Intelligent temperature control method and system for baking equipment

Abstract

The invention relates to the technical field of baking equipment, and discloses an intelligent temperature control method and system for baking equipment, and the method comprises the steps: carrying out the real-time collection of dough parameters, internal state parameters and environment parameters in the baking equipment with timestamps, and synchronously calling the regional feature parameters and historical data of the baking equipment, according to the method, actual state changes of the dough in different regional environments are accurately captured, a detailed basis is provided for subsequent temperature control strategy making, the dynamic control parameter set is generated, the baking temperature can be accurately regulated and controlled according to the real-time regional environments, and the baking efficiency is improved. The baking equipment can realize accurate temperature control on the baking process in different regions, ensures that dough is uniformly and consistently heated, finally greatly improves the quality stability of a baked product, ensures that the baked product can achieve an ideal baking effect under different region climate conditions, improves the baking yield, and is suitable for linked baking workshops.

Description

Technical Field

[0001] The present invention relates to the technical field of baking equipment, and particularly relates to an intelligent temperature control method and system for baking equipment. Background Art

[0002] In the field of baking equipment, the current temperature control means mainly rely on a preset temperature control curve or use a single sensor for feedback. The industry generally uses this static temperature control method, presetting a fixed temperature-time curve based on historical experience. This kind of static temperature control sets the corresponding relationship between temperature and time as a fixed program by solidifying historical experience data to achieve the goal of intelligent temperature control.

[0003] Due to being suitable for different baking requirements, baking equipment is often widely used in diverse scenarios, such as chain baking workshops. Chain baking workshops are often located in different cities. When using the traditional static temperature control method for temperature control, it is often affected by the climate in different regions during baking, and it is difficult to achieve adaptive adjustment of intelligent temperature control during the baking process, ultimately resulting in poor baking effects. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides an intelligent temperature control method and system for baking equipment, which solves the above problems.

[0005] The above technical objectives of the present invention are achieved through the following technical solutions:

[0006] An intelligent temperature control method for baking equipment, comprising:

[0007] S1: Real-time collect the dough parameters, internal state parameters and environmental parameters in the baking equipment with timestamps, and synchronously call the regional characteristic parameters and historical data of the baking equipment to form a multi-dimensional original data set;

[0008] S2: Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, and construct time-synchronized real-time multi-modal time series data based on the dough state evolution feature matrix;

[0009] S3: Analyze the real-time multi-modal time series data to generate temperature control parameters, mine the correlation of the temperature control parameters, and generate a temperature control strategy matrix;

[0010] S4: According to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a correction strategy adapted to the regional climate characteristics;

[0011] S5: Solve the correction strategy to generate a dynamic control parameter set;

[0012] S6: Analyze the dynamic control parameter set and construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

[0013] Furthermore, analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, including:

[0014] Extract from the preprocessed multi-dimensional original data set to obtain absorption peak features, and form an initial feature matrix with the absorption peak features;

[0015] Calculate the initial feature matrix to generate regression parameters;

[0016] Verify the regression parameters to obtain the surface moisture activity index;

[0017] Concatenate the surface moisture activity index with environmental parameters and dough parameters to generate a dough state evolution feature matrix.

[0018] Furthermore, construct time-synchronized real-time multi-modal time series data based on the dough state evolution feature matrix, including:

[0019] Conduct time series correlation analysis on the dough state evolution feature matrix to generate a three-dimensional spatio-temporal data cube containing time series correlations, specifically including: align the timestamps of the dough state feature matrices at each moment, use the dynamic time warping algorithm to calculate the time series similarity matrix of the feature dimensions between adjacent matrices, characterize the cross-moment feature dependencies through the autocorrelation coefficient matrix, and then stack the feature matrices with time series correlation weights along the time axis in sequence to construct a three-dimensional spatio-temporal data cube with the number of Number of features Time steps;

[0020] Dynamically weight the three-dimensional spatio-temporal data cube to generate a time series correlation structure;

[0021] According to the time series correlation structure, perform time-aligned synchronization processing on dough parameters, internal state parameters, and environmental parameters to construct time-synchronized real-time multi-modal time series data.

[0022] Furthermore, analyze the real-time multi-modal time series data to generate temperature control parameters, including:

[0023] Perform standardization processing on the real-time multi-modal time series data to obtain the standardized real-time multi-modal time series data;

[0024] Evaluate the importance of the standardized real-time multi-modal time series data to obtain the importance scores of each parameter;

[0025] Calculate the importance threshold based on the internal state parameters of the baking equipment, specifically including: constructing the probability of the sliding window data of the internal state parameters of the baking equipment using kernel density estimation, and directly extracting the 95% quantile of the distribution function as the importance threshold;

[0026] Compare and screen the importance scores of the parameters with the importance threshold to obtain a set of key parameters;

[0027] Associate the set of key parameters with the temperature control parameters of the internal state parameters to generate temperature control parameters.

[0028] Furthermore, mine the relevance of the temperature control parameters to generate a temperature control strategy matrix, including:

[0029] Calculate the correlation degree distribution characteristic matrix of the temperature control parameters, specifically as follows: for various parameters in the temperature control parameters, calculate the linear correlation between pairwise parameters respectively to obtain the correlation coefficient, analyze the correlation coefficient, draw the parameter correlation degree heat map, observe the distribution of the correlation strength and positive and negative correlation between each parameter, and finally form a complete correlation degree distribution characteristic matrix;

[0030] Calculate the strong correlation threshold according to the correlation degree distribution characteristic matrix;

[0031] Compare and screen the correlation degree values in the correlation degree distribution characteristic matrix with the strong correlation threshold to obtain a set of strongly correlated parameter pairs;

[0032] Construct an initial temperature control strategy matrix according to the set of strongly correlated parameter pairs;

[0033] Optimize the initial temperature control strategy matrix to generate the final temperature control strategy matrix.

[0034] Furthermore, according to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a correction strategy suitable for the regional climate characteristics, including:

[0035] Mine the real-time environmental parameters and regional characteristic parameters to construct an initial correlation matrix;

[0036] Combine the initial correlation matrix with the dough moisture activity data in the real-time dough parameters to obtain the humidity deviation compensation coefficient and the temperature fluctuation compensation gradient;

[0037] Fuse the humidity deviation compensation coefficient and the temperature fluctuation compensation gradient to generate a moisture activity change curve;

[0038] Use the moisture activity change curve to correct the temperature control strategy matrix to generate a correction strategy suitable for the regional climate characteristics.

[0039] Further, solve the correction strategy to generate a set of dynamic control parameters, including:

[0040] Perform multi-parameter decoupling analysis based on the correction strategy to generate a set of climate adaptability rules;

[0041] Constrain the set of climate adaptability rules through a geographical feature constraint vector to form a regional adaptive optimization framework;

[0042] Perform global optimization on the regional adaptive optimization framework to generate a candidate set of dynamic compensation parameters;

[0043] Verify the thermodynamic characteristics of the candidate set of dynamic compensation parameters to generate a set of dynamic control parameters.

