A method and control system for classifying rice grain water absorption performance and optimizing moisture distribution

Abstract

The application discloses a kind of rice water absorption performance classification and moisture distribution optimization method and control system, by analyzing the water absorption performance of rice, moisture diffusion distribution, realizes the accurate control of rice cooking process.First, the water absorption performance of rice is classified using support vector machine algorithm, and the moisture distribution equation is constructed to calculate the internal moisture distribution of rice.Then, according to the moisture distribution, the heat transfer process is optimized.Finally, the heating power is dynamically adjusted, the internal heat distribution of rice is optimized, and the uniformity of moisture absorption is realized.The application accurately controls the whole process of rice cooking by multidimensional data analysis and intelligent algorithm application, effectively improves the consistency and taste of rice quality, and provides a new technical scheme for intelligent cooking equipment.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method and control system for classifying the water absorption performance of rice grains and optimizing moisture distribution. Background Technology

[0002] Rice cooking, as a crucial step in food processing, directly impacts the taste and quality of food, playing a key role in enhancing the consumer experience. The uniformity of rice texture is not only a core aspect of cooking technique but also an important guarantee for meeting diverse dietary needs. However, current cooking methods commonly suffer from uneven rice texture when dealing with different types of rice, particularly exhibiting a soft, sticky outer layer and a firmer interior. This is primarily due to the difficulty of effectively controlling the diffusion of moisture within the rice grain using existing methods, leading to uneven moisture distribution and affecting the final taste.

[0003] Traditional cooking methods typically rely on experience to adjust water volume and heating time, lacking precise adaptation to the characteristics of rice grains. Different rice varieties exhibit significant differences in water absorption and diffusion rates due to variations in internal structure and composition, such as the proportion of different starch types and protein content. The waxy layer and endosperm layer on the surface of the rice grain further hinder water transfer to the interior, resulting in rapid water absorption on the surface while insufficient absorption inside, leading to uneven water distribution. This uneven water diffusion causes a temperature gradient to form between the inside and outside of the rice grain during heating, creating a vapor pressure difference that affects the direction and rate of water migration within the grain. This uneven diffusion process makes it difficult for the core of the rice grain to fully absorb water and be heated, resulting in the technical challenge of differences in taste.

[0004] Therefore, the key issue in improving the taste and quality of rice is to establish a precise moisture transfer model based on the microstructure and composition characteristics of different rice varieties, and to optimize the diffusion process of moisture inside and between rice grains by dynamically controlling the temperature and pressure fields, so as to achieve uniform water absorption and heating from the surface to the core. Summary of the Invention

[0005] This invention provides a method for classifying the water absorption performance of rice grains and optimizing moisture distribution, mainly including:

[0006] Relevant characteristic parameters were obtained from rice variety characteristic data to classify the water absorption performance of rice varieties and obtain the water absorption performance classification results; a moisture diffusion model was constructed to determine the moisture content difference distribution; heating parameters and / or water addition parameters were adjusted to optimize heat transfer within rice grains;

[0007] By combining the distribution of moisture content differences, a transfer model is constructed to determine the final standard deviation of moisture distribution.

[0008] Furthermore, the method includes: obtaining relevant feature parameters from rice variety characteristic data to construct a feature dataset; classifying the water absorption performance of rice varieties using a classification algorithm based on the feature dataset to obtain water absorption performance classification results; constructing a moisture diffusion model based on the water absorption performance classification results and specific feature parameters to calculate the internal moisture diffusion rate of rice grains and determine the moisture content difference distribution; if the moisture content difference distribution exceeds a preset threshold, optimizing the internal heat transfer of rice grains by adjusting heating and water addition parameters to obtain a uniform heat gradient distribution; constructing a transfer model based on the uniform heat gradient distribution and the moisture content difference distribution to optimize the uniformity of rice grain water absorption, dynamically adjusting control parameters, and determining the final moisture distribution standard deviation. Furthermore, the method includes: extracting starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm porosity as initial features from a rice variety characteristic database; normalizing the values ​​of each initial feature to obtain a standardized feature vector; constructing a feature dataset for classification and analyzing the completeness of the feature dataset; if there are missing or abnormal data, marking and supplementing the corresponding data to obtain the final feature dataset.

[0009] Furthermore, the method includes: extracting starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm pore permeability as initial feature data from a rice variety characteristic database; preprocessing the initial feature data by normalizing the value of each feature using a standardization method to obtain a standardized feature vector; constructing a feature dataset for classification based on the standardized feature vector; analyzing the correlation between the features in the feature dataset to determine the completeness of the feature dataset; and marking and supplementing the corresponding data if there is missing or abnormal data in the feature dataset to obtain the final feature dataset.

[0010] Furthermore, the method includes: constructing a water absorption performance classification model; predicting the water absorption performance category of rice varieties using the classification model; marking data with confidence levels below a preset threshold; re-analyzing the marked data using specific feature parameters to adjust the water absorption performance classification results; and outputting a final list of rice varieties with water absorption performance classifications based on the adjusted classification results.

