Starch saccharification reaction state prediction and regulation method

Abstract

The invention provides a starch saccharification reaction state prediction, regulation and control method, which is applied to the technical field of food processing, and comprises the following steps: dividing a reaction system into a plurality of sub-regions, collecting temperature, slurry viscosity, starch particles, enzyme concentration and other sensing data of each sub-region, and calculating a local gelatinization degree index by using a preset gelatinization degree calculation model. According to the method, a saccharification reaction state prediction model is established, saccharification parameters of each sub-region are predicted in combination with the saccharification reaction state prediction model, then a saccharification state prediction result of the whole reaction system is generated, and the type, dosage and time of amylase putting are determined according to the deviation between the prediction result and a preset saccharification target. Therefore, the product quality and the saccharification efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of food processing technology, and in particular to a method for predicting and controlling the state of starch saccharification reaction. Background Technology

[0002] With the improvement of the level of automation in bio-fermentation and starch processing, starch saccharification reaction has been widely used in the production of starch sugar, alcohol, monosodium glutamate and enzyme preparations. As a key step in the conversion of starch into soluble sugars, the saccharification reaction is affected by a variety of factors such as temperature, pH value, enzyme concentration, degree of gelatinization and material mixing uniformity. In order to ensure the stability of the reaction process and product quality, it is usually necessary to predict and control the saccharification state.

[0003] In existing saccharification production systems, the prediction of the saccharification reaction state usually relies on theoretical models based on enzyme-catalyzed reaction kinetics, such as the Michaelis-Menten equation model or its simplified form. These models mainly calculate the rate of reducing sugar formation by establishing a functional relationship between reaction time, temperature, and substrate concentration, and thereby infer the trend of substrate concentration change over time.

[0004] However, the above models are generally based on the ideal assumptions of a uniform reaction system, complete substrate gelatinization, and constant enzyme activity. They ignore non-ideal factors in industrial production environments, such as batch differences in raw materials, uneven local gelatinization of the slurry, limited local mass transfer, and uneven enzyme distribution. Therefore, the existing technology lacks a method for predicting starch saccharification reactions that considers multiple process parameters. This makes the existing prediction models highly accurate only under laboratory-scale conditions, while the prediction error is large in actual continuous or large-scale production environments, resulting in reduced saccharification efficiency in actual production. Summary of the Invention

[0005] This application provides a method for predicting and controlling the saccharification reaction state of starch. The core of this method is to divide the starch slurry into multiple sub-regions in space, collect sensor data of each sub-region and calculate the local gelatinization index, input the local gelatinization index into the saccharification reaction state prediction model for prediction, determine the control strategy based on the deviation between the prediction result and the preset saccharification target, and generate control instructions including the type, dosage and time of amylase administration, thereby improving the saccharification efficiency of the saccharification reaction process.

[0006] To achieve the above objectives, this application adopts the following technical solution: This application provides a method for predicting and controlling the state of starch saccharification reaction, the method including: The starch slurry is spatially divided into multiple sub-regions, and sensor data of the corresponding starch slurry in multiple sub-regions are collected. The sensor data is input into the preset gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region. The local gelatinization degree index is used to characterize whether the starch slurry in the corresponding sub-region is in a state suitable for saccharification reaction. The local gelatinization indexes corresponding to multiple sub-regions are respectively input into the saccharification reaction state prediction model; In the saccharification reaction state prediction model, the local gelatinization degree index is used as a preset parameter. Based on the preset parameter, the saccharification reaction parameters of the corresponding sub-region in the next preset time period are predicted, and the saccharification reaction parameters are used as the corresponding prediction results. The predicted results are compared with the preset saccharification target to obtain the deviation results of the saccharification index. Based on the deviation results, the corresponding control strategy is determined, and the corresponding control command is generated according to the control strategy. The control command is sent to the corresponding execution device. The control strategy may include: the type of amylase, the dosage, and the timing of administration.

[0007] In some possible implementations, the sensing data may include: temperature data, slurry viscosity data, and starch particle data. Inputting the sensing data into a pre-defined gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region may include: The first characteristic parameter during the gelatinization process is calculated based on temperature data and slurry viscosity data. The first characteristic parameter may include: peak viscosity, gelatinization temperature, and viscosity change rate. The second characteristic parameter is calculated based on the starch granule data. The second characteristic parameter may include: the rate of decrease in the number of starch granules. Input the first feature parameter and the second feature parameter into the preset gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region.