[0044] Further, analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation, including:

[0045] Extract the set of dynamic control parameters to generate a real-time fluctuation feature sequence;

[0046] Determine the quantile baseline based on kernel density estimation within each window of the real-time fluctuation feature sequence to generate a dynamic fluctuation threshold;

[0047] Construct a comparison mechanism between the real-time fluctuation value and the dynamic fluctuation threshold, and record the over-threshold anomalies and corresponding deviations;

[0048] Calculate the statistical distribution characteristics of the cumulative deviation based on the statistical distribution algorithm to obtain the warning threshold.

[0049] Further, analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation, and also include:

[0050] When the cumulative deviation exceeds the warning threshold, construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

[0051] An intelligent temperature control system for baking equipment, including:

[0052] Acquisition unit: Perform real-time acquisition with timestamps on the dough parameters, internal state parameters, and environmental parameters inside the baking equipment, and synchronously call the geographical feature parameters and historical data of the baking equipment to form a multi-dimensional original data set;

[0053] Analysis unit: Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, and construct time-synchronized real-time multi-modal time series data based on the dough state evolution feature matrix;

[0054] Mining unit: Analyze the real-time multi-modal time series data to generate temperature control parameters, and mine the correlation of the temperature control parameters to generate a temperature control strategy matrix;

[0055] Correction unit: According to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a correction strategy adapted to the regional climate characteristics;

[0056] Solution unit: Solve the correction strategy to generate a set of dynamic control parameters;

[0057] Construction unit: Analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

[0058] In summary, the present invention mainly has the following beneficial effects:

[0059] By collecting a multi-dimensional raw data set such as dough parameters, internal state parameters, and environmental parameters in the baking equipment in real time, and integrating regional characteristic parameters and historical data, various factor changes during the baking process are comprehensively and dynamically sensed. Based on this, the dough state evolution characteristic matrix and time-synchronized real-time multi-modal time series data are constructed, laying a solid foundation for generating accurate temperature control parameters in the subsequent process. Further correlation analysis and strategy matrix construction enable the temperature control strategy to closely fit the actual baking situation. Then, by combining the real-time environmental parameters and regional characteristic parameters to correct the strategy matrix, a correction strategy adapted to the regional climate characteristics is generated. Finally, the obtained set of dynamic control parameters can flexibly adapt to the baking requirements under different regional climate conditions, realizing the adaptive adjustment of intelligent temperature control, effectively solving the problem of poor temperature control effect affected by regional climate in the traditional method, and improving the stability and consistency of baking quality.

[0060] Through the real-time collection with timestamps of multi-dimensional parameters such as dough, internal state of the equipment, and external environment, the comprehensiveness and accuracy of the data are ensured, and the regional characteristic parameters and historical data are synchronously invoked to enrich the data background. In the subsequent processing, by extracting absorption peak characteristics to form an initial characteristic matrix, the surface moisture activity index is obtained through calculation and verification, and the dough state evolution characteristic matrix is generated. This process fully utilizes intelligent data processing technologies to deeply mine the dough state information. When constructing time-synchronized real-time multi-modal time series data, methods such as the dynamic time warping algorithm are used to accurately capture the temporal correlation between parameters, making the data processing more in line with the dynamic characteristics of the actual baking process. For the analysis and correlation mining of temperature control parameters, technical means such as importance evaluation and kernel density estimation are used to screen key parameters and generate temperature control parameters, and then a temperature control strategy matrix is constructed and optimized. The whole process is highly intelligent, reducing the dependence on manual experience and making the generation of the temperature control strategy more reasonable. In addition, the construction of an adaptive intelligent temperature control mechanism based on deviation accumulation further strengthens the intelligent control ability of the baking equipment.

[0061] Determine the importance threshold through the probability construction method of kernel density estimation, screen out the key parameters, and generate temperature control parameters in association with the temperature control parameters. This series of processes ensures the accuracy of the temperature control parameters. During the construction of the temperature control strategy matrix, not only calculate the correlation degree distribution characteristic matrix, but also draw a heat map to visually present the parameter correlation strength and positive and negative correlation distribution. Based on this, set the strong correlation threshold to screen the set of strong correlation parameter pairs, construct and optimize the temperature control strategy matrix, so that the temperature control strategy can accurately reflect the actual correlation between parameters. At the same time, based on the real-time environmental parameters and regional characteristic parameters to correct the temperature control strategy matrix, fully consider the influence of regional climate factors on baking. By constructing the initial correlation matrix, combining the dough moisture activity data to obtain the compensation coefficient and gradient, and then generating the moisture activity change curve for strategy correction, so that the corrected strategy is closer to the actual situation and improves the temperature control accuracy. Through steps such as multi-parameter decoupling analysis, geographical feature constraint vector constraint, and global optimization, a strictly verified dynamic control parameter set is generated to ensure the accuracy of the temperature control parameters in actual applications. And the adaptive intelligent temperature control mechanism based on deviation accumulation can monitor the fluctuations of the dynamic control parameters in real time, use kernel density estimation to determine the dynamic fluctuation threshold, accurately record the abnormal and deviation beyond the threshold, calculate the cumulative deviation characteristics based on the statistical distribution algorithm to obtain the warning threshold, and make adaptive adjustments in time when the cumulative deviation exceeds the warning threshold, further enhancing the accuracy and reliability of the temperature control process, ensuring that the baked products can achieve the ideal baking effect under different regional climate conditions, improving the baked product yield and quality stability, and being applicable to chain baking workshops. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is the flowchart of the intelligent temperature control method for the baking equipment of the present invention;

[0063] Figure 2 is the block diagram of the intelligent temperature control system for the baking equipment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0065] Refer to Figure 1 , an intelligent temperature control method for baking equipment, including:

[0066] S1: Real-time collect the dough parameters, internal state parameters and environmental parameters in the baking equipment with timestamps, and synchronously call the regional characteristic parameters and historical data of the baking equipment to form a multi-dimensional original data set;

[0067] S2: Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, and construct real-time multi-modal time series data with time synchronization based on the dough state evolution feature matrix;

[0068] S3: Analyze the real-time multi-modal time series data to generate temperature control parameters, mine the correlation of the temperature control parameters, and generate a temperature control strategy matrix;

[0069] S4: According to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a corrected strategy adapted to the regional climate characteristics;

[0070] S5: Solve the corrected strategy to generate a set of dynamic control parameters;

[0071] S6: Analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

[0072] Through the real-time acquisition with timestamps of the dough parameters, internal state parameters, and environmental parameters in the baking equipment, and synchronously invoking the regional characteristic parameters and historical data, a multi-dimensional original data set is formed, laying a solid data foundation for precise temperature control. All kinds of key information in the baking process can be comprehensively mastered. Analyzing the preprocessed data set to generate a dough state evolution feature matrix and construct real-time multi-modal time series data can accurately capture the evolution law of the dough state, making the temperature control strategy more in line with the actual change requirements of the dough. Generating temperature control parameters and mining their correlation to form a temperature control strategy matrix, and correcting the strategy in combination with real-time environment and regional characteristic parameters can adapt to different regional climate characteristics, solving the problem that traditional temperature control is greatly affected by regional climate, ensuring intelligent temperature control in different environments. Solving the corrected strategy to generate a set of dynamic control parameters and constructing an adaptive intelligent temperature control mechanism based on deviation accumulation can dynamically adjust the control parameters according to real-time data, realizing precise and adaptive control of the temperature of the baking equipment, and effectively improving the baking quality and efficiency.

