A method and system for controlling a wafer processing apparatus based on big data modeling

Abstract

A control method and system for tart crust processing equipment based on big data modeling, belonging to the field of food industry equipment control technology, includes the following steps: obtaining dough kneading and proofing process data based on batch identifiers; obtaining pre-processing state data of the stamping equipment for the target dough stamping process, and combining the kneading and proofing process data with a trained stamping quality prediction model to obtain a predicted quality value of the finished tart crust after stamping; comparing the predicted quality value with a preset quality target to determine whether the preset stamping process parameters need adjustment; if so, adjusting the preset stamping process parameters to obtain the execution stamping process parameters; and controlling the stamping equipment to perform the stamping operation on the target dough according to the execution stamping process parameters. This approach achieves forward-looking prediction of the finished tart crust quality and enables timely intervention and adjustment, ultimately improving the stability and consistency of product quality.

Description

Technical Field

[0001] This application relates to the field of food industry equipment control technology, specifically to a control method and system for tart crust processing equipment based on big data modeling. Background Technology

[0002] Tart crust, especially puff pastry, is a popular base for Western-style pastries. Its final texture and appearance depend crucially on the delicate, alternating layers of fat and gluten formed during processing (shortcrust pastry). In modern industrialized food production, tart crust processing typically involves multiple continuous steps, including dough preparation, proofing, pressing, and freezing. Achieving high quality and consistency is key to enhancing market competitiveness. However, due to the natural volatility of raw materials (flour, fat), variations in the production environment, and gradual changes in equipment conditions, consistently replicating tart crusts with ideal layering, uniform thickness, and perfect demolding in mass production has always been a technical challenge for the industry.

[0003] Currently, automated tart crust production lines commonly employ Programmable Logic Controllers (PLCs) for sequential and logical control. Two main control strategies exist: one is program control based on fixed timing and setpoints, where process parameters for each step (such as kneading time, proofing temperature and humidity, and stamping pressure and time) are pre-set and fixed; the other relies on operator experience for intermittent manual intervention and parameter fine-tuning, such as manually adjusting the stamping press pressure or speed based on the dough's feel or sampled finished product appearance. Furthermore, some improvement schemes attempt to introduce simple feedback control within a single step, such as adjusting the dough cutting weight through weighing feedback or stabilizing the proofing chamber's set temperature through temperature sensors.

[0004] Although data acquisition and process control have been gradually introduced into industrial production, for the specific scenario of tart crust processing, existing systems, while typically collecting routine parameters (such as time, temperature, and pressure setpoints), fail to extract and utilize the deeper state characteristics of the dough reflected in this data. Due to the lack of this crucial state information, the system cannot quantitatively assess the true "readiness state" of the dough before stamping, leading to a lack of precise input for subsequent control. This can easily result in lower or unstable product quality after stamping, meaning proactive intervention is impossible, and only rejection or rework can be performed after defects occur, causing losses. Summary of the Invention

[0005] This application provides a control method and system for tart crust processing equipment based on big data modeling, which can solve the technical problems existing in the above-mentioned related technologies.

[0006] In a first aspect, embodiments of this application provide a control method for tart crust processing equipment based on big data modeling, employing the following technical solution: A control method for tart crust processing equipment based on big data modeling, the method comprising the following steps: Obtain the batch identifier of the dough, and based on the batch identifier, obtain the dough kneading process data and proofing process data generated in the kneading and proofing processes; wherein, the kneading process data includes at least the torque characteristic parameters obtained by performing time-frequency domain analysis on the torque signal of the mixing motor; the proofing process data includes temperature and humidity data and proofing time at at least two different spatial locations in the proofing environment; Obtain the pre-processing status data of the stamping equipment for the target dough stamping process to be performed, and construct an input feature vector by combining the dough kneading process data and the proofing process data; The input feature vector is input into a pre-trained stamping quality prediction model to obtain the quality prediction value of the finished tart crust after stamping according to the preset stamping process parameters; the quality prediction value includes at least the tart crust score characterizing the integrity of the tart crust layer structure and the probability of sticking to the mold when the tart crust is demolded. Compare the predicted quality value with the preset quality target, and determine whether the preset stamping process parameters need to be adjusted based on the comparison results. If necessary, the preset stamping process parameters are adjusted according to the pre-established parameter optimization model to improve the predicted quality value, thereby obtaining the stamping process parameters to be executed. The stamping equipment is controlled to perform a stamping operation on the target dough according to the stamping process parameters.

[0007] In conjunction with the first aspect, in one embodiment, the torque characteristic parameter includes a torque fluctuation coefficient, which is the ratio of the standard deviation to the average value of the torque signal of the stirring motor collected at a set sampling frequency over a complete mixing cycle.

[0008] In conjunction with the first aspect, in one embodiment, the torque characteristic parameter includes the proportion of low-frequency signal energy, which is the proportion of signal energy extracted from the 0.1Hz to 5Hz low-frequency band after performing a fast Fourier transform on the torque signal to the total signal energy.

[0009] In conjunction with the first aspect, in one embodiment, the proofing process data includes the vertical temperature difference between the top and bottom of the proofing box along its height direction.

[0010] In conjunction with the first aspect, in one embodiment, the proofing process data includes the vertical humidity difference between the top and bottom of the proofing box along its height direction.

[0011] In conjunction with the first aspect, in one implementation, the step of adjusting the preset stamping process parameters based on a pre-established parameter optimization model, guided by improving the predicted quality value, yields the executed stamping process parameters. The adjustment process includes adjusting the rate of change of the stamping pressure according to the tart crust score to form a protective pressure curve, and / or adjusting the execution parameters of the demolding aid according to the probability of sticking to the mold.

[0012] In conjunction with the first aspect, in one embodiment, adjusting the rate of change of the stamping pressure according to the tart crust score to form a protective pressure curve includes the following steps: When the tart crust score is lower than a first preset threshold, the first pressurization rate of the first stage is reduced; when the tart crust score is higher than or equal to the first preset threshold, the pressurization rate of the first stage is increased or reduced. When the tart crust score is lower than the second preset threshold, the second pressurization rate of the second stage after the first stage is reduced, and a pressure holding platform is inserted during the pressure rise; when the tart crust score is higher than or equal to the second preset threshold, the second pressurization rate of the second stage is adopted or increased, and the pressure holding platform is not inserted.