[0008] In some possible implementations, inputting the first feature parameter and the second feature parameter into a preset gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region may include: The first characteristic parameter is input into the preset gelatinization degree calculation model, and the first gelatinization degree is calculated based on the peak viscosity, gelatinization temperature and viscosity change rate of the first characteristic parameter. The second feature parameter is input into the preset gelatinization degree calculation model. The first gelatinization degree is corrected according to the decrease rate of starch granule number of the second feature parameter to obtain the local gelatinization degree index of the corresponding sub-region.

[0009] In some possible implementations, the sensing data may include: pH value, starch concentration, and amylase concentration. In the saccharification reaction state prediction model, the local gelatinization degree index is used as a preset parameter. Based on the preset parameter, the saccharification reaction parameters of the corresponding sub-region in the next preset time period are predicted, which may include: In the saccharification reaction prediction model, the local gelatinization degree index is used as a preset correction parameter. The substrate conversion rate constant and product formation rate constant in the saccharification reaction prediction model are corrected according to the preset correction parameters to obtain the first set of rate constants; Temperature, pH, starch concentration, and amylase concentration are used as reaction condition parameters. Based on the reaction condition parameters and the first set of rate constants, the saccharification reaction parameters of the corresponding sub-region in the next preset time period are calculated.

[0010] In some possible implementations, the saccharification reaction parameters may include: DE value, dextrin content, reducing sugar formation rate, and reducing sugar concentration. Based on the reaction condition parameters and a first set of rate constants, the saccharification reaction parameters for the corresponding sub-region in the next preset time period are calculated, which may include: In the saccharification reaction prediction model, based on temperature data, pH value, starch concentration, amylase concentration and first rate constant set, the dextrin content and reducing sugar concentration in the next preset time period are calculated. The first rate constant set is used to compensate for the local differences in gelatinization degree and enzyme activity in sub-regions. The reducing sugar concentration is converted into glucose equivalents to obtain the DE value; The rate of reducing sugar production is calculated based on the reducing sugar concentration in multiple sub-regions and a preset time period. DE value, dextrin content, reducing sugar formation rate, and reducing sugar concentration were used as parameters for the saccharification reaction.

[0011] Among some possible implementation methods, the following may also be included: In the saccharification reaction prediction model, the local gelatinization degree index is used as a preset weight parameter, and the saccharification reaction parameters of the corresponding sub-region in the next preset time period are predicted based on the preset weight parameter.

[0012] In some possible implementations, predicting the saccharification reaction parameters of the corresponding sub-region in the next preset time period based on preset weight parameters may include: The starch concentration was adjusted based on preset weight parameters to obtain the substrate concentration parameters; In the saccharification reaction prediction model, based on substrate concentration parameters and reaction condition parameters, the saccharification reaction parameters of the corresponding sub-region in the next preset time period are predicted.

[0013] In some possible implementations, using saccharification reaction parameters as the corresponding prediction results can include: By fusing the saccharification reaction parameters corresponding to multiple sub-regions, the predicted results of starch saccharification reaction are obtained.

[0014] In some possible implementation methods, the predicted results are compared with the preset saccharification target to obtain the deviation results of the saccharification index. Based on the deviation results, the corresponding control strategy is determined, which may include: The predicted results are compared with the preset saccharification target to obtain the deviation results of the saccharification index; Based on the deviation results, determine the type of amylase to be added; Calculate the corresponding dosage and timing of amylase administration based on the deviation results; The type, dosage, and timing of amylase administration are used as corresponding regulatory strategies.

[0015] Among some possible implementation methods, the following may also be included: Based on the control strategy and the saccharification reaction parameters corresponding to the sub-region, the local control instructions corresponding to the sub-region are determined.

[0016] As can be seen from the above technical solution, this application has the following beneficial effects: 1. This application achieves the detection of saccharification reaction by dividing the starch slurry in the reactor into sub-regions and collecting multiple types of sensor data. Compared with the existing technology that only relies on the overall average parameters, the prediction accuracy and control specificity are improved.

[0017] 2. This application uses a saccharification state prediction model based on local gelatinization degree to predict saccharification parameters in each sub-region and compare them with the overall target. This allows for dynamic optimization of enzyme types, dosage, and timing, thereby improving saccharification efficiency and product uniformity.