[0073] In one case of this embodiment, analyzing the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix includes:

[0074] Extract the preprocessed multi-dimensional original data set to obtain absorption peak features, and form an initial feature matrix with the absorption peak features. Specifically, it includes: analyzing the preprocessed multi-dimensional original data set by using the first derivative method to locate the positions of multiple absorption peaks, and recording the key feature values such as the wavelength position and absorbance magnitude of each absorption peak, and then arranging and combining these extracted absorption peak feature values in order to form an initial feature matrix. The rows of the matrix represent different samples, and the columns of the matrix correspond to each absorption peak;

[0075] Calculate the initial feature matrix to generate regression parameters, specifically including: taking the initial feature matrix as the independent variable and the measured surface water activity index as the dependent variable, extracting latent variables by the PLSR method, calculating the covariance that maximizes the independent variable and the dependent variable, calculating the weight vector of the independent variable through iterative optimization, using the leave-one-out cross-validation method, determining the optimal number of latent variables by evaluating the prediction error, and extracting the weight coefficients of each original feature (corresponding absorption peak) from the optimal number of latent variables after dimensionality reduction to generate regression parameters reflecting the strength of its association with the dependent variable;

[0076] Verify the regression parameters to obtain the surface water activity index. The specific calculation formula is as follows:

[0077] ;

[0078] In the formula, represents the surface water activity index, The value range of is The smaller the value of , the lower the surface water activity of the dough. The larger the value of , the higher the surface water activity of the dough. n represents the number of absorption peak features, represents the th absorption peak feature extracted from the preprocessed multi-dimensional original dataset, represents the regression coefficient corresponding to the th absorption peak feature when calculating the regression parameters, is the comprehensive index of environmental parameters,

[0079] Stitch the surface water activity index with the environmental parameters and dough parameters to generate the dough state evolution feature matrix, specifically including: based on the row alignment principle, according to the sample correspondence, horizontally stitch the surface water activity index, environmental parameters, and dough parameters of each sample using the column vector horizontal stitching method in turn to form a new matrix column, and generate the dough state evolution feature matrix, where the rows of the matrix still represent samples, and the columns correspond to the surface water activity index, environmental parameters, and each dimension feature of the dough parameters respectively;

[0080] The absorption peak features are extracted by the first derivative method to accurately analyze the dough spectral data. The first derivative method can accurately locate the positions of spectral absorption peaks, obtain key characteristic values such as wavelength and absorbance, overcome the subjective errors of traditional manual recognition, ensure the objectivity and consistency of feature extraction. The constructed initial feature matrix has samples as rows and absorption peak features as columns, completely retaining the spectral information. When calculating the regression parameters, the PLSR method effectively solves the problem of multi-variable collinearity, maximizes the covariance by extracting latent variables, and determines the optimal number of latent variables in combination with the leave-one-out cross-validation to avoid overfitting, so that the calculation result of the surface moisture activity index can accurately reflect the change of the dough state;

[0081] By aligning and splicing the surface moisture activity index with environmental and dough parameters row by row, a three-dimensional state evaluation system is constructed. By integrating feature weights, environmental and dough parameters, the dynamic migration of surface moisture in the dough is keenly captured. And the columnar splicing of the matrix breaks the data barrier, integrating spectral features, environmental conditions and basic dough parameters, providing an all-round monitoring perspective, and then deeply mining the potential correlations between parameters, such as the influence of environmental factors on moisture activity and the connection between basic dough parameters and spectral features, so as to ensure the real-time monitoring of the dough state during baking.

[0082] In one case of this embodiment, real-time multimodal time-series data synchronized in time is constructed based on the dough state evolution feature matrix, including:

[0083] Perform time-series correlation analysis on the dough state evolution feature matrix to generate a three-dimensional spatio-temporal data cube containing time-series correlations, specifically including: align the timestamps of the dough state feature matrices at each moment, calculate the time-series similarity matrix of the feature dimensions between adjacent matrices using the dynamic time warping algorithm, characterize the cross-moment feature dependencies through the autocorrelation coefficient matrix, and then stack the feature matrices with time-series correlation weights along the time axis in sequence to construct a three-dimensional spatio-temporal data cube with dimensions (number of features number of features number of time steps), where each spatio-temporal unit stores the feature correlation intensity corresponding to the moment, realizing the time synchronization and spatio-temporal coupling characterization of multimodal time-series data;

[0084] Perform dynamic weighting on the three-dimensional spatio-temporal data cube to generate a time-series correlation structure, specifically including: by constructing a query-key-value matrix in the time dimension, calculating the weight coefficients of the feature correlations at each time step based on scaled dot-product attention, generating a time-series attention matrix after Softmax normalization, multiplying the time-series attention matrix element by element with the feature correlation matrix corresponding to the time step in the cube, realizing the dynamic adjustment of the cross-moment correlation intensity, arranging the feature correlation matrices at each time step after dynamic adjustment in chronological order to generate a dynamic graph structure containing time-series dependence weights, that is, the time-series correlation structure. The nodes of the time-series correlation structure represent feature dimensions, and the edge weights represent the spatio-temporal coupling strength after dynamic weighting;

[0085] According to the temporal correlation structure, perform time-aligned synchronization processing on the dough parameters, internal state parameters, and environmental parameters to construct real-time multimodal temporal data with time synchronization, specifically including: based on the dynamic edge weights of the temporal correlation structure, use the weighted dynamic time warping algorithm to align multi-source parameter sequences, match the original time series of dough parameters, internal state parameters, and environmental parameters with the corresponding feature nodes in the correlation structure respectively, use the edge weights as path constraint weights, calculate the optimal warping path of each modal sequence to the reference time axis, aggregate the highly correlated feature information of adjacent time steps through a graph attention network, use cubic spline interpolation to fill the missing values after alignment, and then use tensor splicing to integrate the multimodal data into a three-dimensional matrix of (time Parameter type parameter value), where the time dimension realizes millisecond-level synchronization through Newton interpolation method, forming a real-time updatable and time-synchronized multimodal temporal data set;

[0086] By constructing real-time multimodal temporal data with time synchronization of the dough state evolution feature matrix, it is possible to achieve accurate monitoring and characterization of the dough state during the baking process. Among them, the generated three-dimensional spatio-temporal data cube finely depicts the dynamic changes of the dough state on the time axis and the correlations between features, providing strong data support for in-depth understanding of the physical and chemical changes of the dough. Secondly, when constructing the temporal correlation structure, the dynamic weighting and attention mechanism are used to effectively capture the changes in the correlation strength of each feature dimension of the dough at different time steps, and can more accurately reflect the actual evolution process of the dough, which is of great significance for optimizing the temperature control in baking and improves the stability of intelligent temperature control;

[0087] By integrating various modal data such as dough parameters, internal state parameters, and environmental parameters, a real-time multimodal temporal data set with time synchronization is constructed. On the one hand, technologies such as the weighted dynamic time warping algorithm and the graph attention network are used to solve the problem of time alignment of multi-source parameters, ensure the accurate matching of different modal data on the time axis, and completely retain the spatio-temporal coupling relationship between data. On the other hand, means such as cubic spline interpolation and tensor splicing are used to not only fill the missing values after alignment, but also realize the efficient integration and millisecond-level synchronous update of data, providing a high-quality data basis for real-time monitoring and analysis of the dough state, which provides comprehensive and real-time dough state information for the intelligent temperature control of baking equipment, assists it in making decisions quickly, and thus improves the baking efficiency.