[0011] Furthermore, the method includes: using a support vector machine algorithm to perform classification training based on the feature dataset to construct a water absorption performance classification model; using the classification model to predict the feature dataset and determine the water absorption performance category of the rice variety; if the confidence level of the prediction result is lower than a preset threshold, then labeling the corresponding data; performing secondary analysis on the labeled data in conjunction with specific feature parameters to adjust the classification result; and outputting a list of rice varieties in different categories based on the adjusted classification result to determine the final water absorption performance classification output.

[0012] Furthermore, the method includes: constructing an initial analysis dataset based on the water absorption performance classification results, combined with the waxy layer barrier coefficient and endosperm pore permeability; extracting features of surface water penetration depth and core water diffusion delay to obtain a preliminary feature set; constructing a diffusion equation based on the water diffusion coefficient based on the preliminary feature set; calculating the rice grain water diffusion rate through the diffusion equation to obtain rate distribution data; and analyzing the matching relationship between surface saturation and core hysteresis based on the rate distribution data to determine the moisture content difference distribution.

[0013] Furthermore, the method includes: constructing an initial analysis dataset based on the water absorption performance classification results, combined with the waxy layer barrier coefficient and endosperm pore permeability; extracting features of surface water penetration depth and core water diffusion delay from the initial analysis dataset to obtain a preliminary feature set; constructing a diffusion equation based on the water diffusion coefficient based on the preliminary feature set; calculating the internal water diffusion rate of rice grains using the diffusion equation to obtain rate distribution data; and analyzing the matching relationship between surface saturation and core hysteresis based on the rate distribution data to determine the moisture content difference distribution. Further, the method includes: obtaining initial moisture distribution data within rice grains; triggering an adjustment mechanism if the difference in initial moisture distribution data exceeds a preset threshold to determine the adjustment range of heating time; calculating the heat distribution change based on the adjustment range to obtain dynamic adjustment parameters; adjusting the internal moisture distribution of rice grains to obtain real-time deviation data; and analyzing the heat distribution uniformity to determine the final adjustment scheme.

[0014] Furthermore, the method includes: obtaining initial moisture distribution data inside the rice grain based on the moisture content difference distribution; if the difference in the initial moisture distribution data exceeds a preset threshold, triggering an adjustment mechanism to determine the adjustment range of heating time and water volume; calculating the heat distribution change using a heat transfer model based on the adjustment range to obtain dynamic adjustment parameters; adjusting the moisture distribution inside the rice grain using the dynamic adjustment parameters to obtain real-time deviation data; analyzing the heat distribution uniformity based on the real-time deviation data to determine the final adjustment scheme. Furthermore, the method includes: applying a support vector machine algorithm to classify and train the water absorption performance of rice varieties to construct a classification model; predicting the feature dataset using the classification model to determine the water absorption performance category of the rice varieties; marking data with confidence levels below a preset threshold; re-analyzing the marked data using specific feature parameters to adjust the water absorption performance classification results; and outputting the final list of rice varieties classified by water absorption performance based on the adjusted classification results.

[0015] Furthermore, the method includes: combining monitoring data of surface infiltration and core diffusion to analyze differences in moisture distribution and determine a preliminary distribution state; constructing a transfer equation based on the diffusion coefficient for the preliminary distribution state to obtain the moisture transfer law; analyzing the relationship between water absorption uniformity and distribution based on the moisture transfer law; and predicting the trend of moisture distribution change through the updated distribution data to determine the final moisture distribution standard deviation.

[0016] Furthermore, the method includes: constructing an initial analysis dataset based on the water absorption performance classification results, combined with the waxy layer barrier coefficient and endosperm pore permeability; extracting features of surface water penetration depth and core water diffusion delay to obtain a preliminary feature set; constructing a diffusion equation based on the water diffusion coefficient based on the preliminary feature set; calculating the rice grain water diffusion rate through the diffusion equation to obtain rate distribution data; and using a support vector machine algorithm to perform a correlation analysis on surface water saturation and core water lag based on the preliminary feature set to determine the distribution pattern of the water content difference between surface saturation and core lag.

[0017] Furthermore, the method includes: extracting starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm porosity as initial features from a rice variety characteristic database; performing numerical normalization on the starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm porosity using a standardization method to obtain standardized feature vectors; constructing a feature dataset for classification and analyzing the completeness of the feature dataset; if there are missing or abnormal data, marking and supplementing the corresponding data to obtain the final feature dataset. This invention also provides a rice cooker control system utilizing rice grain water absorption performance classification and moisture distribution optimization methods, including a unit for executing the methods described above.

[0018] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0019] This invention discloses a method for classifying rice grain water absorption performance and optimizing moisture distribution, as well as a rice cooker control system. By analyzing the water absorption performance and moisture diffusion distribution of different rice varieties, precise control of the rice cooking process is achieved. First, the water absorption performance of different rice varieties is classified using a support vector machine algorithm, and a moisture diffusion equation is constructed to calculate the internal moisture distribution of the rice grains. Then, the heat transfer process is optimized based on the moisture distribution. Finally, the heating power is dynamically adjusted to optimize the internal heat distribution of the rice grains, achieving uniform moisture absorption. This invention, through multi-dimensional data analysis and the application of intelligent algorithms, precisely controls the entire rice cooking process, effectively improving the consistency and taste of rice quality, and providing a new technical solution for intelligent cooking equipment. Attached Figure Description

[0020] Figure 1This is a flowchart of a method for classifying the water absorption performance of rice grains and optimizing moisture distribution according to the present invention.