[0018] 3. This application achieves refined saccharification control by distributing the overall control strategy to each sub-region and executing it automatically. Compared with manual or single control methods, it can reduce saccharification deviation and improve enzyme utilization and reaction stability. Attached Figure Description

[0019] The present application will be further described below with reference to the accompanying drawings.

[0020] Figure 1 A flowchart of the first method for predicting and controlling the state of starch saccharification reaction provided in this application; Figure 2 A flowchart of the second method for predicting and controlling the state of starch saccharification reaction provided in this application; Figure 3 A flowchart of the third method for predicting and controlling the state of starch saccharification reaction provided in this application. Detailed Implementation

[0021] The terms "first," "second," and "third," etc., used in this application specification, claims, and drawings are used to distinguish different objects, not to limit a specific order.

[0022] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0023] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first: Current methods for predicting the saccharification state of starch typically involve collecting overall physicochemical parameters of the starch slurry in the saccharification reactor, primarily including temperature, pH, viscosity, and dissolved oxygen concentration. Some systems also employ near-infrared spectroscopy, online chromatography, or optical sensing techniques to acquire indirect indicators such as the degree of starch gelatinization, reducing sugar concentration, or saccharification rate. These collected parameters are usually input into empirical formulas, kinetic models, or statistical regression models to calculate saccharification state indicators, such as degree of gelatinization, saccharification rate, or DE value, which characterize whether the starch slurry is in a suitable state for enzymatic hydrolysis. Based on these indicators, the system can use models to predict the changing trends of the saccharification reaction in the next time period, including the saccharification rate and the amount of reducing sugar produced. This allows for the control and adjustment of process parameters, such as adjusting reactor temperature, pH, or amylase dosage, enabling continuous monitoring and management of the saccharification process. The entire prediction process is based on overall slurry parameters and primarily relies on global average data for analysis and calculation.

[0024] In existing technologies, the measurement of starch gelatinization degree mainly relies on two methods: one is a rapid viscosity analyzer. This method continuously measures the viscosity curve of starch slurry under heating and stirring conditions as it changes with temperature, reflecting the changes in the flow characteristics of starch particles during heating, expansion, rupture, and gelatinization. The analyzer output curve contains several key characteristic parameters, such as gelatinization initiation temperature, peak viscosity, valley viscosity, and stable viscosity. These parameters can be used to indirectly estimate the degree of starch gelatinization, thereby assessing whether the slurry as a whole has reached a state suitable for saccharification. This method is relatively simple to operate and can provide overall dynamic viscosity information, but it mainly relies on... While reflecting the average state of the slurry, it is difficult to reflect the heterogeneity and local differences within the slurry's internal space. Another method is focused beam reflection measurement, which involves emitting a high-energy laser beam into the slurry and measuring the scattering signal reflected from the particle surface by the beam to obtain the particle size distribution, number, and morphological changes of starch particles in real time. As the starch particles expand, rupture, and gelatinize during heating, the dynamic changes in particle number and size can be captured, thereby indirectly calculating the degree of gelatinization of the slurry. This method is suitable for continuous saccharification processes, but it usually only reflects particle changes within the laser detection area and is difficult to comprehensively describe the spatial distribution of the entire slurry.

[0025] Research has found that in existing starch saccharification production systems, the prediction of the saccharification reaction status typically relies on enzyme-catalyzed reaction kinetic models. Researchers generally establish empirical functional relationships between reaction time, temperature, substrate concentration, and sugar production rate through experimental methods to predict the progress of the saccharification reaction. These models are mostly derived from laboratory-scale measurements of enzyme-catalyzed reactions. The core idea is that, under fixed temperature and pH conditions, enzymes in starch slurry catalyze the substrate reaction to produce reducing sugars or other measurable products. By monitoring the changes in the concentration of these products over time, the degree of starch conversion in the reaction system can be indirectly reflected. Therefore, predictive models often use the rate of change of substrate concentration as the core calculation indicator to determine the status and trend of the saccharification reaction.