[0088] In one case of this embodiment, analyze the real-time multimodal temporal data to generate temperature control parameters, including:

[0089] Perform standardization processing on the real-time multimodal temporal data to obtain the standardized real-time multimodal temporal data;

[0090] Perform importance assessment on the standardized real-time multimodal time-series data to obtain the importance scores of each parameter. Specifically, for the standardized real-time multimodal time-series data, use the locally weighted linear regression algorithm for importance assessment. First, determine the weight values according to the sample time-series distance of the real-time multimodal time-series data. The closer to the target sample, the greater the weight. Then, introduce the Gaussian kernel function, estimate the parameters by the least squares method, solve the objective function of minimizing the weighted mean square error, and obtain the regression coefficients of each parameter. Use the recursive feature elimination algorithm to screen features according to the magnitude of the regression coefficients. The greater the absolute value of the regression coefficient, the higher the importance score of the feature, so as to obtain the importance scores of each parameter;

[0091] Calculate the importance threshold according to the internal state parameters of the baking equipment. Specifically, use kernel density estimation to construct the probability of the sliding window data of the internal state parameters (temperature, pressure) of the baking equipment, and directly extract the 95% quantile of the distribution function as the importance threshold. This threshold ensures that only significant features are retained at the 95% confidence level by dynamically capturing the boundaries of the high-density intervals of the state parameters, realizing the strong correlation and self-adaptation between the threshold and the real-time working conditions of the baking equipment;

[0092] Compare and screen the importance scores of the parameters with the importance threshold to obtain the key parameter set. Specifically, compare the importance scores of each parameter with the importance threshold. If the importance score of the parameter is greater than the importance threshold, it indicates that this parameter is a significant feature and is included in the key parameter set; if the importance score of the parameter is less than or equal to the importance threshold, the parameter is excluded. After this screening, finally obtain the key parameter set adapted to the real-time working conditions of the baking equipment;

[0093] Associate the key parameter set with the temperature control parameter of the internal state parameter to generate the temperature control parameter. The specific calculation formula is as follows:

[0094] ;

[0095] In the formula, represents the temperature control parameter, represents the importance score of the th parameter, K represents the key parameter set, represents the temperature control parameter of the internal state parameter, represents the internal state parameter of the baking equipment;

[0096] Through the standardization process and dynamic feature screening of real-time multimodal time-series data, the improvement of data quality and the adaptive extraction of key parameters are achieved. The standardization process eliminates the dimensional differences and distribution offsets of multimodal parameters, making heterogeneous time-series data such as temperature and pressure comparable, establishing a unified benchmark for subsequent analysis, and avoiding evaluation biases caused by parameter unit or magnitude differences. On this basis, a combined method of locally weighted linear regression and recursive feature elimination is adopted. The Gaussian kernel function is used to dynamically assign sample weights. While retaining the local characteristics of time series, the regression coefficients are iteratively optimized by the least squares method to accurately capture the dynamic influence weights of parameters at different time points on temperature control. Compared with traditional fixed-window feature selection, the weighted mechanism strengthens the time-series correlation of neighboring samples, reducing the feature importance evaluation error. Combined with the recursive feature elimination algorithm, using the absolute value of the regression coefficient as the judgment basis, redundant parameter interferences are effectively eliminated, ensuring the dynamic update of the key parameter set and its strong correlation with the operating state of the baking equipment, laying a high-precision data foundation for the generation of subsequent temperature control parameters;

[0097] Through the dynamic threshold modeling of kernel density estimation and the multi-dimensional parameter fusion calculation, the adaptive optimization of temperature control parameters and the precise matching of equipment operating conditions are achieved. Based on the sliding window kernel density estimation method, dynamic probability modeling is performed on state parameters such as temperature and pressure. By extracting the 95% quantile threshold, the limitations of traditional fixed thresholds are broken through, enabling the feature screening boundary to automatically adjust with the changes in the operating conditions of the baking equipment. This threshold mechanism significantly improves the robustness of feature screening under abnormal conditions such as high temperature and high pressure by capturing the high-density distribution intervals of state parameters in real time. In the stage of generating temperature control parameters, the importance scores of key parameters are coupled with the internal state parameters of the baking equipment, which not only strengthens the decision-making weights of high-importance parameters but also balances the influence of the real-time state of the baking equipment through the normalization process of the denominator term, achieving the coordinated improvement of energy efficiency and product quality in the baking process.

[0098] In one case of this embodiment, the correlation of temperature control parameters is mined to generate a temperature control strategy matrix, including:

[0099] Calculate the correlation degree of temperature control parameters to obtain a correlation degree distribution feature matrix, specifically including: for the operation data of various parameters such as temperature, humidity, and equipment power in temperature control parameters, calculate the linear correlation between pairwise parameters respectively to obtain the correlation coefficient. By analyzing the correlation coefficient, draw a parameter correlation degree heat map, observe the correlation strength and positive and negative correlation distribution between parameters, and finally form a complete correlation degree distribution feature matrix;

[0100] For any two parameters XA and YA, calculate the correlation coefficient of the two parameters , and the calculation formula is as follows:

[0101] ;

[0102] Among them, represents the number of data samples, and respectively represent the th sample values of any two parameters XA and YA, represents index variable of and respectively represent the sample means of any two parameters XA and YA, The value range of

[0103] Set the color mapping rule: establish the correspondence between the correlation coefficient and color, and stipulate that the positive correlation coefficient (0 1) uses the red color system, and the closer the correlation coefficient is to 1, the darker the red; the negative correlation coefficient (-1 0) uses the blue color system, and the closer the correlation coefficient is to -1, the darker the blue; when the correlation coefficient is close to 0 (such as 0.2), use light color or white;

[0104] Create a two-dimensional table, arrange the temperature control parameters (such as temperature, humidity, equipment power, etc.) in the horizontal and vertical columns of the table respectively. For each cell in the table, the two parameters corresponding to the row-column intersection are filled with the corresponding color according to the calculated correlation coefficient and the color mapping rule, so as to form a heat map of parameter correlation degree. By observing the color distribution of the heat map, intuitively judge the correlation strength and positive and negative correlation between parameters. The darker the color of the cell, the higher the correlation strength between the corresponding parameters. The red cell represents a positive correlation between the two parameters, and the blue cell represents a negative correlation between the two parameters. The lighter the color of the cell, the weaker the correlation strength between the corresponding parameters. Recreate a two-dimensional matrix, and the rows and columns are also identified by temperature control parameters. Fill the correlation coefficients between each two parameters calculated before into the corresponding cells of the matrix to form a complete correlation degree distribution characteristic matrix;