[0021] Figure 2 This is a flowchart of another method for classifying the water absorption performance of rice grains and optimizing moisture distribution according to the present invention.

[0022] Figure 3 This is a flowchart of another method for classifying the water absorption performance of rice grains and optimizing moisture distribution according to the present invention.

[0023] Figure 4 This is a flowchart of another method for classifying the water absorption performance of rice grains and optimizing moisture distribution according to the present invention.

[0024] Figure 5 This is a schematic diagram of the logic structure of a rice cooker control system for classifying rice grain water absorption performance and optimizing moisture distribution, provided in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0026] like Figure 1-2 This embodiment of a method for classifying the water absorption performance of rice grains and optimizing moisture distribution may specifically include:

[0027] Steps S101-S102: Obtain starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm pore permeability from the rice variety characteristic database. Use the support vector machine algorithm to classify the water absorption performance of rice varieties using the aforementioned characteristics as feature vectors. Determine the water absorption performance category as high, medium, or low to obtain the water absorption performance classification result.

[0028] The rice varietal characteristics database can be a self-built database. Legitimate data sources for the database include, but are not limited to, the following: the National Rice Data Center, covering rice variety resources, gene data, breeding materials, and cultivation techniques, supporting searches by variety name and characteristic; the International Rice Research Institute (IRRI), providing the Germplasm Resources Information Network (GRIN), containing gene and agronomic trait data for wild and cultivated rice; relevant journal articles, seed industry companies and industry reports, field surveys or collaborations with institutions, agricultural product trading platforms, and data obtained through experimental measurement and analysis of commercially available rice samples of different varieties. From the rice varietal characteristics database, starch water absorption and swelling rate, protein water-binding capacity, wax layer barrier coefficient, and endosperm porosity are obtained to construct an initial characteristic dataset, resulting in a set of rice varietal characteristics.

[0029] Data preprocessing was performed on a set of rice variety characteristics. Standardization methods were used to normalize the starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm porosity, obtaining standardized feature vectors. Based on these standardized feature vectors, a support vector machine (SVM) algorithm was applied to train a classification model for rice variety water absorption performance, determining the initial classification boundaries. The classification model was used to predict the standardized feature vectors, determining the rice variety's water absorption performance category. If the confidence level of the prediction result was lower than a preset threshold, the rice variety characteristic data was labeled, obtaining a preliminary classification result. For the labeled data, a secondary analysis was performed using endosperm porosity and wax layer barrier coefficient. If both indicators were higher than the preset threshold, the classification result was adjusted to a high category, obtaining a corrected classification result. Based on the corrected classification result, lists of rice varieties classified as high, medium, and low categories were output, determining the final water absorption performance classification output. The final water absorption performance classification output generated structured data of the classification results, which was stored in a database for persistent recording.

[0030] Specifically, data is obtained from a rice variety characteristics database. Assuming the database contains 100 rice varieties, each variety includes starch water absorption and swelling rate (range 0.5-2.0, unit g / g), protein water-binding capacity (range 0.1-0.5, unit g / g), waxy layer barrier coefficient (range 0.01-0.1, unit not specified), and endosperm porosity (range 0.001-0.01, unit cm² / s). Taking a certain rice variety as an example, the data are: starch water absorption and swelling rate 1.8, protein water-binding capacity 0.4, waxy layer barrier coefficient 0.03, and endosperm porosity 0.008.

[0031]

[0032] C(I) represents the classification result based on the comprehensive water absorption index I. When I > 0.8, it is classified as high water absorption performance; when 0.4 ≤ I ≤ 0.8, it is classified as medium water absorption performance; and when I < 0.4, it is classified as low water absorption performance. This classification standard is based on experimental data and is used to evaluate the water absorption performance of rice.

[0033]

[0034] I represents the comprehensive water absorption index, ω i Let x represent the weight of the i-th feature. i Let represent the i-th feature value. In this example, the weight vector is [0.4, 0.3, 0.2, 0.1]. The comprehensive water absorption index is calculated by weighted summation and used for the final classification judgment.

[0035]

[0036] CV represents the average accuracy of cross-validation, k represents the number of folds (k=5 in this example), and Acc... i Let represent the accuracy at the i-th fold. 5-fold cross-validation divides the dataset into 5 parts, using 4 parts for training and 1 part for validation each time, repeating 5 times, and taking the average accuracy as the model performance evaluation metric.

[0037]

[0038] X represents the eigenvector, and x1, x2, x3, and x4 represent the four standardized eigenvalues. In this example, the eigenvector is [1.833, 0.667, -0.5, 1.2], which contains various physicochemical properties of the meter.