[0026] In the process of model building, researchers usually assume that the reaction system is highly homogeneous in space. Specifically, the temperature, concentration and enzyme distribution in each part of the slurry are consistent. The reaction process can be regarded as the average behavior of the whole system. The substrate is assumed to be in a fully gelatinized state to ensure that starch molecules are uniformly exposed to the enzyme action sites and there is no reaction lag or inhibition caused by differences in physical structure. The model also assumes that the enzyme activity remains constant throughout the reaction process and is not affected by temperature fluctuations, substrate changes or by-product accumulation. Based on these idealized conditions, the prediction system can approximate the saccharification reaction state using overall parameters such as average temperature, overall pH value, or average viscosity. However, as saccharification processes gradually move towards continuous and large-scale production, prediction models established under laboratory conditions reveal significant limitations in industrial environments. Batch variations in raw materials significantly impact the system's reaction characteristics. Starch particles from different sources or processed using different techniques exhibit differences in particle size distribution, crystallinity, and branching ratio, leading to variations in their gelatinization initiation temperature, peak viscosity, and water absorption and swelling characteristics. Under the same heating or stirring conditions, the gelatinization progress of different batches of raw materials often differs, making it impossible to maintain a consistent degree of gelatinization within the system. In industrial reactors, due to the large slurry volume and limited stirring efficiency, localized formations can easily occur within the slurry. Due to temperature and viscosity gradients, some regions are fully heated and have good fluidity, while others, limited by stirring or pipeline structure, may experience insufficient heat transfer or incomplete material mixing. This localized unevenness directly leads to uneven gelatinization and limited substrate mass transfer, resulting in incomplete substrate gelatinization or insufficient enzyme contact with the substrate in some areas. Consequently, the local saccharification rate is significantly lower than the overall average. This leads to a fundamental problem with existing saccharification reaction state prediction methods: they neglect the coupling effects between multiple process variables in industrial production, especially the impact of gelatinization degree on saccharification reaction prediction. Therefore, when laboratory models are directly applied to continuous or large-scale production, their prediction errors increase significantly, the predicted results deviate from the actual saccharification state, and ultimately, the saccharification efficiency decreases.

[0027] Example 1 To address the aforementioned problems, this application provides a method for predicting and controlling the state of starch saccharification reaction. Please refer to [link / reference]. Figure 1 .

[0028] S101, Starch Slurry Division and Data Acquisition: In the saccharification reactor, to ensure the detection of the dynamics and non-uniformity of the reaction system, a partitioning strategy combining fluid dynamics modeling and physical parameter measurement was adopted. The reaction system was divided into multiple independent spatial sub-regions. The partitioning process comprehensively considered the slurry flow field distribution, differences in heat transfer efficiency, enzyme distribution uniformity, and raw material particle size distribution characteristics. Specifically, based on the type of agitator, blade angle, and slurry volume, a velocity and temperature field distribution model inside the reactor was first established using fluid numerical simulation. The simulation results clearly identified the locations of temperature gradients and local flow dead zones. The location of the stirring may lead to uneven mixing and insufficient gelatinization. Therefore, the reactor is divided into three functional sub-regions: the upper high-temperature homogenization zone, located at the top of the reactor, is driven by the stirring blades to generate strong circulation flow. It has a high temperature and relatively uniform enzyme distribution, and mainly undertakes heat transfer and gelatinization in the early stage of the reaction; the middle main reaction zone, which is the core area of ​​the entire saccharification reaction, with the highest substrate concentration, enzyme activity and reaction rate. Starch gelatinization and saccharification occur simultaneously. This zone is the main monitoring target of the prediction model; and the lower deposition zone, located near the bottom of the reactor, has poor fluidity and slow heat conduction. It often contains incompletely gelatinized starch particles, which is an important source of uneven saccharification.

[0029] For example, in an industrial saccharification tank, a paddle circulation flow model can be used to divide the area into three sub-regions with four detection units in each layer, for a total of 12 data acquisition nodes. Data is collected by setting an integrated temperature and viscosity composite probe and an optical particle detection module at each node. Of course, the above is just an example, and those skilled in the art can set it according to the actual specific situation, without making specific limitations here.

[0030] To accurately reflect the actual state of starch gelatinization and saccharification in each sub-region, a multi-type sensor joint acquisition system is adopted. For example, several sensing nodes are set in each functional layer. Each node includes a temperature probe, a viscosity probe, and an optical particle detection module to synchronously acquire reaction parameters of the corresponding sub-region. Each sensing unit is connected to the edge acquisition controller through a multi-channel data bus. Three temperature and three viscosity sensors are arranged in the upper high-temperature homogenization region to detect the heat conduction rate and viscosity change trend during the gelatinization initiation stage. Four temperature, four viscosity, and four particle detection sensors are arranged in the middle main reaction region to capture the saccharification rate, particle decomposition progress, and enzyme distribution uniformity. Two temperature probes and two optical particle detection probes are arranged in the lower deposition region to detect the residual amount of starch particles and the heat transfer hysteresis effect in the deposition layer. Through the above arrangement, the acquired sensing data includes temperature data, reflecting the heat energy distribution and heat exchange efficiency of different sub-regions; slurry viscosity data, used to infer the degree of local gelatinization reaction and the molecular chain breakage state; and starch particle data, with particle number and particle size changes recorded by the optical detection module to assess whether gelatinization is sufficient.