[0105] According to the correlation degree distribution characteristic matrix, calculate the strong correlation threshold, specifically including: obtain the correlation degree data of all parameter pairs from the correlation degree distribution characteristic matrix, calculate the mean and standard deviation of the correlation degrees of all parameter pairs, and set the strong correlation threshold as the mean plus twice the standard deviation according to the normal distribution characteristics to obtain the strong correlation threshold;

[0106] Compare and screen the correlation values in the correlation degree distribution feature matrix with the strong correlation threshold to obtain a set of strong correlation parameter pairs, which specifically includes: traversing the correlation degree distribution feature matrix, extracting the correlation values of each parameter pair one by one, comparing them with the strong correlation threshold one by one, screening out the parameter pairs whose correlation values are greater than or equal to the strong correlation threshold, and integrating these qualified parameter pairs into a set to construct a set of strong correlation parameter pairs;

[0107] Construct an initial temperature control strategy matrix based on the set of strong correlation parameter pairs, which specifically includes: determining the matrix dimension, where the rows represent temperature and the columns correspond to parameter pairs, extracting the correlation characteristics of the parameter pairs in the set of strong correlation parameter pairs, assigning positive values for positive correlation, negative values for negative correlation, and zero for no correlation, grading according to the correlation strength, assigning high weights for strong correlation, medium weights for medium correlation, and discarding weak correlation, and then filling the matrix elements to construct an initial temperature control strategy matrix;

[0108] Optimize the initial temperature control strategy matrix to generate a final temperature control strategy matrix, which specifically includes: when optimizing the initial temperature control strategy matrix, based on the actual operation data of the baking equipment, continuously collect temperature-related data, including set temperature, actual temperature, temperature fluctuation range, and equipment operation parameters, etc., compare these data with the target temperature of the initial temperature control strategy matrix to determine the temperature deviation situation, calculate the error between the actual temperature and the target temperature using the mean square error, and use this as a basis to determine the adjustment direction and step size of the matrix weights, apply the gradient descent algorithm, calculate the gradient of the error with respect to each weight, thereby determining the adjustment direction and step size of the weights, and then update each weight in the initial temperature control strategy matrix, apply the updated matrix to the actual operation of the baking equipment, observe whether the temperature of the baking equipment can approach the target temperature faster and whether the temperature fluctuation is reduced, collect operation data again, continuously repeat the above process, adjust the matrix weights until the temperature control reaches the expected stability and response speed, and then use the initial temperature control strategy matrix that shows good temperature control effect as the temperature control strategy matrix to achieve long-term stable temperature control;

[0109] By deeply exploring the complex relationships among temperature control parameters, calculating the pairwise linear correlations of various parameters such as temperature, humidity, and equipment power, comprehensive and accurate correlation coefficients are obtained, forming a correlation degree distribution characteristic matrix. According to the characteristics of the normal distribution, a strong correlation threshold is set, and a set of strongly correlated parameter pairs is screened out, making the construction of the temperature control strategy matrix targeted. Further, an initial temperature control strategy matrix is constructed based on the set of strongly correlated parameter pairs, and corresponding weights are assigned according to different correlation characteristics, initially realizing the effective integration of the mutual influences among temperature control parameters. In the optimization stage, combined with actual operation data, advanced optimization methods such as the gradient descent algorithm are used to continuously adjust the matrix weights so that it can accurately adapt to the actual temperature control requirements. The finally generated temperature control strategy matrix can significantly improve the stability and response speed of the temperature control system, effectively reduce temperature fluctuations, make the temperature control of baking equipment more accurate and efficient, provide a stable and reliable temperature environment for the baking process, and improve the quality and production efficiency of baked products;

[0110] By setting color mapping rules, the abstract correlation coefficients are transformed into intuitive color information, enabling staff to quickly and clearly observe the correlation strength and positive / negative correlation between each temperature control parameter. The depth change of the color intuitively reflects the tightness of the correlation between parameters, and the distinct contrast between red and blue clearly indicates the positive / negative correlation relationship, greatly reducing the difficulty and complexity of data analysis, facilitating technicians to quickly identify key parameter pairs, providing a powerful visual basis for formulating effective temperature control strategies. At the same time, from the screening process of the correlation degree distribution characteristic matrix to the set of strongly correlated parameter pairs, and then to the construction and optimization of the temperature control strategy matrix, the entire process constructs a systematic and intelligent temperature control decision support system that can dynamically adjust the strategy according to actual data, realizing the adaptive adjustment of temperature control during the operation of baking equipment.

[0111] In one case of this embodiment, according to real-time environmental parameters and regional characteristic parameters, the temperature control strategy matrix is corrected to generate a corrected strategy adapted to the regional climate characteristics, including:

[0112] Explore the real-time environmental parameters and regional characteristic parameters to construct an initial correlation matrix, specifically including: preprocess the real-time environmental parameters (such as temperature, humidity, air pressure, etc.) and regional characteristic parameters (such as annual average temperature, monsoon intensity, precipitation distribution, etc.), remove noise and outliers, use the principal component analysis algorithm to explore key features, and standardize the data to the interval to eliminate the dimension difference, construct an initial correlation matrix with regions as rows and parameters as columns, calculate the correlation values of each parameter with the target climate characteristic through the Pearson correlation coefficient, and fill the matrix, thereby constructing an initial correlation matrix that can reflect the correlation between regional climate characteristics and environmental parameters;

[0113] Combine the initial correlation matrix with the dough moisture activity data in the real-time dough parameters to obtain the humidity deviation compensation coefficient and the temperature fluctuation compensation gradient, specifically including: perform normalization processing on the real-time dough moisture activity data to obtain a normalized dough moisture activity matrix, perform matrix multiplication on the humidity-related rows in the initial correlation matrix and the normalized dough moisture activity matrix to obtain the humidity deviation compensation coefficient, and perform gradient descent calculation on the temperature-related rows in the initial correlation matrix and the normalized dough moisture activity matrix to obtain the temperature fluctuation compensation gradient;

[0114] Fuse the humidity deviation compensation coefficient and the temperature fluctuation compensation gradient to generate a moisture activity change curve, specifically including: adopt the weighted average method, multiply the humidity deviation compensation coefficient by its corresponding weight coefficient (0.6), multiply the temperature fluctuation compensation gradient by its corresponding weight coefficient (0.4), add the two results to obtain a fusion value, use the time as the horizontal axis and the fusion value as the vertical axis, and use the cubic spline interpolation algorithm to interpolate the discrete fusion values on the time axis to generate a continuous moisture activity change curve;

[0115] Use the moisture activity change curve to correct the temperature control strategy matrix to generate a corrected strategy adapted to the regional climate characteristics, specifically including: extract the fusion values at each time point in the moisture activity change curve, adopt the linear mapping algorithm, decompose the fusion values into a temperature correction amount and a humidity correction amount according to a preset proportionality coefficient (temperature correction coefficient 0.3, humidity correction coefficient 0.7), and through matrix dot multiplication operation, apply the temperature correction amount and the humidity correction amount at the corresponding time points to the temperature control column and the humidity control column of the temperature control strategy matrix respectively to complete the row-by-row correction of the matrix elements and generate a corrected strategy adapted to the regional climate characteristics;