[0013] In conjunction with the first aspect, in one embodiment, the step of adjusting the execution parameters of the demolding aid device according to the probability of mold sticking risk, Based on the risk level to which the sticking risk probability belongs, the corresponding set of execution parameters for the demolding assistance device is selected from the predefined collaborative strategy library; the set of execution parameters for the demolding assistance device includes different ejection speeds and action sequences of the ejection mechanism, different vibration parameters and triggering times of the high-frequency vibrator, and whether to use spray-assisted demolding and the timing of spray triggering.

[0014] In conjunction with the first aspect, in one embodiment, after controlling the stamping equipment to perform a stamping operation on the target dough according to the stamping process parameters, the method further includes the following steps: Obtain the actual product quality data of the target dough after the stamping process is completed, and associate this data with other process data and stamping process parameters involved in the target dough, and add it as a new historical data sample to the historical dataset of dough under the same batch identifier; The stamping quality prediction model and the parameter optimization model are updated using the historical dataset.

[0015] Secondly, embodiments of this application provide a control system for tart crust processing equipment based on big data modeling, employing the following technical solution: A control system for tart crust processing equipment based on big data modeling, used to implement the control method for tart crust processing equipment based on big data modeling as described above, comprises: The data retrieval module is configured to obtain the batch identifier of the dough, and based on the batch identifier, obtain the dough kneading process data and proofing process data generated in the kneading and proofing processes; wherein, the kneading process data includes at least the torque characteristic parameters obtained by performing time-frequency domain analysis on the torque signal of the mixing motor; the proofing process data includes temperature and humidity data at at least two different spatial locations in the proofing environment and the proofing time; The quality prediction module is configured to acquire the pre-processing state data of the stamping equipment for the target dough stamping process, combine the dough kneading process data and the proofing process data to construct an input feature vector; input the input feature vector into a pre-trained stamping quality prediction model to obtain the quality prediction value of the tart crust finished product after stamping according to preset stamping process parameters; the quality prediction value includes at least the tart crust score characterizing the integrity of the tart crust layer structure and the probability of sticking to the mold when the tart crust is demolded; The comparison and adjustment module is configured to compare the predicted quality value with the preset quality target, and determine whether the preset stamping process parameters need to be adjusted based on the comparison result; if so, adjust the preset stamping process parameters according to the pre-established parameter optimization model to improve the predicted quality value, thereby obtaining the execution stamping process parameters; and control the stamping equipment to perform stamping operation on the target dough according to the execution stamping process parameters.

[0016] The beneficial effects of the technical solutions provided in this application include: This application provides a control method and system for tart crust processing equipment based on big data modeling. By utilizing collectable data or state information of the dough during the kneading and proofing stages, it further analyzes and obtains gradient information on the uniformity of the dough's structural distribution and the dough's relaxation state. By combining this deep state information, which is difficult to quantify or is often overlooked in traditional methods, with the pre-processing state data of the stamping equipment, a comprehensive input feature vector is constructed. This enables the stamping quality prediction model to more accurately assess the "preparation state" of the dough before stamping, achieving proactive prediction of the tart crust's finished product quality rather than post-processing inspection, and allowing for timely intervention and adjustment in the future. On the other hand, after making a forward-looking prediction of the quality of the finished tart crust, this application uses a parameter optimization model to dynamically adjust the stamping process parameters based on the prediction results, forming an adaptive control strategy. Through this data-driven optimization, the execution of the stamping process parameters can better adapt to the characteristics of the current dough, thereby effectively improving the integrity of the tart crust's layer structure, reducing the risk of sticking to the mold, and ultimately improving the stability and consistency of product quality. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an embodiment of the control method for tart crust processing equipment based on big data modeling in this application. Figure 2 This is a schematic diagram of the functional modules of an embodiment of the tart crust processing equipment control system based on big data modeling in this application. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0019] In the traditional tart crust processing, although some process parameters of the dough are obtained before stamping, these parameters cannot effectively reflect the deep state information of the dough, such as the uniformity of the distribution of the internal structure and the relaxation state of the dough. This deep state information cannot be quantitatively evaluated, which further leads to the lack of precise input parameters for controlling the stamping process, resulting in fluctuations or substandard quality of the tart crust products obtained from the batch of dough.

[0020] To address these issues, this application provides a control method and system for tart crust processing equipment based on big data modeling.

[0021] Firstly, referring to Figure 1 This application provides a control method for tart crust processing equipment based on big data modeling, the method including the following steps: S100. Obtain the batch identifier of the dough, and obtain the dough kneading process data and proofing process data generated in the kneading process and proofing process based on the batch identifier; The data for the dough mixing process includes at least torque characteristic parameters obtained by performing time-frequency domain analysis on the torque signal of the mixing motor. Specifically, the torque characteristic parameters are values ​​extracted from the time-frequency domain analysis of the torque signal of the mixing motor. For example, they may include statistical characteristics of the torque signal or the energy percentage within a specific frequency range. These parameters can be used to quantify the uniformity of structural distribution in the dough, such as the distribution of fats and gluten.

[0022] The proofing process data includes environmental conditions and time information for the dough during the proofing stage. Specifically, it includes temperature and humidity data at at least two different spatial locations in the proofing environment, as well as proofing time. The proofing process data is used to further characterize the gradient information of the dough's relaxation state. The gradient information reflects differences that may be caused by uneven temperature and humidity, and affect the dough's deformation behavior during stamping and the final product quality.

[0023] S200: Obtain the pre-processing status data of the stamping equipment for the target dough stamping process to be executed, and construct the input feature vector by combining the dough kneading process data and the proofing process data; The pre-processing status data includes the operating status information of the stamping equipment before it performs the stamping operation. For example, this may include die temperature, equipment running time, ambient temperature, etc., all of which affect the stability of the stamping process and product quality. The feature vector, formed by combining pre-processing state data, surface processing data, and proofing process data, is a multi-dimensional data representation resulting from integrating the surface processing data, proofing process data, and pre-processing state data of the stamping equipment. This vector serves as the input to the subsequent stamping quality prediction model, containing more comprehensive information affecting the tart crust forming quality.