[0031] S102, Calculate the local degree of gelatinization based on the data, please refer to [link / reference]. Figure 2 The collected temperature data, slurry viscosity data, and starch particle data are input into the preset gelatinization degree calculation model to calculate the local gelatinization degree index of each sub-region, so as to characterize whether the starch slurry in that sub-region is in a suitable state for saccharification reaction.

[0032] The gelatinization degree calculation model refers to the calculation of the local gelatinization degree index of the slurry by collecting temperature, slurry viscosity and starch particle data in each sub-region and using existing technologies such as rapid viscosity analyzer and focused beam reflection measurement. This model calculates the local gelatinization degree index of each sub-region by imitating the temperature and viscosity curve change characteristics in rapid viscosity analyzer measurement and the particle number and particle size change characteristics in focused beam reflection measurement. This index is used to characterize whether the starch slurry in that region is suitable for saccharification reaction.

[0033] For example, temperature data from each sub-region is combined with slurry viscosity data to extract features such as peak viscosity, gelatinization temperature, and viscosity change rate, similar to the measurement principle of a rapid viscosity analyzer. These features indirectly reflect the gelatinization state of the slurry. By using data such as particle number change, particle size distribution, and particle descent rate, the detection results of focused beam reflection measurement are simulated to characterize the degree of starch particle breakage and gelatinization during heating and stirring. The simulated temperature and viscosity first characteristic parameters and the simulated particle second characteristic parameters are fused together and calculated. An adjustable weighting coefficient is set, and the weight value is jointly determined by the temperature range and enzyme activity state. For example, the weight of the first characteristic parameter is increased in the high-temperature stage, and the weight of the second characteristic parameter is increased in the particle disintegration stage. Through this weighting mechanism, the gelatinization degree estimation of different reaction stages is adjusted. The fusion result is compared with historical data to correct for deviations. The final output local gelatinization degree index is expressed in the form of a 0~1 range, where 1 represents complete gelatinization and 0 represents no gelatinization. Of course, the above is just an example, and those skilled in the art can set it according to the actual specific situation. No specific limitation is made here.

[0034] S103, based on the local gelatinization degree index, the saccharification reaction state was predicted, and the prediction results were obtained. Please refer to [link / reference]. Figure 3 After calculating the local gelatinization index of each sub-region, the local gelatinization index is input as a preset parameter into the saccharification reaction state prediction model. Combined with the reaction condition parameters such as temperature, pH value, starch concentration and amylase concentration of each sub-region, the saccharification reaction parameters of each sub-region in the next preset time period are predicted, and finally the saccharification state prediction result of the entire reaction system is generated.

[0035] For clarity and brevity in the description of the following embodiments, a brief explanation of some terms is provided: Local gelatinization index: refers to the degree of gelatinization of starch slurry in each sub-region, calculated through step S102. It is used to characterize whether the region has reached the conditions for suitable saccharification reaction. Its calculation method combines the temperature and viscosity characteristics simulated by a rapid viscosity analyzer with the particle breakage characteristics simulated by focused beam reflection measurement.

[0036] Saccharification reaction state prediction model: Based on the existing saccharification kinetics theory, the model is used to predict the saccharification reaction parameters of each sub-region by combining local gelatinization degree and reaction condition parameters, including dextrin content, reducing sugar concentration, DE value and reducing sugar formation rate, etc. The model adjusts the reaction rate constant by using local gelatinization degree as a correction or weighting parameter to correct the decrease in prediction accuracy caused by uneven local gelatinization.

[0037] Reaction condition parameters: including temperature, pH value, starch concentration, amylase concentration, etc. in each sub-region, used to describe the local environmental conditions of the saccharification reaction.

[0038] The first set of rate constants is based on the substrate conversion rate and product generation rate of the sub-region after correction of the local gelatinization index, and is used to compensate for local gelatinization unevenness and enzyme activity differences.