[0116] By deeply integrating the real-time environmental parameters and the regional characteristic parameters, construct an initial correlation matrix that accurately reflects the correlation between the regional climate and environmental parameters, and further generate a moisture activity change curve, providing a strong basis for the correction of the temperature control strategy matrix. In the data mining and analysis stage, use algorithms such as principal component analysis and Pearson correlation coefficient to effectively remove noise and outliers, and accurately extract key features, ensuring the accuracy and reliability of the data. Based on this, the generated moisture activity change curve can accurately capture the dynamic trend of the dough moisture activity with the change of the environment, making the corrected temperature control strategy matrix more suitable for the dough fermentation requirements under the actual regional climate conditions, significantly improving the pertinence and effectiveness of the temperature control strategy, and thus providing more accurate temperature and humidity control for the dough fermentation process, which is beneficial to improving the stability and consistency of the dough fermentation quality;

[0117] By implementing dynamic adjustment of intelligent temperature control strategies, the adaptability of the dough fermentation process to regional climates is greatly enhanced. The humidity deviation compensation coefficient and temperature fluctuation compensation gradient are integrated to generate a water activity change curve, and this curve is used to correct the temperature control strategy matrix, enabling the temperature control strategy to respond in real time to changes in environmental parameters and regional climates. This dynamic correction mechanism avoids the inadaptability problems that may occur with traditional fixed temperature control strategies in different regional environments, reduces the risk of dough baking failure caused by climate differences, and improves the success rate and efficiency of fermentation. At the same time, data processing technologies are adopted to ensure the scientificity and rationality of the correction process, providing a stable and accurate temperature and humidity environment for dough baking, which helps to improve the quality and taste of dough products and meet the requirements of dough baking processes in different regions.

[0118] In one case of this embodiment, the correction strategy is solved to generate a set of dynamic control parameters, including:

[0119] Based on the correction strategy, multi-parameter decoupling analysis is implemented to generate a set of climate adaptability rules, specifically including: introducing a feature extraction algorithm to accurately extract the parameter features in the correction strategy to determine the key influencing factors, applying the principal component analysis algorithm to decouple the key parameters to separate independent multi-parameters, analyzing the performance of the decoupled parameters under different climate conditions through a clustering algorithm, and finally, based on the analysis results of the parameters, using a decision tree algorithm to construct and generate a set of climate adaptability rules;

[0120] The set of climate adaptability rules is constrained by a geographical feature constraint vector to form a regional adaptive optimization framework, specifically including: quantifying and normalizing geographical feature parameters (such as latitude, altitude, annual average temperature, etc.) to construct a geographical feature constraint vector, embedding the geographical feature constraint vector as a prior condition into the decision space of the set of climate adaptability rules, and using a constrained non-linear programming algorithm with the compatibility between the control parameters output by the rule set and geographical features as the objective function, solving the optimal solution through the Lagrange multiplier method, and iteratively adjusting the parameter thresholds and weight coefficients in the rule set. Finally, a regional adaptive optimization framework including geographical feature constraints is formed;

[0121] Global optimization is performed on the regional adaptive optimization framework to generate a candidate set of dynamic compensation parameters, specifically including: using the genetic algorithm for global optimization with the degree of fit between the control parameters output by the regional adaptive optimization framework and actual requirements as the objective function. In the genetic algorithm, the control parameters are converted into a chromosome coding method, and through selection, crossover, and mutation operations, iterative calculations are performed to calculate the fitness of each generation of individuals, and individuals with high fitness are selected. At the same time, the simulated annealing algorithm is introduced to jump out of the local optimum. After multiple iterations, parameters are finally extracted from high-quality individuals to form a candidate set of dynamic compensation parameters;

[0122] Verify the thermodynamic characteristics of the dynamic compensation parameter candidate set to generate a dynamic control parameter set, specifically including: taking the candidate parameters in the dynamic compensation parameter candidate set as control variables, evaluating the thermodynamic responses under different compensation parameters through the parameter response simulation algorithm, collecting the actual temperature distribution data using an infrared thermal imager, and calculating the root mean square error (RMSE < 0.5°C as the threshold) between the two to verify the thermal balance accuracy. Use the Monte Carlo sampling technique (sampling times: 1000 times) to sample and analyze the dynamic compensation parameter candidate set to test its robustness. For the parameters that do not meet the standards, use the sequential quadratic programming algorithm to optimize and adjust them under the condition of satisfying the Lagrange multiplier constraints. Finally, verify the thermodynamic stability under the optimized parameters through the time series analysis algorithm, and screen out the parameters that meet the accuracy and stability requirements, thereby generating a dynamic control parameter set;

[0123] Through the fine multi-parameter decoupling analysis and the comprehensive application of a variety of advanced algorithms, the depth optimization of the correction strategy is realized. First, introduce the feature extraction algorithm to accurately locate the key influencing factors, and then use the principal component analysis algorithm to decouple the key parameters to separate the independent multi-parameters, which lays the foundation for accurately constructing the rule set according to different climate conditions. Then, use the clustering algorithm to clearly present the performance of each parameter under different climate conditions, and the decision tree algorithm constructs a climate adaptability rule set based on the analysis results. In addition, embed the geographical feature constraint vector into the rule set decision space and use the constrained nonlinear programming algorithm to solve the optimal solution, so that the optimization framework can fully adapt to the geographical features of a specific region, greatly improving the regional adaptability and accuracy of the control parameters, and solving the problem that the previous control strategies are difficult to be evenly applicable in different climate and geographical environments, providing a solid support for realizing refined dynamic control;

[0124] By using the genetic algorithm for global optimization, convert the control parameters into chromosome coding, and continuously iterate through operations such as selection, crossover, and mutation, effectively improving the fit between the control parameters and the actual requirements, avoiding the limitations of local optimal solutions. At the same time, introduce the simulated annealing algorithm to further jump out of the local optimum and make the optimization process more comprehensive and effective. In the subsequent thermodynamic characteristics verification link, use methods such as the parameter response simulation algorithm, comparison of measured data, and Monte Carlo sampling technique to strictly screen and test the robustness of the dynamic compensation parameter candidate set, ensuring that the finally generated dynamic control parameter set can meet high-standard requirements in terms of thermal balance accuracy and stability. For the parameters that do not meet the standards, use the sequential quadratic programming algorithm for optimization and adjustment, and finally verify its thermodynamic stability through the time series analysis algorithm. This series of rigorous verification and optimization processes greatly enhance the reliability and practicality of the control parameters.

[0125] In a case of this embodiment, the dynamic control parameter set is analyzed to construct an adaptive intelligent temperature control mechanism based on deviation accumulation, including:

[0126] Extract the dynamic control parameter set to generate a real-time fluctuation feature sequence, specifically including: extracting key parameters from the dynamic control parameter set, including but not limited to temperature, pressure, flow rate, etc., calculating the mean value of the key parameters using a sliding window algorithm, taking a window of 5 sample points, using the difference algorithm to obtain the difference between adjacent sample points, generating a difference sequence, and then based on this difference sequence, calculating its standard deviation, taking the square root of the sum of the squares of each element in the difference sequence minus its mean value, and dividing by the sample size minus one to quantify the degree of fluctuation, and finally forming a real-time fluctuation feature sequence.