[0024] S300. Input the input feature vector into the pre-trained stamping quality prediction model to obtain the quality prediction value of the finished tart crust after stamping according to the preset stamping process parameters; the quality prediction value includes at least the tart crust score characterizing the integrity of the tart crust layer structure and the probability of sticking to the mold when the tart crust is demolded. S400: Compare the predicted quality value with the preset quality target, and determine whether the preset stamping process parameters need to be adjusted based on the comparison results. S500. If necessary, adjust the preset stamping process parameters according to the pre-established parameter optimization model to improve the quality prediction value, and obtain the stamping process parameters to be executed. S600: Control the stamping equipment to perform stamping operation on the target dough according to the stamping process parameters.

[0025] Specifically, the method provided in this application first obtains the batch identifier of the dough, and then obtains the dough kneading and proofing process data generated during the kneading and proofing processes based on the batch identifier. The batch identifier can be obtained in various ways, such as manually inputting the dough batch number into the control interface, or automatically identifying it by scanning the barcode or QR code on the dough container. The kneading and proofing process data can be queried from a production database or read in real time from a data acquisition system connected to the kneading and proofing equipment. The kneading process data can be obtained by collecting the real-time torque signal of the mixing motor during the kneading process and performing simple statistical analysis on the signal, such as calculating its average value, variance, or peak value, as a torque characteristic parameter. The proofing process data can be obtained by placing a temperature and humidity sensor at the top and bottom of the proofing box to record the temperature and humidity changes throughout the proofing cycle, and combining this with the total proofing time to characterize the relaxation state of the dough.

[0026] Furthermore, the method provided in this application will acquire the pre-processing state data of the stamping equipment for the target dough stamping process, which can be acquired through the equipment's built-in sensors or control system; subsequently, it will combine the dough kneading process data and the proofing process data to construct an input feature vector that can comprehensively reflect the influencing conditions of dough stamping.

[0027] Subsequently, the aforementioned input feature vector is fed into a pre-trained stamping quality prediction model to obtain the predicted quality value of the finished tart crust after stamping according to preset stamping process parameters. This stamping quality prediction model can be a machine learning model trained on historical production data; for example, it can be a multilayer perceptron or support vector machine model. It predicts various quality indicators of the tart crust by learning the relationship between input features and the final tart crust quality. The predicted quality value includes at least a tart crust score characterizing the structural integrity of the crust layers and the probability of the crust sticking to the mold during demolding. The tart crust score can be a value between 0 and 100, used to quantify structural integrity. The probability of sticking to the mold can be a value between 0 and 1, representing the likelihood of the tart crust sticking together during demolding.

[0028] Next, the predicted quality value is compared with the preset quality target. Based on the comparison results, it is determined whether the preset stamping process parameters need to be adjusted. The preset quality target can be set by the production manager according to product standards. For example, it may require the tart crust score to be no less than 85 points and the probability of sticking to the mold to be no more than 0.1. The comparison process can be a simple logical judgment. For example, if the predicted tart crust score is lower than the preset target, or the probability of sticking to the mold is higher than the preset target, then it is determined that parameter adjustments are needed.

[0029] If adjustments are deemed necessary, the preset stamping process parameters are adjusted based on a pre-established parameter optimization model, guided by the goal of improving the predicted quality, to obtain the final stamping process parameters. The parameter optimization model can be a mapping selection model based on predetermined rules, which selects appropriate adjustment schemes from a predefined strategy library according to the specific circumstances of the predicted quality. For example, if the tart crust score is low, the model might suggest increasing the stamping pressure or extending the holding time; if the risk of sticking to the mold is high, the model might suggest adjusting the demolding speed or increasing the strength of the demolding aid.

[0030] Finally, the stamping equipment is controlled to perform a stamping operation on the target dough according to the stamping process parameters. The stamping process parameters are sent to the controller of the stamping equipment (e.g., a programmable logic controller, PLC), which precisely controls various actions of the stamping equipment based on these parameters, including stamping pressure, stamping speed, holding time, demolding sequence, etc., to ensure that the dough can be stamped and shaped in the best condition.

[0031] The following example will provide a more detailed explanation of the above technical solution: On the tart crust production line, a batch of dough marked "M20231101-001" has completed the kneading and proofing processes and is about to enter the stamping stage. Traditional control methods usually operate according to preset fixed stamping parameters without considering the actual state of the dough in that batch, which may lead to inconsistencies in the quality of the final tart crust product.

[0032] To address this, the method proposed in this application first obtains the batch identifier "M20231101-001" of the dough batch. Based on this identifier, the system retrieves data generated by the dough during the kneading and proofing processes from a historical database. Specifically, during the kneading process, the torque signal of the mixing motor is collected throughout the entire kneading cycle and subjected to time-frequency domain analysis to extract torque characteristic parameters such as the average value and standard deviation of the torque signal. These parameters are used to characterize the uniformity of oil distribution in the batch of dough. For example, large torque fluctuations may indicate uneven oil distribution. During the proofing process, temperature and humidity sensors installed at the top and bottom of the proofing box record temperature and humidity data throughout the proofing process, which, combined with the proofing time, are used to characterize the gradient information of the dough's relaxation state. For example, a large temperature or humidity difference between the top and bottom may indicate unevenness in the dough's relaxation state.

[0033] Simultaneously, the system acquires pre-processing status data of the stamping equipment, such as the current temperature of the stamping die being 35 degrees Celsius, the stamping equipment having run continuously for 8 hours, and the ambient temperature of the workshop being 22 degrees Celsius. Subsequently, the acquired data from the dough kneading process (such as torque characteristic parameters), the proofing process (such as temperature and humidity data and proofing time), and the pre-processing status data of the stamping equipment (such as die temperature and equipment running time) are integrated to construct a comprehensive input feature vector.

[0034] The input feature vector is fed into a pre-trained stamping quality prediction model. This model learns the complex relationship between dough condition, equipment condition, and final tart crust quality based on extensive historical production data. Upon receiving the input vector, the model immediately outputs a predicted quality value for the tart crust after the batch of dough has been stamped according to the current preset stamping process parameters. For example, the model predicts a crust score of 78 points and a sticking probability of 0.25 for this batch of tart crusts.

[0035] Next, the system compares these predicted quality values ​​with the preset quality targets. The preset quality targets are: a tart crust score of no less than 85 points and a sticking probability of no more than 0.15. Since the predicted tart crust score (78 points) is lower than the target (85 points), and the sticking probability (0.25) is higher than the target (0.15), the system determines that the current preset stamping process parameters need to be adjusted.