[0039]

[0040] Z i Let x represent the standardized eigenvalues. i Let μ represent the original feature value, σ represent the feature mean, and σ represent the feature standard deviation. This is the Z-score standardization formula, used to convert features of different dimensions into a standard normal distribution with a mean of 0 and a standard deviation of 1, ensuring that each feature has the same weight in the model. During data preprocessing, Z-score standardization is used, subtracting the mean from each feature value and dividing by the standard deviation to ensure consistent feature dimensions. For example, the standardized value of starch water absorption swelling rate is (1.8-1.25) / 0.3=1.833. The feature vector is constructed as [1.833, 0.667, -0.5, 1.2] (assuming the standardized values ​​of other features). The Support Vector Machine (SVM) algorithm is used, with a radial basis function (RBF) kernel, parameters C=1.0, gamma=0.1, and the model is optimized through 5-fold cross-validation.

[0041] In the training data, 80% (80 types of rice) was used for training, and 20% (20 types of rice) was used for testing. The SVM model categorized water absorption performance into three classes: high, medium, and low. The classification criteria were based on experimental data: high (comprehensive water absorption index > 0.8), medium (0.4-0.8), and low (< 0.4). The comprehensive water absorption index was calculated by weighted summation of features, with weights of [0.4, 0.3, 0.2, 0.1]. The index for the example rice type was 1.8×0.4 + 0.4×0.3 + 0.03×0.2 + 0.008×0.1 = 0.846, classifying it as high. The model's prediction accuracy was 85%. The confusion matrix showed a prediction accuracy of 90% for the high class, 80% for the medium class, and 85% for the low class. To ensure logical rigor, the importance of features was analyzed, revealing that starch water absorption swelling rate contributed the most to the classification (weight 0.4), as it is directly related to the water diffusion rate. Ultimately, the example rice variety was classified as having high water absorption, validating the effectiveness of the model.

[0042] Step S103: Based on the water absorption performance classification results, combined with the wax layer barrier coefficient and endosperm pore permeability, a diffusion equation based on the water diffusion coefficient is constructed for the surface water penetration depth and core water diffusion delay of rice grains. The internal water diffusion rate of rice grains is calculated, and the distribution of the water content difference between surface water saturation and core water lag is determined.

[0043] Based on the water absorption performance classification results, and combined with data on the waxy layer barrier coefficient and endosperm pore permeability, an initial analysis dataset was constructed. Features were extracted for the surface water penetration depth and core water diffusion delay of rice grains, resulting in a preliminary feature set of surface and core water distribution. Based on this preliminary feature set, a support vector machine algorithm was used to analyze the correlation between surface water saturation and core water lag, determining the distribution pattern of the water content difference between surface saturation and core lag. For this distribution pattern, a diffusion equation based on the water diffusion coefficient (determined by both the waxy layer barrier and endosperm pore permeability) was constructed to calculate the internal water diffusion rate of rice grains, obtaining rate distribution data. The rate distribution data was used to analyze the matching relationship between water diffusion rate and surface saturation. If the rate distribution data was below a preset threshold, the corresponding rice grain data was marked, resulting in a marked rate anomaly set. Based on the marked rate anomaly set and combined with the endosperm pore permeability characteristics, a secondary calibration of core water lag was performed, obtaining calibrated lag distribution data. Based on the calibrated hysteresis distribution data, the dynamic balance between surface water penetration depth and core water hysteresis is analyzed to determine the final water content difference distribution result. Using the final water content difference distribution result, structured distribution data records are generated and stored in a pre-established database to obtain persistent distribution analysis records.

[0044] Specifically, based on the classification results of rice grain water absorption performance, and combined with the barrier coefficient of the waxy layer and the permeability of the endosperm pores, a water diffusion model is constructed to analyze the water penetration depth of the rice grain surface and the core water diffusion delay, and then calculate the internal water diffusion rate and the distribution of water content difference.

[0045]

[0046] This represents the maximum difference in water content between the surface and the core, where C surface C represents the surface moisture content. core This represents the core moisture content, and the calculated maximum difference is 45%.

[0047]

[0048] This represents the endosperm porosity of rice, with a value of 0.006 cm² / s, which is a parameter describing the ability of water to pass through the endosperm tissue of rice grains.

[0049]

[0050] This is the iterative formula for solving the diffusion equation using the finite difference method, used to numerically simulate the process of water infiltration from the surface to the core. Here, C(x,t) represents the water concentration at location x at time t, Δt represents the time step, and Δx represents the spatial step.

[0051]

[0052] This is the formula for calculating the resistance to water osmosis, where R... w The value represents the water penetration resistance, α represents the barrier coefficient of the wax layer (0.05), and δ represents the surface thickness (0.01cm). The calculated resistance value is 0.08.

[0053]

[0054] This is the formula for calculating the water diffusion rate, where v represents the diffusion rate, D represents the diffusion coefficient (0.004 cm² / s), ΔC represents the concentration difference, and Δx represents the grain radius (0.2 cm). The calculated average diffusion rate from the surface to the core is approximately 0.012 cm / s.

