food in real time based on the real-time moisture content decay curve to generate the current cooking stage of the food; performing moisture decay trend evolution on the current cooking stage according to a moisture content change quantitative correlation model to obtain a moisture decay trend of the current stage; A rolling time window moisture attenuation prediction is performed according to the moisture attenuation trend to obtain moisture attenuation prediction values for multiple time windows.

6. The cooking appliance control method according to claim 1, wherein: The specific steps of step S5 are: Analyze the optimal demand state based on the personalized taste characteristics of multiple ingredients to obtain the optimal demand state for each ingredient; Calculating the optimal moisture content according to the optimal demand state to generate the optimal moisture content state of each food ingredient; Performing multi-objective collaborative cooking parameter planning according to the optimal moisture content state to obtain multi-objective collaborative cooking parameters; Extracting the latest water content of the food based on the food weight monitoring parameters; The cooking parameters are dynamically adjusted for the latest moisture content of the food according to the multi-objective collaborative cooking parameters and the moisture attenuation prediction value, thereby generating dynamically adjusted cooking parameters.

7. The cooking appliance control method according to claim 1, wherein: The specific steps of step S6 are: Instant cooking control based on dynamically adjusted cooking parameters and the moisture content parameters of ingredients after cooking; Performing a final cooking deviation calculation on the moisture content parameters of each ingredient based on the optimal moisture content state of the ingredient, thereby obtaining cooking deviation feedback information; Iterative parameter regulation and optimization are performed on the cooking deviation feedback information to build an intelligent cooking parameter control model.

8. A control device for a cooking appliance, characterized in that: The method for controlling the cooking appliance according to claim 1 comprises: The moisture content decay curve module is used to identify the weight monitoring parameters of the ingredients in the cooking utensils, perform multi-point moisture content decay analysis on the ingredients, and construct a real-time moisture content decay curve; The quantitative correlation module is used to identify the real-time cooking environment state parameters, perform quantitative correlation analysis based on the real-time moisture content decay curve, and build a quantitative correlation model for moisture content changes; An image recognition module is used to obtain monitoring images from cooking utensils, perform deep target visual recognition, and mine personalized taste preferences to generate personalized taste characteristics for multiple ingredients. The moisture decay prediction module is used to identify the real-time cooking state based on the real-time moisture decay curve and the quantitative correlation model of moisture content change, and to perform rolling time window moisture decay prediction to obtain moisture decay prediction values for multiple time windows; A cooking parameter adjustment module, configured to calculate the optimal moisture content based on the user's personalized taste characteristics of multiple ingredients, and dynamically adjust the cooking parameters based on the moisture decay prediction value, thereby generating dynamically adjusted cooking parameters; The cooking control module is used to perform instant cooking control based on dynamic adjustment of cooking parameters, iteratively optimize parameter regulation, and build an intelligent cooking parameter control model.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the cooking appliance control method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cooking appliance control method according to any one of claims 1 to 7 are implemented.

Images