[0036] If adjustments are needed, the system adjusts the preset stamping process parameters based on a pre-established parameter optimization model, aiming to improve the tart crust score and reduce the probability of mold sticking. For example, the parameter optimization model might suggest increasing the peak value of the stamping pressure curve by 5% and extending the holding time by 1 second if the tart crust score is low; simultaneously, it might suggest increasing the ejection speed of the demolding aid by 10% if the probability of mold sticking is high. This results in a new set of stamping process parameters.

[0037] Finally, the system sends this set of optimized stamping process parameters to the controller of the stamping equipment. The stamping equipment then performs the stamping operation on the dough batch identified as "M20231101-001" based on these new parameters. In this way, the stamping process can be dynamically adjusted according to the actual state of the dough and the real-time status of the equipment, thereby improving the quality and consistency of the finished tart crust.

[0038] Based on the above scheme, it can be seen that, compared with the traditional tart crust processing methods that rely on fixed process parameters or operator experience for adjustment, this application utilizes collectable data or state information of the dough during the kneading and proofing stages to achieve further analysis and obtain deeper state information, thereby at least analyzing the uniformity of the dough's structural distribution and the gradient information of the dough's relaxation state. These are key factors that are difficult to quantify or are ignored in traditional methods. By combining this deeper state information with the pre-processing state data of the stamping equipment, a comprehensive input feature vector is constructed, enabling the stamping quality prediction model to more accurately assess the dough's "preparation state" before stamping. Thus, this method achieves forward-looking prediction of the tart crust quality, rather than post-processing detection, and allows for timely intervention and adjustment. Furthermore, after completing the forward-looking prediction of the tart crust quality, this application chooses to use a parameter optimization model to dynamically adjust the stamping process parameters based on the prediction results, thereby forming an adaptive control strategy. Through this data-driven optimization, the execution of the stamping process parameters can better adapt to the characteristics of the current dough, thereby effectively improving the integrity of the tart crust's layer structure, reducing the risk of sticking to the mold, and ultimately improving the stability and consistency of product quality.

[0039] Furthermore, in some embodiments, the torque characteristic parameter will specifically include a torque fluctuation coefficient, which is the ratio of the standard deviation to the average value of the torque signal of the stirring motor collected at a set sampling frequency over a complete mixing cycle.

[0040] A complete kneading cycle refers to the entire time from the start to the end of the kneading process. During this period, the mixing motor thoroughly mixes and kneads the dough, shaping it into a dough with specific gluten strength and elasticity. The complete kneading cycle is usually preset by the control system of the kneading equipment; for example, it can be set to 10 minutes, 15 minutes, or 20 minutes, depending on the dough recipe and the desired dough state.

[0041] The torque fluctuation coefficient is a statistical index used to quantify the dispersion of the torque signal of a stirring motor relative to its average level during the dough kneading process. In this embodiment, the torque fluctuation coefficient is calculated by collecting the torque signal generated by the stirring motor equipped with a torque sensor within a complete dough kneading cycle at a set sampling frequency, and then calculating the ratio of the standard deviation to the average value of these signals. Alternatively, the torque fluctuation coefficient can also be calculated using other statistical methods, such as the ratio of the root mean square error (RMSE) to the average value, or by calculating the ratio of the peak-to-valley difference of the torque signal to the average value to characterize its fluctuation degree. The set sampling frequency refers to the number of sample points collected per unit time when collecting the stirring motor torque signal. The sampling frequency can be determined based on the stirring motor speed, the dynamic characteristics of the dough kneading process, and the required data accuracy. In this embodiment, 10-50Hz is preferred for easier implementation by industrial PLCs and data acquisition cards. For PLCs and acquisition cards capable of collecting higher frequency signals, it can also be set to 100Hz or 200Hz in other embodiments to ensure that subtle changes in the torque signal can be captured.

[0042] This application's embodiments introduce a torque fluctuation coefficient as a torque characteristic parameter in the dough kneading process data, thereby effectively quantifying the uniformity of the target dough's structural distribution. Specifically, in the dough kneading process of tart crust processing, the mixing motor generates different torque responses to dough in different states. The more uniform the distribution of fat in the dough and the more complete the gluten network formation, the smaller the fluctuation of the mixing torque. Conversely, when the blades cut and squeeze insufficiently dispersed fat lumps or excessively large gluten aggregates, a momentary, "collision-like" sudden increase in resistance occurs, manifesting as high-frequency peaks or burrs on the torque curve. Therefore, by collecting the torque signals generated by the mixing motor throughout the entire complete kneading cycle at a set sampling frequency, and then calculating the ratio of the standard deviation to the average value of these torque signals—that is, the torque fluctuation coefficient—this ratio can dimensionlessly and objectively quantify the relative degree of change in internal resistance of the dough during the kneading process, thus accurately characterizing the uniformity of fat and gluten distribution in the dough. By incorporating this quantified torque fluctuation coefficient into the dough kneading process data, and combining it with other proofing process data and the pre-processing status data of the stamping equipment, a more representative input feature vector can be constructed, which can significantly improve the accuracy of the pre-trained stamping quality prediction model in predicting the quality of the tart crust.

[0043] Furthermore, in some embodiments, this application further proposes that the torque characteristic parameter includes the proportion of low-frequency signal energy, wherein the proportion of low-frequency signal energy is the ratio of the signal energy extracted from the 0.1Hz to 5Hz low-frequency band after performing a fast Fourier transform on the torque signal to the total signal energy.

[0044] The low-frequency signal energy ratio is a quantitative indicator obtained through frequency domain analysis of the torque signal from the mixing motor. It aims to reflect the macroscopic uniformity and stability of the dough's internal structure during the kneading process. Specifically, this indicator can reflect the slow dynamic processes related to changes in the dough's internal structure that may be overlooked by traditional time-domain analysis (such as torque fluctuation coefficient). The specific calculation method involves performing a Fast Fourier Transform (FFT) on the acquired torque signal. The FFT decomposes complex time-domain signals into a superposition of different frequency components, revealing hidden periodicity, harmonic components, and energy distribution at different frequencies. This transformation can be performed by a dedicated digital signal processor (DSP) chip, an embedded controller (such as an ARM processor) equipped with corresponding mathematical library functions, or processed by an industrial computer connected to the device using software algorithms.