[0039] Here is an explanation of some parameter names: Dextrin content: Dextrin is an intermediate product of partial starch hydrolysis, with a molecular weight lower than starch but higher than monosaccharides. Dextrin content indicates the proportion or concentration of starch partially hydrolyzed into dextrin during the saccharification reaction. It reflects the degree of reaction in the early and middle stages of saccharification; a high dextrin content indicates that starch has begun to hydrolyze but has not yet completely formed reducing sugars. As a parameter of the saccharification reaction, dextrin content can be used to assess enzyme activity and the uniformity of the reaction.

[0040] DE value: also known as glucose equivalent, represents the ratio of the total amount of reducing sugar in the saccharification product to the amount of starch that has been completely hydrolyzed into glucose. It is usually expressed as a percentage. The higher the DE value, the deeper the degree of saccharification, and the more starch has been hydrolyzed into more absorbable reducing sugars. The DE value is an important indicator for measuring the starch saccharification endpoint and product quality, and can guide the amylase dosing strategy.

[0041] Reducing sugar concentration: The concentration of monosaccharides and oligosaccharides in the reaction system that can be detected by chemical or enzymatic methods. It is a directly measurable indicator in the actual saccharification product. The reducing sugar concentration directly determines the sweetness and fermentability of the final saccharified liquid. In the prediction model, it is used together with the DE value, dextrin content and formation rate to comprehensively evaluate the saccharification status of each subregion.

[0042] Reducing sugar formation rate: The amount of reducing sugar produced per unit time during saccharification. It can usually be calculated by measuring the change of reducing sugar concentration in the reaction system over time. The reducing sugar formation rate reflects reaction kinetics and is an important parameter for evaluating enzyme activity and reaction efficiency. By monitoring the reducing sugar formation rate in different sub-regions, problems such as uneven saccharification or uneven enzyme distribution in the reactor can be detected.

[0043] For example, the local gelatinization index of each sub-region is used as a preset correction parameter and input into the saccharification reaction prediction model. The model corrects the original reaction rate constant based on this parameter to form a first set of rate constants. Data such as temperature, pH value, starch concentration, and amylase concentration of each sub-region are collected synchronously through the acquisition node in step S101. The model combines the first set of rate constants with these reaction condition parameters to predict the saccharification reaction parameters in the next preset time period. This includes converting the predicted reducing sugar concentration into glucose equivalent to obtain the DE value, calculating the reducing sugar formation rate based on the change in reducing sugar concentration within the time period, and outputting the dextrin content of each sub-region. The DE value, dextrin content, reducing sugar formation rate, and reducing sugar concentration together constitute the saccharification reaction parameters of the sub-region. In addition, the local gelatinization index can also be used as a weighting parameter to locally adjust the starch concentration to form a substrate concentration parameter. The saccharification prediction model predicts the saccharification reaction parameters of each sub-region based on the adjusted substrate concentration parameter and reaction condition parameters, so that the prediction results more accurately reflect the actual saccharification situation in the sub-region.

[0044] The saccharification reaction parameters of all sub-regions are fused to generate a prediction result of the saccharification state of the overall reaction system, including the average DE value, overall dextrin content, total reducing sugar production rate and local deviation analysis. The fusion process can also identify uneven saccharification areas in the reactor, providing a reference for subsequent amylase dosing strategies.

[0045] S104. Based on the prediction results, implement corresponding control strategies. Compare the predicted saccharification state of the overall reaction system obtained in step S103 with the preset saccharification target, calculate the overall deviation of saccharification indicators, including the magnitude and direction of the deviation of each key indicator relative to the target value, and determine the overall control strategy based on the overall deviation results, including the type of amylase to be added, the total dosage, and the overall addition time window. After generating the overall strategy, combine the local prediction results of each sub-region to form the corresponding local control instructions for each sub-region, and rationally allocate the overall strategy to each sub-region to achieve refined saccharification control. Send the control instructions to the execution device (such as pump, stirrer, enzyme dosing device) to ensure that each sub-region operates according to the strategy.

[0046] Specifically, the predicted saccharification results are compared with the preset saccharification target to calculate the overall deviation, including the levels and magnitudes of each key indicator. Based on the type of deviation (e.g., excessively low DE value, excessively high dextrin content, insufficient reducing sugar production rate), the main control direction is determined. Based on the overall deviation analysis and raw material characteristics (starch type, particle size distribution, degree of pregelatinization), suitable enzymes are selected. α-Amylase: mainly used to break down starch macromolecules and reduce slurry viscosity, suitable for areas with insufficient gelatinization; β-amylase: used to produce dextrin, suitable for situations with low reducing sugar concentration; glucoamylase: used to increase the rate of reducing sugar production and DE value, suitable for situations with low overall saccharification. Multiple enzymes can be automatically combined and used according to different deviation index weights to form a mixed enzyme strategy.