[0055]

[0056] This is a partial differential equation for moisture diffusion, where C represents moisture concentration, t represents time, x represents spatial coordinates, and D represents the diffusion coefficient (value 0.004 cm² / s). This equation describes the variation of moisture in rice grains over time and space. First, the waxy layer barrier coefficient of a certain rice variety was extracted from the database as 0.05, and the endosperm pore permeability was 0.006 cm² / s. It was assumed that this rice variety had been classified as having medium water absorption capacity. Using these data, a partial differential equation based on the moisture diffusion coefficient was constructed, setting the diffusion coefficient D to 0.004 cm² / s. The equation was solved using the finite difference method to simulate the process of moisture permeation from the surface to the core. The calculation showed that the surface moisture saturation reached 80% within the first 10 minutes, while the core moisture content was only 30%, forming a significant lag. Further analysis of the moisture diffusion rate, using the formula v = D × ΔC / Δx, where ΔC is the concentration difference and Δx is the rice grain radius of 0.2 cm, yielded an average diffusion rate of approximately 0.012 cm / s from the surface to the core. Subsequently, a two-dimensional grid model was constructed to analyze the difference in moisture content between the surface and the core of the rice grain. The grain was divided into 100 micro-units, and the change in moisture content over time in each unit was calculated. It was found that even after the surface layer was saturated, the core moisture still needed approximately 30 minutes to reach equilibrium, with a maximum content difference of 45%. To refine the analysis, the surface thickness of the rice grain was introduced as an auxiliary parameter, assumed to be 0.01 cm. Combined with the barrier coefficient of the waxy layer, the water penetration resistance was calculated, revealing that the surface moisture accumulation rate is significantly affected by the barrier coefficient, with a resistance value of 0.08. Through these calculations and analyses, a complete chain was formed, from data extraction to model construction and then to rate and distribution calculations, comprehensively revealing the internal moisture migration patterns of the rice grain.

[0057] Step S104: If the difference in moisture content is greater than a preset threshold of 5 mass percentages, then by adjusting the heating duration and the volume of water added, the heat transfer process inside the rice grain is optimized to address the surface moisture saturation and core moisture lag, thereby obtaining a uniform heat gradient distribution from the surface to the core.

[0058] By analyzing the difference between moisture content and distribution, initial moisture distribution data within the rice grain is obtained, determining the initial deviation between surface saturation and core lag. Based on the initial moisture distribution data, a preset threshold is compared. If the distribution difference exceeds the preset threshold, an adjustment mechanism is triggered, obtaining a reference range for the heating time and water volume to be adjusted. Within this reference range, a pre-established heat transfer model is used to calculate the heat distribution change from the surface to the core within the rice grain, obtaining dynamic adjustment parameters for heat transfer. These dynamic adjustment parameters, combined with heating time and water volume, are used to adjust the moisture content within the rice grain, obtaining real-time deviation data between surface saturation and core lag. Based on the real-time deviation data, the degree of matching between the heat distribution and the uniform gradient is analyzed. If the matching degree does not meet the preset standard, the heat transfer parameters are iteratively adjusted to determine the final adjustment scheme. Using the final adjustment scheme, the moisture adjustment process within the rice grain is continuously monitored, obtaining uniform gradient distribution data from the surface to the core to determine if the preset equilibrium conditions are met. The final state of moisture content and heat distribution is recorded using the uniform gradient distribution data, obtaining complete adjustment process data.

[0059] Specifically, for cases where the difference in internal moisture content of a rice grain exceeds a preset threshold of 5% by mass, the system first collects real-time moisture data from the surface and core using sensors. Assuming the surface moisture content is 75% and the core moisture content is 65%, the calculated difference is 10%, exceeding the threshold and triggering an optimization process. Next, the system invokes a heat transfer simulation algorithm based on the heat conduction equation... T / t=k ²T, where k is the thermal conductivity set to 0.002 cal / cm·s·℃, is used to calculate the current heat gradient. The surface temperature is found to be 80℃, and the core temperature is 60℃, with a gradient difference of 20℃, indicating uneven heat distribution. Subsequently, the system automatically calculates the heating duration using a preset heating adjustment model. The initial duration is 20 minutes, and based on the difference ratio of 10% / 5%=2, the heating duration is extended to 20×1.2=24 minutes to enhance heat transfer to the core. Simultaneously, based on the relationship model between moisture difference and added water volume, the system sets the initial added water volume to 100 ml, and adjusts it to 100×1.1=110 ml based on the difference ratio, promoting heat penetration by increasing humidity. Furthermore, the system uses a three-dimensional heat distribution grid algorithm to divide the rice grains into 50 units, iteratively calculating the temperature change of each unit. The predicted adjustment results in a surface temperature drop to 78℃, a core temperature increase to 70℃, and a gradient difference reduction to 8℃. Finally, the system analyzes the adjusted moisture distribution. The surface moisture content is expected to be 77%, and the core moisture content is 70%. The difference is reduced to 7%, which is still higher than the threshold but tends to be optimized. The system records this data and links it to the rice heat treatment database to provide a reference for subsequent batches, forming a closed-loop logic from data acquisition, heat simulation to parameter adjustment, to ensure that the heat transfer process is gradually uniform.