[0045] The extraction of signal energy in the low-frequency band from 0.1Hz to 5Hz refers to selecting all frequency components between 0.1Hz and 5Hz from the obtained spectrum data after completing the Fast Fourier Transform, and calculating the total energy contained in these components. In dough kneading, the blade rotation cycle is typically between 0.5 seconds and several seconds, corresponding to 1-2Hz. The entire macroscopic cycle of the dough being lifted, folded, and rolled back in the mixing bowl typically takes several seconds to tens of seconds, corresponding to 0.1-1Hz. Furthermore, the hydration and polymerization of gluten proteins in dough is a relatively slow kinetic process. Its structure undergoes stretching, folding, and reorganization under shear force, requiring continuous mechanical energy input and exhibiting a stable power dominated by viscous dissipation, with spectral characteristics concentrated in the low-frequency range below 5Hz. In contrast, fats, especially solid fats, are stretched into continuous films during repeated folding and rolling. This plastic deformation process consumes energy continuously and steadily, also exhibiting a power spectrum in the low-frequency range below 5Hz. Therefore, this specific low-frequency range is preferred in this embodiment because it is more effective in corresponding to the macroscopic movement rhythm of the mixing paddle, the overall mixing and tumbling cycle of the dough, and the physical process of gluten and fat slowly reorganizing or aggregating in the dough during the actual dough kneading process.

[0046] The aforementioned proportion of the total signal energy refers to the sum of the extracted signal energy in the 0.1Hz to 5Hz low-frequency band and the total signal energy across the entire effective frequency range. By calculating this proportion, the differences in absolute energy values ​​caused by factors such as the power of the mixing motor, the total amount of dough, or the sensitivity of the sensor can be eliminated, thus obtaining a relative and more robust index for accurately characterizing the uniformity of fat distribution in the dough and the stability of the kneading process.

[0047] Therefore, the solution in this application uses the proportion of low-frequency signal energy as a torque characteristic parameter in the dough kneading process data, which, together with the aforementioned torque fluctuation coefficient, constitutes a comprehensive characterization of the time-frequency domain analysis of the stirring motor torque signal. This combination allows the subsequent input feature vector to more comprehensively and deeply reflect the internal state of the dough during the kneading stage, especially the macroscopic distribution uniformity of gluten and fat. When the input feature vector is input into a pre-trained stamping quality prediction model, the model can obtain a more accurate quality prediction value for the finished tart crust based on richer and more accurate dough state information, including tart crust scores and the probability of sticking to the mold.

[0048] Furthermore, in some embodiments, the proofing process data needs to include the vertical temperature difference between the top and bottom of the proofing box along its height direction.

[0049] The vertical temperature difference refers to the temperature difference at different locations along the height of the proofing chamber. Several methods can be used to accurately obtain this crucial parameter. For example, a temperature sensor, such as a PT100 resistance temperature detector (RTD) or a thermocouple, can be installed at both the top and bottom of the proofing chamber. The vertical temperature difference is obtained by collecting the temperature values ​​at these two locations in real time and calculating the difference. Alternatively, a movable temperature probe can be deployed to scan the vertical direction of the proofing chamber, thereby obtaining the temperature distribution at different heights and calculating the temperature difference between the top and bottom.

[0050] The solution in this application incorporates the vertical temperature difference between the top and bottom of the proofing box along its height into the proofing process data, enabling the system to more precisely capture the differences in dough state caused by temperature stratification within the proofing box. Specifically, when there is a significant vertical temperature difference within the proofing box, such as a higher temperature at the top than at the bottom, the yeast activity in the top dough is higher, the gas production rate is faster, and the gluten network is more relaxed; while the bottom dough may ferment more slowly and have a relatively dense gluten structure. If this unevenness in the relaxation state of the upper and lower doughs is not quantified, it will cause the stamping quality prediction model to deviate when assessing the state of the dough before entering the stamping process. By measuring and inputting the vertical temperature difference, the stamping quality prediction model can identify and learn the influence of this temperature gradient on the physical properties of the dough (such as hardness, elasticity, and internal stress), thereby more accurately representing the overall relaxation state and gradient distribution of the dough when constructing the input feature vector. This allows the model to more accurately predict the quality prediction value of the finished tart crust, including the crust score and the probability of sticking to the mold, providing a more reliable decision-making basis for subsequent adjustments to the stamping process parameters.

[0051] Furthermore, in some embodiments, the proofing process data also includes the vertical humidity difference between the top and bottom of the proofing box along its height direction.

[0052] The vertical humidity difference between the top and bottom of the proofing box refers to the difference in humidity value between the top and bottom areas along the vertical direction inside the proofing box. This parameter directly reflects the uniformity or gradient of humidity distribution within the proofing box and has a significant impact on assessing the surface moisture evaporation rate of dough at different heights and the risk of crust formation.

[0053] The vertical humidity difference can be calculated by installing at least one high-precision humidity sensor at the top and bottom of the proofing box to collect humidity data in real time, and then using a controller or data processing unit to calculate the difference.

[0054] This application aims to better capture the relaxation gradient information of dough during the proofing process by introducing the vertical humidity difference between the top and bottom of the proofing box along the height direction into the proofing process data. Specifically, in tart crust processing, the purpose of dough proofing is to relax the internal structure of the dough to achieve suitable extensibility and elasticity for stamping. However, when the humidity inside the proofing box is uneven, especially in the vertical direction, there may be a significant difference in humidity between the top and bottom. This vertical humidity difference leads to different surface water activities of the dough at different heights, which in turn affects the degree of drying of the dough surface and the formation of a hard crust. When there is a large vertical humidity difference in the proofing box, for example, the humidity at the bottom is higher than that at the top, the surface of the dough in the top area may lose water faster, making it easier to form a dense and poorly extensible hard crust. This hard crust not only restricts the normal flow of the dough in the stamping mold, resulting in incomplete filling of the tart crust edges, but may also crack during stamping due to insufficient extensibility, seriously affecting the integrity of the tart crust's layer structure (i.e., tart crust score). Therefore, this embodiment incorporates this vertical humidity difference as part of the proofing process data, combining it with the dough mixing process data and the pre-processing state data of the stamping equipment to construct an input feature vector. This input feature vector is then fed into a pre-trained stamping quality prediction model. Since the vertical humidity difference can more accurately reflect the true state of the dough during the proofing stage, especially the potential risk of surface crust formation, the stamping quality prediction model can more accurately predict the quality of the finished tart crust based on this.