[0127] Determine the quantile baseline within each window of the real-time fluctuation feature sequence to generate a dynamic fluctuation threshold, specifically including: within each window of the real-time fluctuation feature sequence, using the kernel density estimation algorithm to fit the data distribution with a Gaussian kernel function to obtain a probability density function, calculating the cumulative distribution function of the probability density function through numerical integration, using the binary search algorithm to determine the value corresponding to a cumulative probability of 0.95 in the cumulative distribution function as the 95% quantile baseline. Since the data in each window is different, the fitted distribution and the calculated baseline change accordingly. Combining the standard deviation of the data within the window, generating upper and lower fluctuation thresholds around the dynamic baseline based on the outlier determination criterion of the normal distribution, and then generating a dynamic fluctuation threshold;

[0128] Construct a comparison mechanism between the real-time fluctuation value and the dynamic fluctuation threshold, record the over-threshold anomalies and the corresponding deviations, specifically including: using a numerical comparison algorithm to compare each fluctuation value in the real-time fluctuation feature sequence with the corresponding dynamic fluctuation threshold one by one. If the fluctuation value exceeds the upper threshold or is lower than the lower threshold, it is determined as an over-threshold anomaly, record the anomaly time and the deviation between the fluctuation value and the threshold, and at the same time count the number of over-threshold anomalies. Among them, calculate the absolute difference between this fluctuation value and the corresponding threshold (the upper threshold if it exceeds the upper limit, and the lower threshold if it is lower than the lower limit), which is the deviation;

[0129] Calculate the statistical distribution characteristics of the cumulative deviation based on the statistical distribution algorithm to obtain a warning threshold. The specific calculation formula is as follows:

[0130] ;

[0131] In the formula, represents the warning threshold, represents the length of the real-time fluctuation feature sequence, represents the th feature in the real-time fluctuation feature sequence, represents the cumulative deviation, and respectively corresponding to and weight coefficients, indicating the number of times of over-threshold anomalies;

[0132] Through refined real-time fluctuation feature analysis, the precise capture and quantitative evaluation of the dynamic changes of temperature control parameters are realized. By using the sliding window algorithm and differential calculation, the real-time fluctuation features of key parameters such as temperature and pressure can be effectively extracted. The fluctuation degree is quantified by the standard deviation to form a feature sequence containing the details of parameter dynamic changes. Combining the dynamic fluctuation threshold constructed by kernel density estimation, the baseline can be adaptively adjusted according to the actual distribution of each window data, avoiding the problem of insufficient adaptability of traditional fixed thresholds to complex fluctuation scenarios. This dynamic threshold generation method based on data distribution characteristics can not only sensitively identify abnormal fluctuations but also filter out minor perturbations within the normal range, providing a scientific and reliable benchmark for subsequent anomaly determination and significantly improving the dynamic adaptability of the temperature control system to the non-linear and time-varying baking process;

[0133] By combining the comparison mechanism with the statistical distribution algorithm, a deviation accumulation early warning system with self-learning ability is constructed. By comparing the fluctuation value with the dynamic threshold in real time, the over-threshold anomalies can be immediately captured and the deviation data can be recorded, providing precise anomaly positioning and quantitative evaluation for the system. The early warning threshold calculated based on the statistical distribution characteristics weights and calculates multi-dimensional information such as real-time fluctuation features, accumulated deviation, and the number of anomalies, enabling the early warning strategy to be dynamically adjusted according to the actual deviation state of the baking process. This design that combines data statistical laws with real-time deviation feedback not only realizes the cumulative correction of temperature control deviation but also endows the system with the ability of "experience learning", enabling it to continuously optimize the early warning strategy during long-term operation, reducing the frequency of manual intervention, and improving the autonomous control accuracy and stability of the equipment in complex baking environments, providing an intelligent deviation control guarantee for high-quality baking.

[0134] In one case of this embodiment, analyzing the dynamic control parameter set and constructing an adaptive intelligent temperature control mechanism based on deviation accumulation further includes:

[0135] When the accumulated deviation exceeds the early warning threshold, constructing an adaptive intelligent temperature control mechanism based on deviation accumulation, specifically including: when the real-time fluctuation value is higher than the upper limit of the dynamic fluctuation threshold, it is recorded as a positive deviation (requiring temperature reduction); if is lower than the lower limit of the dynamic fluctuation threshold, it is recorded as a negative deviation (requiring temperature increase). When the real-time fluctuation value is within the dynamic fluctuation threshold, no adjustment is made;

[0136] The adaptive intelligent temperature control mechanism is divided into three levels, as follows:

[0137] Mild: When the absolute value of the ratio of the accumulated deviation to the early warning threshold Within 20%, the temperature set value is fine-tuned by 1°C each time. When there is a positive deviation, the set temperature is decreased to avoid over-baking. When there is a negative deviation, the set temperature is increased to ensure the baking progress.

[0138] Medium: When the absolute value of the ratio of the cumulative deviation to the warning threshold is between 21% - 50%, the power of the heating element is adjusted according to the deviation ratio. When there is a positive deviation, the power is decreased by 15% - 30% to avoid continuous overheating. When there is a negative deviation, the power is increased by 15% - 30% to reach the target temperature as soon as possible.

[0139] Severe: When the absolute value of the ratio of the cumulative deviation to the warning threshold is greater than 50%, it enters the emergency mode. When there is a positive deviation, some heating elements are turned off to quickly cool down. When there is a negative deviation, all heating elements operate at full power to quickly heat up.

[0140] For the problem of difficult temperature control adaptation and poor baking effect caused by climate differences when the chain baking workshop is applied across regions, through the hierarchical deviation response strategy, according to the ratio of the cumulative deviation to the warning threshold, it is divided into three-level adjustment modes: mild, medium, and severe, which accurately adapt to the temperature control requirements in different regional environments. In the humid southern region, if the temperature fluctuates due to high environmental humidity, it is fine-tuned by 1°C in the case of mild deviation to avoid excessive dough fermentation. In the dry northern environment, the heating power is adjusted proportionally in the case of medium deviation to prevent overshoot or energy waste. When extreme climate causes severe deviation, the heating elements are quickly turned on and off in the emergency mode to quickly correct the temperature, so that the equipment can always maintain stable temperature control accuracy under complex and variable regional climate conditions, effectively solving the problem of insufficient adaptability of traditional technologies and ensuring that the baking quality transcends regional limitations.

[0141] Reference Figure 2 , an intelligent temperature control system for baking equipment, comprising:

[0142] Acquisition unit: Real-time acquisition with time stamps of the dough parameters, internal state parameters, and environmental parameters in the baking equipment, and synchronously call the regional characteristic parameters and historical data of the baking equipment to form a multi-dimensional original data set.

[0143] Analysis unit: Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, and construct real-time multi-modal time series data synchronized with time based on the dough state evolution feature matrix.

[0144] Mining unit: Analyze the real-time multi-modal time series data to generate temperature control parameters, and mine the correlation of the temperature control parameters to generate a temperature control strategy matrix.

[0145] Correction unit: According to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a correction strategy adapted to the regional climate characteristics.