[0060] Step S105: Based on the internal structural characteristics of the rice grains, calculate the heat transfer rate and moisture diffusion rate to obtain dynamic distribution data of surface moisture saturation and core moisture lag. If the deviation between surface moisture saturation and core moisture lag exceeds a preset threshold, adjust the heat input rate using a pre-established heat transfer model to determine the optimized heat distribution uniformity parameters.

[0061] For heat distribution and uniformity parameters, relevant steam distribution data is acquired through a real-time monitoring system. A support vector machine algorithm is used to predict the uniformity trend, yielding preliminary steam distribution uniformity trend data. Based on this trend data, combined with core accumulation and steam adjustment attributes, the real-time range of core steam accumulation is analyzed to determine if the current distribution deviates from a preset standard. If the real-time range of core steam accumulation exceeds the preset standard, the steam adjustment parameters are optimized using a pre-established heat distribution adjustment model, resulting in an adjusted steam distribution parameter set. Based on this adjusted parameter set, combined with real-time monitoring data, the dynamic changes in heat distribution and uniformity parameters are analyzed to determine if the steam distribution is approaching a uniform state. If the steam distribution has not yet reached a uniform state, iterative calculation tools are used to fine-tune the steam adjustment parameters based on core accumulation and real-time range data, resulting in a further optimized distribution scheme. Using the optimized distribution scheme, combined with trend and predictive analysis data, the steam distribution uniformity trend is continuously monitored to determine if it meets the preset standard. Based on the final steam distribution uniformity trend data, the dynamic adjustment process of heat distribution and core accumulation is recorded, obtaining complete steam distribution uniformity information.

[0062] Specifically, the system first collects real-time heat distribution data from different regions of the rice grain using a built-in heat distribution sensor matrix. Assuming a surface heat density of 5.2 J / cm³ and a core heat density of 3.8 J / cm³, the calculated heat distribution uniformity parameter is 0.73, indicating a certain deviation in heat distribution. Next, the system combines this parameter with real-time monitored steam distribution data. Assuming a current steam distribution uniformity of 0.65, a support vector machine algorithm is used for analysis. The model, trained on 5000 historical heat and steam distribution samples, predicts that the steam distribution uniformity will rise to 0.68 within the next 5 minutes. Based on the prediction, the system further calculates the real-time adjustment range for core steam accumulation. Assuming a current accumulation of 0.55, the algorithm output determines the adjustment range to be between 0.52 and 0.58. Subsequently, the system calls the uniformity evaluation module to compare the predicted uniformity of 0.68 with the preset standard value of 0.70. Analysis shows that the current value does not meet the standard, automatically triggering the adjustment mechanism. The system, by referencing a rice grain moisture distribution database, obtains the current rice variety's moisture parameter as 0.25. Combined with the heat distribution uniformity parameter of 0.73, it generates an adjustment strategy to fine-tune the core steam accumulation degree to 0.57, thereby improving overall uniformity. The entire process forms a closed-loop logic from data acquisition and algorithm prediction to automatic adjustment, ensuring that the steam distribution gradually approaches the preset standard.

[0063] Based on the steam pressure difference, the response frequency of the pressure valve is dynamically adjusted to control the heating power output and the opening range of the pressure valve, thereby optimizing the internal steam distribution.

[0064] Step S106: Based on the moisture content difference distribution, construct a transfer equation based on the moisture diffusion coefficient for the surface moisture penetration depth and core moisture diffusion delay, optimize the water absorption uniformity of rice grains from the surface to the core, and determine the final standard deviation of the rice moisture distribution.

[0065] Real-time monitoring values ​​of surface infiltration and core diffusion in rice grains are acquired to analyze the differences in moisture content distribution and determine the preliminary moisture distribution status. Based on the preliminary moisture distribution status, a transfer equation based on the diffusion coefficient is constructed to address the delay factor between surface infiltration and core diffusion, thus obtaining the transfer law of moisture in the rice grain structure. Through the transfer law, the relationship between water absorption uniformity and moisture distribution is analyzed. If the difference between the surface infiltration rate and the core diffusion rate exceeds a preset threshold, the steam balance parameters are adjusted to determine the updated distribution difference data. Based on the updated distribution difference data, a pre-established regression model is used to predict the changing trend of moisture distribution in the rice grain structure, obtaining a dynamic water absorption uniformity index. The correlation between moisture distribution and standard deviation is monitored through the dynamic water absorption uniformity index. If the standard deviation does not reach a preset range, the diffusion coefficient-related parameters are adjusted to determine the optimized transfer equation result. Based on the optimized transfer equation result, moisture distribution data inside the rice grain is continuously collected to analyze the coordination between surface infiltration and core diffusion, obtaining the final standard deviation of cooked rice moisture distribution. By using the final standard deviation of the rice moisture distribution, combined with steam balance parameters, the stability of water absorption uniformity in the rice grain structure is verified, and the complete business processing results are determined.