[0055] Furthermore, in some embodiments, during step S500, adjusting the preset stamping process parameters based on a pre-established parameter optimization model to improve the quality prediction value, and obtaining the stamping process parameters to be executed, The adjustment process includes adjusting the rate of change of the stamping pressure according to the tart crust score to form a protective pressure curve, and / or adjusting the execution parameters of the demolding aid according to the probability of sticking to the mold.

[0056] The rate of change of stamping pressure refers to how quickly the pressure value changes over time when the stamping equipment applies pressure to the target dough. This rate directly affects the stress distribution and deformation process of the dough's internal layered structure. The protective pressure curve is a non-linear pressure application mode, whose core objective is to avoid damaging the dough's layered structure due to excessively rapid or excessive pressure application during the stamping process. This curve can be represented as a piecewise linear, exponential, or S-shaped pressure rise trajectory. Its characteristic is that a relatively gentle pressure is applied in the initial stage when the dough structure is relatively fragile, while the pressure application rate can be appropriately increased after the dough structure gradually stabilizes or reaches a certain density. By adjusting the rate of change of stamping pressure, the slope, segment points, or maximum pressure value of the pressure rise can be dynamically adjusted according to the tart crust score to adapt to the needs of different dough states. The demolding aid device refers to a device or mechanism used to help the tart crust separate smoothly from the mold. Common demolding aid devices include, but are not limited to, ejection mechanisms, high-frequency vibrators, and spraying devices (such as oil spraying or air spraying devices). The execution parameters of the demolding aid device refer to the specific working settings of the demolding aid device when performing the demolding operation. For example, for an ejection mechanism, the execution parameters may include ejection speed, ejection stroke, ejection timing, etc.; for a high-frequency vibrator, they may include vibration frequency, vibration amplitude, vibration duration, triggering timing, etc.; for a spraying device, they may include spray volume, spray pressure, spraying timing, etc. By adjusting the execution parameters of the demolding aid device, the working mode and specific parameters of the demolding aid device can be dynamically selected or modified according to the predicted probability of sticking to the mold, in order to cope with different sticking risks.

[0057] Specifically, after receiving the tart crust score and sticking risk probability from the stamping quality prediction model, the parameter optimization model intelligently adjusts the rate of pressure change applied by the stamping equipment based on the tart crust score. When the tart crust score is low, the model generates a smoother protective pressure curve to avoid damaging the layered structure of the dough during stamping; when the tart crust score is high, a faster pressurization rate can be used to improve production efficiency. Simultaneously, the parameter optimization model dynamically adjusts the operating mode and parameters of the demolding aid device based on the predicted sticking risk probability. When the sticking risk probability is high, the model selects a stronger or more coordinated demolding aid strategy, such as increasing the ejection speed, increasing vibration intensity, or advancing the spraying timing, to ensure smooth demolding of the tart crust; when the sticking risk probability is low, gentler demolding parameters can be used to reduce energy consumption and equipment wear.

[0058] In this embodiment, adjusting the rate of change of the stamping pressure based on the tart crust score specifically includes the following steps: When the tart crust score is lower than the first preset threshold, the first pressurization rate of the first stage is reduced; when the tart crust score is higher than or equal to the first preset threshold, the pressurization rate of the first stage is increased or reduced. When the tart crust score is lower than the second preset threshold, the second pressurization rate of the second stage after the first stage is reduced, and a pressure holding platform is inserted during the pressure rise; when the tart crust score is higher than or equal to the second preset threshold, the second pressurization rate of the second stage is adopted or increased, and a pressure holding platform is not inserted.

[0059] The first preset threshold is a critical value used to determine whether the tart crust score meets a certain standard. This threshold can be set based on historical production data, product quality standards, or expert experience. The second preset threshold is another critical value used to judge the tart crust score, usually lower than or equal to the first preset threshold, used to more finely distinguish the tart crust structure. Its specific value can also be manually adjusted according to actual production needs and product characteristics.

[0060] The first stage, the initial pressurization rate, refers to the speed at which the pressurizing equipment applies pressure to the target dough during the initial pressurization process. This rate directly affects the deformation and layering of the dough during initial pressure. The second stage, the second pressurization rate, refers to the speed at which the pressurizing equipment applies pressure to the target dough in the subsequent stages of the pressurization process. This pressurization rate has a significant impact on the final thickness, density, and layering stability of the tart crust. The holding pressure plateau refers to the stage during the pressurization process where, after the pressure reaches a certain set value, that pressure is maintained for a period of time without further increase. The purpose of this plateau is to provide the internal structure of the dough with sufficient relaxation and stabilization time, which helps the fat and gluten layers to better combine or separate, thereby optimizing the layering structure of the tart crust. The holding pressure plateau can be a fixed pressure value and duration, or it can be dynamically adjusted according to the characteristics of the dough.

[0061] This application embodiment achieves adaptive adjustment of the stamping pressure change rate by performing phased and threshold-based judgments on the tart crust score, and introduces a pressure-holding platform mechanism to form a protective pressure curve. Specifically, in the first stage of the stamping process, the system first obtains the tart crust score output by the stamping quality prediction model. When the tart crust score is lower than a first preset threshold, it indicates that the layer structure integrity of the dough may be poor. At this time, the system will reduce the first pressurization rate of the first stage to reduce the impact on the dough and avoid further damage to the fragile structure of the dough due to excessively rapid pressurization. Conversely, if the tart crust score is higher than or equal to the first preset threshold, it indicates that the dough structure is relatively ideal, and the system can adopt or increase the first stage pressurization rate to improve production efficiency. Based on this, in the second stage of the stamping process, the system will again compare the tart crust score with the second preset threshold. When the tart crust score is lower than the second preset threshold, this usually means that the layer structure of the dough still has certain risks or needs further optimization in the second stage of stamping. At this time, the system will reduce the second pressurization rate of the second stage and strategically insert a pressure-holding platform during the pressure rise. By reducing the pressurization rate and increasing the holding time, the dough's internal structure can be given ample opportunity to relax and adjust, promoting the stable formation of the fat and gluten layers, thereby effectively protecting and optimizing the integrity of the tart crust's layer structure. Conversely, if the tart crust score is higher than or equal to the second preset threshold, it indicates that the dough structure is stable and good. The system can then adopt or increase the second pressurization rate in the second stage without needing to insert a holding platform, ensuring pressing efficiency and the quality of the final product.