[0047] Based on the overall deviation range and the local gelatinization index of each sub-region, the total dosage is determined, and the dosage is allocated to each sub-region: the dosage is increased for areas with incomplete gelatinization or delayed saccharification, and the dosage is maintained or slightly reduced for areas close to the target to avoid over-saccharification. The determination of the dosage also takes into account enzyme activity, reaction temperature, slurry concentration and stirring efficiency to ensure maximum enzyme utilization.

[0048] The timing of enzyme addition is adjusted based on the predicted saccharification progress and reaction kinetics: for areas with insufficient initial saccharification, enzymes can be added earlier or continuously to initiate the reaction; for areas that are close to the target in the middle and later stages, enzyme addition can be delayed or added in stages to avoid local over-saccharification. The timing of enzyme addition is also combined with the circulation characteristics of the stirring paddle and the temperature gradient to ensure that the enzyme is evenly distributed in the sub-region.

[0049] The overall strategy is quantitatively allocated to each sub-region, forming local control instructions: including the enzyme type, dosage, and time for each sub-region. The instructions are then sent to the corresponding execution devices (pumps, enzyme dispensers, etc.) to ensure fully automated execution.

[0050] For example, if the prediction shows that the overall DE value is low, while the upper and middle layers are close to the target and the lower deposition area has low gelatinization, then glucoamylase will be selected as the main enzyme, with a small amount of α-amylase as an auxiliary. The total dosage will be determined according to the overall deviation, and the enzyme will be added to the lower layer in advance to improve the saccharification progress. If the rate of reducing sugar generation is too fast and causes the local DE value to exceed the standard, the dosage of enzyme in the corresponding area will be reduced and the addition time will be delayed to prevent over-saccharification.

[0051] This application divides the starch slurry in the saccharification reactor into multiple independent sub-regions. Combined with a local gelatinization degree calculation model based on a rapid viscosity analyzer and focused beam reflection measurement principle, it assesses the gelatinization degree of each sub-region and uses this assessment as input to a saccharification reaction state prediction model. This model predicts the saccharification reaction parameters of each sub-region and the overall system. Based on the comparison between the overall prediction results and the preset saccharification target, it generates amylase dosing strategies for the whole system and sub-regions, including enzyme type, dosage, and dosing time. This enables precise control over saccharification unevenness, enzyme activity differences, and heat transfer efficiency differences within the reactor. Compared to existing technologies, this application significantly improves the uniformity and controllability of the saccharification reaction, enhances prediction accuracy and saccharification efficiency, optimizes enzyme resource utilization, improves production flexibility and automation, and ensures the stability of the saccharification endpoint and product quality, thereby achieving efficient saccharification reaction management.

[0052] The foregoing has shown and described the basic principles, main features, and advantages of this application. Those skilled in the art should understand that this application is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this application. Various changes and modifications can be made to this application without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of this application as claimed. The scope of protection of this application is defined by the appended claims and their equivalents.

Claims

1. A method for predicting and controlling the state of a starch saccharification reaction, characterized by, The method comprises: spatially dividing the starch slurry into a plurality of sub-regions, and collecting sensing data of the corresponding starch slurry in the plurality of sub-regions; inputting the sensing data into a preset gelatinization degree calculation model to calculate a local gelatinization degree index of the corresponding sub-region, the local gelatinization degree index being used to represent whether the starch slurry of the corresponding sub-region is in a state suitable for saccharification reaction; inputting the local gelatinization degree indexes of the plurality of sub-regions into a saccharification reaction state prediction model respectively; in the saccharification reaction state prediction model, taking the local gelatinization degree indexes as preset parameters, predicting saccharification reaction parameters of the corresponding sub-region in a next preset time period according to the preset parameters, and taking the saccharification reaction parameters as corresponding prediction results; comparing the prediction results with a preset saccharification target to obtain a deviation result of a saccharification index, determining a corresponding control strategy according to the deviation result, forming a corresponding control instruction according to the control strategy, and sending the control instruction to a corresponding execution device, the control strategy including: a type, a dosage, and a time of amylase delivery.