[0066] Specifically, the system uses a high-precision moisture sensor array to monitor the moisture content of the rice grain's surface and core in real time. The surface moisture content was found to be 35.0%, while the core moisture content was 28.0%, resulting in a moisture content difference of 7.0%, indicating uneven water absorption distribution. The system then invoked a moisture penetration depth detection module, measuring the surface moisture penetration depth to be 0.12 cm. Combining this with the rice grain porosity database (the current rice variety has a porosity of 0.40), the moisture diffusion equation was used to... C / t=D ²C (where D is the moisture diffusion coefficient, set to 0.0015 cm² / s) simulates the transfer of moisture from the surface to the core. The system inputs moisture content difference and penetration depth data into a random forest algorithm. The model is trained based on 3000 historical samples, including moisture content difference, penetration depth, and rice variety characteristics. It predicts a core moisture diffusion delay of 0.85 s, and the correlation coefficient between diffusion delay and moisture content difference is calculated to be 0.82, showing a high correlation between the two. Based on the prediction results, the system dynamically adjusts the heating time. The initial heating time is set to 300 s. When the moisture content difference exceeds 5.0%, the algorithm automatically extends the heating time to 360 s to promote moisture diffusion. The system further combines the rice variety database, adjusts the moisture diffusion coefficient to 0.0017 cm² / s, and recalculates that after 5 minutes, the core moisture content rises to 30.5%, the surface moisture content drops to 33.8%, and the moisture content difference is optimized to 3.3%. To ensure uniform water absorption, the system analyzes the standard deviation of moisture distribution. The initial standard deviation is 2.5%. Through the optimized diffusion coefficient and heating time, the predicted standard deviation is reduced to 1.8%, forming a closed-loop logic from data acquisition and algorithm prediction to parameter optimization.

[0067] Based on steam balance data and real-time monitoring values ​​of surface infiltration and core diffusion, a transfer equation is constructed using the diffusion coefficient. By analyzing the differences in moisture content distribution, the transfer law of moisture in the rice grain structure is obtained, the relationship between water absorption uniformity and moisture distribution is determined, and moisture distribution data inside the rice grain is continuously collected to judge the coordination between surface infiltration and core diffusion, thus obtaining the final standard deviation of rice moisture distribution.

[0068] The process involves acquiring steam balance data and real-time monitoring values ​​of surface infiltration and core diffusion. A preset threshold is used to determine distribution differences. If the difference exceeds the threshold, the steam balance parameters are adjusted to obtain updated moisture distribution data. Based on the updated moisture distribution data, a transfer equation based on the diffusion coefficient is constructed and solved using the finite difference method to determine the moisture transfer pattern within the rice grain structure. The rate difference between surface infiltration and core diffusion is analyzed based on this transfer pattern. If the rate difference exceeds a preset threshold, the diffusion coefficient is adjusted to obtain optimized transfer equation parameters. Based on the optimized transfer equation parameters, moisture distribution data within the rice grain is continuously collected. A K-means clustering algorithm is used to determine the coordination between surface infiltration and core diffusion, obtaining a water absorption uniformity index. The standard deviation of the moisture distribution is calculated using this water absorption uniformity index. If the standard deviation does not reach a preset range, the steam balance parameters are iteratively adjusted to determine updated moisture distribution data. Based on the updated moisture distribution data, a pre-established linear regression model is used to predict the trend of moisture distribution changes, yielding the final standard deviation of the cooked rice moisture distribution. The coordination between surface infiltration and core diffusion was verified by the final standard deviation of rice moisture distribution, and the stability of the water absorption uniformity index was determined.

[0069] Specifically, the system first collects internal steam pressure data through its built-in steam balance analysis module. Combined with multi-point moisture detectors, it obtains a surface moisture content of 38.2% and a core moisture content of 25.6%, calculating a distribution difference of 12.6% and recording a steam balance index of 0.75. Next, the system uses a moisture transfer monitoring algorithm to analyze the surface infiltration rate in real time, measuring it to be 0.08 cm / min. Simultaneously, it calculates the diffusion rate using a core diffusion model, obtaining a core diffusion rate of 0.03 cm / min. The compatibility index between the two is analyzed to be 0.62, indicating a significant inconsistency between surface infiltration and core diffusion. The system then constructs a transfer equation based on the diffusion coefficient, setting an initial diffusion coefficient of 0.002 cm² / s, and uses the equation... M / t=D The system simulates the water transfer pattern and, based on the density of the current rice variety (1.2 g / cm³) in the rice grain structure parameter database, derives the water distribution curve within the rice grain, calculating the transfer time difference from the surface to the core as 1.2 seconds. To further analyze water absorption uniformity, the system calls the distribution difference analysis module, comparing the water distribution data with 5000 sets of data in the historical sample database. Using a support vector machine algorithm, the predicted water absorption uniformity index is 0.68, and the correlation coefficient between water distribution and the uniformity index is calculated to be 0.78, indicating a strong correlation between the two. The system continuously collects water distribution data every 30 seconds using a multi-point detector, updating the surface moisture content to 36.5%, the core moisture content to 27.3%, and the distribution difference to 9.2%. The system also reassesses the coordination index between surface infiltration and core diffusion as 0.71. Finally, the system uses the statistical analysis module to calculate the standard deviation of moisture distribution. The initial value is 3.1%. Iterative optimization is performed by combining the transfer equation and the coordination index to predict the final standard deviation as 2.2%. The results are then stored in the database, forming a complete logical chain from data collection to pattern analysis and distribution optimization.