[0146] Solution unit: Solve the correction strategy to generate a set of dynamic control parameters;

[0147] Construction unit: Analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

[0148] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent temperature control method for a baking device, characterized in that, Including: S1: Real-time collect the dough parameters, internal state parameters, and environmental parameters in the baking equipment with timestamps, and synchronously call the regional characteristic parameters and historical data of the baking equipment to form a multi-dimensional original data set; S2: Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, and construct time-synchronized real-time multi-modal time series data based on the dough state evolution feature matrix; S3: Analyze the real-time multi-modal time series data to generate temperature control parameters, mine the correlation of the temperature control parameters, and generate a temperature control strategy matrix; S4: According to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a corrected strategy adapted to the regional climate characteristics; S5: Solve the corrected strategy to generate a dynamic control parameter set; S6: Analyze the dynamic control parameter set to construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

2. The intelligent temperature control method for a baking device according to claim 1, wherein, Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution feature matrix, including: Extract the absorption peak characteristics from the preprocessed multi-dimensional original data set, and form an initial feature matrix with the absorption peak characteristics; Calculate the regression parameters for the initial feature matrix; Verify the regression parameters to obtain the surface moisture activity index; Concatenate the surface moisture activity index with the environmental parameters and dough parameters to generate a dough state evolution feature matrix.

3. The intelligent temperature control method for a baking device according to claim 2, wherein, Construct time-synchronized real-time multi-modal time series data based on the dough state evolution feature matrix, including: Perform time series correlation analysis on the dough state evolution feature matrix to generate a three-dimensional spatio-temporal data cube containing time series correlations, specifically including: align the timestamps of the dough state feature matrices at each moment, calculate the time series similarity matrix of the feature dimensions between adjacent matrices using the dynamic time warping algorithm, characterize the cross-moment feature dependencies through the autocorrelation coefficient matrix, and then stack the feature matrices with time series correlation weights along the time axis in turn to construct a three-dimensional spatio-temporal data cube with the number of features X the number of features X the number of time steps; Perform dynamic weighting on the three-dimensional spatio-temporal data cube to generate a time series correlation structure; According to the time series correlation structure, perform time-aligned synchronization processing on the dough parameters, internal state parameters, and environmental parameters to construct time-synchronized real-time multi-modal time series data.

4. The intelligent temperature control method for a baking device according to claim 3, characterized in that, Analyze the real-time multi-modal time series data to generate temperature control parameters, including: Perform standardization processing on the real-time multi-modal time series data to obtain the standardized real-time multi-modal time series data; Perform importance evaluation on the standardized real-time multi-modal time series data to obtain the importance scores of each parameter; According to the internal state parameters of the baking equipment, calculate the importance threshold, specifically including: use kernel density estimation to construct the probability of the sliding window data of the internal state parameters of the baking equipment, and directly extract the 95% quantile of the distribution function as the importance threshold; Compare and screen the importance scores of the parameters with the importance threshold to obtain a set of key parameters; Associate the set of key parameters with the temperature control parameters of the internal state parameters to generate temperature control parameters.

5. The intelligent temperature control method for a baking device according to claim 4, characterized in that, Mine the correlation of the temperature control parameters to generate a temperature control strategy matrix, including: Calculate the correlation degree of temperature control parameters to obtain the correlation degree distribution characteristic matrix, which is specifically as follows: For various types of parameters in the temperature control parameters, calculate the linear correlation between each pair of calculation parameters respectively to obtain the correlation coefficient. By analyzing the correlation coefficient, draw the parameter correlation degree heat map, observe the distribution of the correlation strength and positive and negative correlation between each parameter, and finally form a complete correlation degree distribution characteristic matrix; Calculate the strong correlation threshold according to the correlation degree distribution characteristic matrix; Compare and screen the correlation degree values in the correlation degree distribution characteristic matrix with the strong correlation threshold to obtain the set of strong correlation parameter pairs; Construct an initial temperature control strategy matrix according to the set of strong correlation parameter pairs; Optimize the initial temperature control strategy matrix to generate the final temperature control strategy matrix.

6. The intelligent temperature control method for a baking device according to claim 5, wherein, According to the real-time environmental parameters and regional characteristic parameters, correct the temperature control strategy matrix to generate a correction strategy adapted to the regional climate characteristics, including: Mine the real-time environmental parameters and regional characteristic parameters to construct an initial correlation matrix; Combine the initial correlation matrix with the dough moisture activity data in the real-time dough parameters to obtain the humidity deviation compensation coefficient and the temperature fluctuation compensation gradient; Fuse the humidity deviation compensation coefficient and the temperature fluctuation compensation gradient to generate a moisture activity change curve; Use the moisture activity change curve to correct the temperature control strategy matrix to generate a correction strategy adapted to the regional climate characteristics.

7. An intelligent temperature control method for a baking device according to claim 6, characterized in that, Solve the correction strategy to generate a set of dynamic control parameters, including: Implement multi-parameter decoupling analysis based on the correction strategy to generate a set of climate adaptation rules; Constrain the set of climate adaptation rules through the geographical feature constraint vector to form a regionally adaptive optimization framework; Perform global optimization on the regionally adaptive optimization framework to generate a candidate set of dynamic compensation parameters; Verify the thermodynamic characteristics of the candidate set of dynamic compensation parameters to generate a set of dynamic control parameters.

8. An intelligent temperature control method for a baking device according to claim 7, characterized in that, Analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation, including: Extract the set of dynamic control parameters to generate a real-time fluctuation feature sequence; Determine the quantile baseline based on kernel density estimation within each window of the real-time fluctuation feature sequence to generate a dynamic fluctuation threshold; Construct a comparison mechanism between the real-time fluctuation value and the dynamic fluctuation threshold, and record the over-threshold anomalies and corresponding deviations; Calculate the statistical distribution characteristics of the cumulative deviation based on the statistical distribution algorithm to obtain the warning threshold.

9. The intelligent temperature control method for a baking device according to claim 8, characterized in that Analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation, and also include: When the cumulative deviation exceeds the warning threshold, construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

10. An intelligent temperature control system for a baking device, which is applied to the intelligent temperature control method for a baking device according to any one of claims 1-9, and is characterized in that, Including: Acquisition unit: Real-time collect the dough parameters, internal state parameters and environmental parameters in the baking equipment with timestamps, and synchronously call the regional characteristic parameters and historical data of the baking equipment to form a multi-dimensional original data set; Analysis unit: Analyze the preprocessed multi-dimensional original data set to generate a dough state evolution characteristic matrix, and construct time-synchronized real-time multi-modal time series data based on the dough state evolution characteristic matrix; Mining unit: Analyze the real-time multi-modal time series data to generate temperature control parameters, mine the correlation of the temperature control parameters, and generate a temperature control strategy matrix; Correction Unit: Correct the temperature control strategy matrix according to real-time environmental parameters and regional characteristic parameters to generate a correction strategy adapted to the regional climate characteristics; Solution Unit: Solve the correction strategy to generate a set of dynamic control parameters; Construction Unit: Analyze the set of dynamic control parameters to construct an adaptive intelligent temperature control mechanism based on deviation accumulation.

Images