2. The method of claim 1, wherein, The sensing data includes temperature data, slurry viscosity data, and starch particle data, and the inputting of the sensing data into the preset gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region comprises: calculating a first characteristic parameter in a gelatinization process according to the temperature data and the slurry viscosity data, the first characteristic parameter including: peak viscosity, gelatinization temperature, and viscosity change rate; calculating a second characteristic parameter according to the starch particle data, the second characteristic parameter including: starch particle number reduction rate; inputting the first characteristic parameter and the second characteristic parameter into the preset gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region.

3. The method of claim 2, wherein, The inputting of the first characteristic parameter and the second characteristic parameter into the preset gelatinization degree calculation model to calculate the local gelatinization degree index of the corresponding sub-region comprises: inputting the first characteristic parameter into the preset gelatinization degree calculation model, and calculating a first gelatinization degree according to the peak viscosity, the gelatinization temperature, and the viscosity change rate of the first characteristic parameter; inputting the second characteristic parameter into the preset gelatinization degree calculation model, and correcting the first gelatinization degree according to the starch particle number reduction rate of the second characteristic parameter to obtain the local gelatinization degree index of the corresponding sub-region.

4. The method of claim 1, wherein, The sensing data includes pH value, starch concentration, and amylase concentration, and the inputting of the local gelatinization degree index as the preset parameter into the saccharification reaction state prediction model to predict the saccharification reaction parameters of the corresponding sub-region in the next preset time period according to the preset parameter comprises: in the saccharification reaction prediction model, taking the local gelatinization degree index as a preset correction parameter, correcting a substrate conversion rate constant and a product generation rate constant in the saccharification reaction prediction model according to the preset correction parameter to obtain a first rate constant set; The temperature data, the pH value, the starch concentration and the amylase concentration are taken as reaction condition parameters, and based on the reaction condition parameters and the first rate constant set, a saccharification reaction parameter of a corresponding sub-region in a next preset time period is calculated.

5. The method of claim 4, wherein, The saccharification reaction parameter includes a DE value, a dextrin content, a reducing sugar generation rate and a reducing sugar concentration, and the calculation of the saccharification reaction parameter of the corresponding sub-region in the next preset time period based on the reaction condition parameters and the first rate constant set includes: In the saccharification reaction prediction model, the dextrin content and the reducing sugar concentration in the next preset time period are calculated based on the temperature data, the pH value, the starch concentration, the amylase concentration and the first rate constant set, and the first rate constant set is used to compensate for the local gelatinization degree and enzyme activity difference of the sub-region; The reducing sugar concentration is converted into a glucose equivalent to obtain a DE value; The reducing sugar generation rate is calculated according to the reducing sugar concentration and the preset time period of the plurality of sub-regions; The DE value, the dextrin content, the reducing sugar generation rate and the reducing sugar concentration are taken as the saccharification reaction parameters.

6. The method of claim 4, wherein, Further comprising: In the saccharification reaction prediction model, the local gelatinization degree index is taken as a preset weight parameter, and the saccharification reaction parameter of the corresponding sub-region in the next preset time period is predicted according to the preset weight parameter.

7. The method of claim 6, wherein, The prediction of the saccharification reaction parameter of the corresponding sub-region in the next preset time period according to the preset weight parameter includes: Based on the preset weight parameter, the starch concentration is adjusted to obtain a substrate concentration parameter; In the saccharification reaction prediction model, the saccharification reaction parameter of the corresponding sub-region in the next preset time period is predicted based on the substrate concentration parameter and the reaction condition parameter.

8. The method of claim 1, wherein, The saccharification reaction parameter is taken as a corresponding prediction result, which includes: The saccharification reaction parameters of the plurality of sub-regions are fused to obtain a prediction result of the starch saccharification reaction.

9. The method of claim 1, wherein, The prediction result is compared with a preset saccharification target to obtain a deviation result of a saccharification index, and according to the deviation result, a corresponding control strategy is determined, which includes: The prediction result is compared with a preset saccharification target to obtain a deviation result of a saccharification index; According to the deviation result, the type of amylase to be put in is determined; According to the deviation result, the dosage and the time of putting in the corresponding amylase are calculated; The type, the dosage and the time of putting in the amylase are taken as the corresponding control strategy.

10. The method of claim 9, wherein, Further comprising: Based on the control strategy and the saccharification reaction parameter of the sub-region, a local control instruction corresponding to the sub-region is determined.

Images