[0070] This application also provides a rice cooker control system for classifying rice grain water absorption performance and optimizing moisture distribution, including a unit for performing the method described in any of the preceding claims.

[0071] Figure 5 This is a schematic diagram of the logic structure of a rice cooker control system for classifying rice grain water absorption performance and optimizing moisture distribution, provided as an embodiment of this application. Figure 5As shown, the system 1 of this embodiment includes a processing unit 11, a storage unit 12, and a transceiver unit 13. The processing unit 11 is used to process data, the storage unit 12 is used to store data, and the transceiver unit 13 is used to send and receive data. The processing unit 11, the storage unit 12, and the transceiver unit 13 cooperate with each other to implement the above-described method. The beneficial effects of the embodiments of this application have been described in the above-described method and will not be repeated here.

[0072] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0073] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0074] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0075] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0076] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0077] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0078] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0079] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for classifying the water absorption performance of rice grains and optimizing moisture distribution, characterized in that, include: Relevant characteristic parameters are obtained from rice variety characteristic data, and the water absorption performance of rice varieties is classified to obtain the water absorption performance classification results; A moisture diffusion model is constructed to determine the moisture content difference distribution; heating parameters and / or water addition parameters are adjusted to optimize heat transfer within the rice grain; based on the internal structural characteristics of the rice grain, the heat transfer rate and moisture diffusion rate are calculated to obtain dynamic distribution data of surface moisture saturation and core moisture lag; if the deviation between surface moisture saturation and core moisture lag exceeds a preset threshold, the heat input rate is adjusted using the pre-established heat transfer model to determine the optimized heat distribution uniformity parameters; the steam pressure gradient change is predicted to determine the steam pressure difference; combining the moisture content difference distribution and the steam pressure difference, a transfer model is constructed to determine the final moisture distribution standard deviation.

2. The method as described in claim 1, characterized in that, Also includes: Starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm porosity were extracted from the rice variety characteristic database as initial features. The initial feature values ​​are normalized to obtain standardized feature vectors; Construct a feature dataset for classification and analyze it to determine the completeness of the feature dataset; If there is missing or abnormal data, the corresponding data will be marked and supplemented to obtain the final feature dataset.

3. The method as described in claim 1 or 2, characterized in that, Also includes: Construct a classification model for water absorption performance; The classification model is used to predict the water absorption performance of the feature dataset and determine the category of rice varieties. Mark data with a confidence level below a preset threshold; The labeled data were re-analyzed based on specific characteristic parameters to adjust the water absorption performance classification results. Based on the adjusted classification results, output the final list of rice varieties classified by water absorption performance.

4. The method for classifying the water absorption performance of rice grains and optimizing moisture distribution as described in claim 1, characterized in that, Also includes: Based on the water absorption performance classification results, and combined with the wax layer barrier coefficient and endosperm pore permeability, an initial analysis dataset was constructed. Extract features of surface water penetration depth and core water diffusion delay to obtain a preliminary feature set; Based on the preliminary feature set, a diffusion equation based on the moisture diffusion coefficient is constructed; The diffusion rate of moisture in rice grains was calculated using the diffusion equation, and the rate distribution data was obtained. Based on the rate distribution data, the matching relationship between surface saturation and core hysteresis is analyzed to determine the distribution of moisture content differences.

5. The method as described in claim 1, characterized in that, Also includes: Obtain initial moisture distribution data inside rice grains; If the difference in the initial moisture distribution data exceeds the preset threshold, the adjustment mechanism is triggered to determine the adjustment range of the heating time; Based on the adjustment range, calculate the change in heat distribution and obtain the dynamic adjustment parameters; Adjust the internal moisture distribution of rice grains and obtain real-time deviation data; Analyze the uniformity of heat distribution to determine the final adjustment scheme.

6. The method as described in claim 3, characterized in that, Also includes: The support vector machine algorithm was applied to classify and train the water absorption performance of rice varieties, and a classification model was constructed.

7. The method as described in claim 4, characterized in that, Also includes: Based on the preliminary feature set, the support vector machine algorithm is used to conduct a correlation analysis on surface water saturation and core water lag, and to determine the distribution pattern of the water content difference between surface saturation and core lag.

8. The method as described in claim 2, characterized in that, It also includes: using a standardization method to numerically normalize the starch water absorption swelling rate, protein water absorption binding force, wax layer barrier coefficient, and endosperm pore permeability to obtain standardized feature vectors.

9. A rice cooker control system applying the method described in claims 1 to 8, characterized in that, Includes a unit for performing the method according to any one of claims 1 to 8.

Images