[0062] Furthermore, in some embodiments, adjusting the execution parameters of the demolding aid device according to the probability of mold sticking risk specifically includes: Based on the risk level to which the sticking risk probability belongs, the corresponding set of execution parameters for the demolding assistance device is selected from the predefined collaborative strategy library; the set of execution parameters for the demolding assistance device includes different ejection speeds and action sequences of the ejection mechanism, different vibration parameters and triggering times of the high-frequency vibrator, and whether to use spray-assisted demolding and the timing of spray triggering.

[0063] The risk level to which the sticking risk probability belongs refers to dividing the sticking risk probability output by the stamping quality prediction model into several discrete risk levels, such as low risk, medium risk, and high risk. This division can be based on a preset probability threshold. The predefined collaborative strategy library is a database or lookup table that stores pre-optimized combinations of demolding auxiliary device execution parameters for different sticking risk levels and / or specific dough characteristics. The parameter combinations in this strategy library can be optimized and verified through experiments, simulations, or based on expert experience, aiming to ensure the best demolding effect under different risk scenarios. For example, it can be a mapping table that associates each risk level with a specific set of demolding parameters; or it can be a rule-based system that dynamically generates parameter sets based on risk levels and other contextual information. The demolding auxiliary device execution parameter set includes different ejection speeds and action sequences of the ejection mechanism, different vibration parameters and triggering times of the high-frequency vibrator, and whether to use spray-assisted demolding and the spray triggering time. Specifically, the ejection speed of the ejection mechanism can be set to a constant speed, segmented speed variation, or gradual speed variation. Its timing can be immediate activation after the mold is fully opened, activation after a specific delay, or synchronization with the mold's opening and closing action. The vibration parameters of the high-frequency vibrator can include vibration frequency, amplitude, and duration, and its triggering timing can be before, during, or after ejection. Spray-assisted demolding can be selected for activation as needed. If activated, the spray type (e.g., release agent, water mist), spray volume, and spray triggering timing (e.g., before dough placement, after stamping, or during demolding) can be set.

[0064] The embodiment of this application further refines the sticking risk probability obtained from the stamping quality prediction model into specific risk levels. Based on this, it intelligently selects a set of execution parameters for a multi-functional demolding auxiliary device from a pre-established collaborative strategy library, thereby achieving refined and collaborative control of the tart crust demolding process. When the system predicts that the target dough has a sticking risk under the current preset stamping process parameters, it no longer simply adjusts a single demolding parameter, but first assesses the severity of the risk and classifies it into the corresponding risk level. Subsequently, the system retrieves the most suitable set of execution parameters for the demolding auxiliary device from the predefined collaborative strategy library based on the risk level. In this way, various auxiliary devices can better cooperate during the demolding process, forming a three-dimensional, multi-layered demolding process, significantly improving the success rate and efficiency of demolding.

[0065] Furthermore, in some embodiments, after step S600, where the stamping equipment is controlled to perform a stamping operation on the target dough according to the stamping process parameters, the following steps are also included: S700. Obtain the actual product quality data of the target dough after stamping, associate this data with other process data and stamping process parameters involved in the target dough, and add it as a new historical data sample to the historical dataset of dough under the same batch identifier. S800 updates the stamping quality prediction model and parameter optimization model using historical datasets.

[0066] This solution aims to obtain the true quality indicators of the tart crust after the stamping operation. This can be achieved through manual visual inspection and scoring, or through image analysis of the finished tart crust using an automated visual inspection system. The purpose of this correlation operation is to integrate the actual product quality data after stamping with information from the entire production process that led to that quality result. Other process data typically refers to the data generated during the dough kneading and proofing processes of that batch of dough, reflecting the state of the dough before stamping. Stamping process parameters refer to the parameters actually used in this stamping operation, such as stamping pressure curve, stamping speed, holding time, and demolding auxiliary parameters. During correlation, data matching and merging are performed in the database using a unified batch identifier, timestamp, or production order number. Ultimately, the complete correlated data records are used as new training samples to expand the existing historical dataset. The historical dataset will then be used as the basis for training and optimizing the stamping quality prediction model and parameter optimization model. The same batch identifier ensures that newly added data samples maintain logical consistency with the historical production data of the same batch of dough, helping the model learn the quality fluctuation patterns within the batch. Update methods can include incremental learning, which involves training the existing model with only new data in small batches to adapt to the new data distribution; periodic full retraining, which involves retraining the model with the entire historical dataset after accumulating a certain amount of new data; or transfer learning, which involves transferring knowledge learned from old data to the new model and fine-tuning it with new data.

[0067] Based on this update mechanism, the stamping quality prediction model can learn the latest mapping relationship between production conditions and product quality, thereby improving its accuracy in predicting the quality of future tart crust products. Simultaneously, the parameter optimization model can adjust its optimization strategy based on the latest actual quality feedback, enabling it to more accurately guide the stamping equipment to achieve preset quality targets when recommending stamping process parameters.

[0068] Secondly, this application proposes a control system for tart crust processing equipment based on big data modeling, which includes a data retrieval module, a quality prediction module, and a comparison and adjustment module.

[0069] The data retrieval module is configured to obtain the batch identifier of the dough, and based on the batch identifier, obtain the dough kneading process data and proofing process data generated in the kneading and proofing processes; wherein, the kneading process data includes at least the torque characteristic parameters obtained by performing time-frequency domain analysis on the torque signal of the mixing motor, which are used to characterize the uniformity of the dough structure distribution; the proofing process data includes temperature and humidity data at at least two different spatial locations in the proofing environment and proofing time, which are used to characterize the gradient information of the dough relaxation state; The quality prediction module is configured to acquire the pre-processing status data of the stamping equipment for the target dough stamping process, and combine the dough kneading process data and the proofing process data to construct an input feature vector; the input feature vector is input into a pre-trained stamping quality prediction model to obtain the quality prediction value of the tart crust finished product after stamping according to the preset stamping process parameters; the quality prediction value includes at least the tart crust score characterizing the integrity of the tart crust layer structure and the probability of sticking to the mold when the tart crust is demolded; The comparison and adjustment module is configured to compare the predicted quality value with the preset quality target, and determine whether the preset stamping process parameters need to be adjusted based on the comparison result; if so, the preset stamping process parameters are adjusted according to the pre-established parameter optimization model to improve the predicted quality value, so as to obtain the execution stamping process parameters; and the stamping equipment is controlled to perform stamping operation on the target dough according to the execution stamping process parameters.

[0070] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0071] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0072] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0073] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0074] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0075] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0076] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A control method for tart crust processing equipment based on big data modeling, characterized in that, The method includes the following steps: Obtain the batch identifier of the dough, and based on the batch identifier, obtain the dough kneading process data and proofing process data generated in the kneading and proofing processes; wherein, the kneading process data includes at least the torque characteristic parameters obtained by performing time-frequency domain analysis on the torque signal of the mixing motor; the proofing process data includes temperature and humidity data and proofing time at at least two different spatial locations in the proofing environment; Obtain the pre-processing status data of the stamping equipment for the target dough stamping process to be performed, and construct an input feature vector by combining the dough kneading process data and the proofing process data; The input feature vector is input into a pre-trained stamping quality prediction model to obtain the quality prediction value of the finished tart crust after stamping according to the preset stamping process parameters; the quality prediction value includes at least the tart crust score characterizing the integrity of the tart crust layer structure and the probability of sticking to the mold when the tart crust is demolded. Compare the predicted quality value with the preset quality target, and determine whether the preset stamping process parameters need to be adjusted based on the comparison results. If necessary, the preset stamping process parameters are adjusted according to the pre-established parameter optimization model to improve the predicted quality value, thereby obtaining the stamping process parameters to be executed. The stamping equipment is controlled to perform a stamping operation on the target dough according to the stamping process parameters.

2. The control method for tart crust processing equipment based on big data modeling as described in claim 1, characterized in that, The torque characteristic parameter includes the torque fluctuation coefficient, which is the ratio of the standard deviation to the average value of the torque signal of the stirring motor collected at a set sampling frequency within a complete mixing cycle.

3. The control method for tart crust processing equipment based on big data modeling as described in claim 2, characterized in that, The torque characteristic parameters include the proportion of low-frequency signal energy, which is the proportion of signal energy extracted from the 0.1Hz to 5Hz low-frequency band after performing a fast Fourier transform on the torque signal to the total signal energy.

4. The control method for tart crust processing equipment based on big data modeling as described in claim 1, characterized in that, The proofing process data includes the vertical temperature difference between the top and bottom of the proofing box along its height direction.

5. The control method for tart crust processing equipment based on big data modeling as described in claim 1, characterized in that, The proofing process data includes the vertical humidity difference between the top and bottom of the proofing box along its height direction.

6. The control method for tart crust processing equipment based on big data modeling as described in claim 1, characterized in that, The preset stamping process parameters are adjusted based on a pre-established parameter optimization model to improve the predicted quality, resulting in the execution stamping process parameters. The adjustment process includes adjusting the rate of change of the stamping pressure according to the tart crust score to form a protective pressure curve, and / or adjusting the execution parameters of the demolding aid according to the probability of sticking to the mold.

7. The control method for tart crust processing equipment based on big data modeling as described in claim 6, characterized in that, The step of adjusting the rate of change of the stamping pressure according to the tart crust score to form a protective pressure curve includes the following steps: When the tart crust score is lower than a first preset threshold, the first pressurization rate of the first stage is reduced; when the tart crust score is higher than or equal to the first preset threshold, the pressurization rate of the first stage is increased or reduced. When the tart crust score is lower than the second preset threshold, the second pressurization rate of the second stage after the first stage is reduced, and a pressure holding platform is inserted during the pressure rise; when the tart crust score is higher than or equal to the second preset threshold, the second pressurization rate of the second stage is adopted or increased, and the pressure holding platform is not inserted.

8. The control method for tart crust processing equipment based on big data modeling as described in claim 6, characterized in that, In the process of adjusting the execution parameters of the demolding aid device according to the probability of mold sticking risk, Based on the risk level to which the sticking risk probability belongs, the corresponding set of execution parameters for the demolding assistance device is selected from the predefined collaborative strategy library; the set of execution parameters for the demolding assistance device includes different ejection speeds and action sequences of the ejection mechanism, different vibration parameters and triggering times of the high-frequency vibrator, and whether to use spray-assisted demolding and the timing of spray triggering.

9. The control method for tart crust processing equipment based on big data modeling as described in claim 1, characterized in that, After controlling the stamping equipment to perform a stamping operation on the target dough according to the stamping process parameters, the method further includes the following steps: Obtain the actual product quality data of the target dough after the stamping process is completed, and associate this data with other process data and stamping process parameters involved in the target dough, and add it as a new historical data sample to the historical dataset of dough under the same batch identifier; The stamping quality prediction model and the parameter optimization model are updated using the historical dataset.

10. A control system for tart crust processing equipment based on big data modeling, used to implement the control method for tart crust processing equipment based on big data modeling as described in any one of claims 1-9, characterized in that, It includes: The data retrieval module is configured to obtain the batch identifier of the dough, and based on the batch identifier, obtain the dough kneading process data and proofing process data generated in the kneading and proofing processes; wherein, the kneading process data includes at least the torque characteristic parameters obtained by performing time-frequency domain analysis on the torque signal of the mixing motor; the proofing process data includes temperature and humidity data at at least two different spatial locations in the proofing environment and the proofing time; The quality prediction module is configured to acquire the pre-processing state data of the stamping equipment for the target dough stamping process, combine the dough kneading process data and the proofing process data to construct an input feature vector; input the input feature vector into a pre-trained stamping quality prediction model to obtain the quality prediction value of the tart crust finished product after stamping according to preset stamping process parameters; the quality prediction value includes at least the tart crust score characterizing the integrity of the tart crust layer structure and the probability of sticking to the mold when the tart crust is demolded; The comparison and adjustment module is configured to compare the predicted quality value with the preset quality target, and determine whether the preset stamping process parameters need to be adjusted based on the comparison result; if so, adjust the preset stamping process parameters according to the pre-established parameter optimization model to improve the predicted quality value, thereby obtaining the execution stamping process parameters; and control the stamping equipment to perform stamping operation on the target dough according to the execution stamping process parameters.

Images