Method for optimizing enzyme dosage in psicolase conversion process based on machine learning

Abstract

The invention relates to the technical field of enzyme preparation production, in particular to a method for optimizing the enzyme dosage in the psicose conversion process based on machine learning, and the method comprises the following steps: obtaining historical data of the psicose conversion process, and training a machine learning model based on the historical data; constructing an enzyme dosage optimization model, introducing constraint conditions of the conversion process into the model, and solving the optimal enzyme dosage of the current conversion process by utilizing the trained machine learning model; in the conversion process, environment dependent variable parameters are detected in real time, parameter change thresholds are set, and when the parameters exceed the threshold range, the required enzyme putting amount is recalculated; and according to the calculated optimal enzyme putting amount, constructing a PID algorithm to control the enzyme putting process. According to the method, machine learning, an optimization algorithm and an automatic control technology are comprehensively applied, intelligent optimization control over the psicoketase conversion process is achieved, the conversion efficiency is improved, the product quality is guaranteed, and remarkable technical and economic benefits are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of enzyme preparation production, and in particular to a method for optimizing enzyme dosage in a psicase conversion process based on machine learning. Background Art

[0002] Allulose is a low-calorie functional sweetener that is widely used in the food, beverage and other industries due to its excellent sweetness characteristics and extremely low caloric value. The production of allulose usually relies on an enzyme-catalyzed conversion process, in which allulose is produced by isomerization of fructose. In this process, the amount of enzyme used directly affects the efficiency and cost of the reaction. Excessive enzyme use increases production costs, while insufficient enzymes may lead to incomplete conversion, reducing yield and efficiency. Therefore, how to reasonably control the amount of enzyme added to ensure efficient conversion of the substrate while controlling costs is a key issue that needs to be urgently addressed in the allulose enzyme conversion process.

[0003] However, the application of existing technologies in the conversion process of allulose enzyme has shortcomings. First, most traditional enzyme delivery strategies rely on experience or fixed delivery schemes, and fail to make full use of historical data for intelligent optimization. Due to the large fluctuations in reaction conditions (such as temperature, pH value, raw material concentration, etc.) of different batches, the fixed enzyme delivery amount cannot adapt to different reaction environments, resulting in low enzyme utilization and increased production costs. Secondly, the existing optimization methods often fail to comprehensively consider the complex relationship between multiple reaction parameters (such as temperature, pH, stirring speed, etc.) and conversion rate, and lack the ability to dynamically adjust the enzyme delivery amount when the reaction conditions fluctuate. In addition, environmental dependent variables (such as temperature and pH value) often fluctuate during the conversion process, and the existing control technology cannot flexibly respond to these fluctuations, which in turn affects the activity and conversion efficiency of the enzyme. Finally, most of the existing enzyme delivery control methods are static control, lacking a real-time feedback adjustment mechanism, and are difficult to adjust according to the actual situation during the reaction process, resulting in inaccurate enzyme delivery, affecting the stability and economic benefits of production.

[0004] In view of this, the present invention proposes a method for optimizing enzyme dosage in the psicase conversion process based on machine learning. Summary of the invention

[0005] To achieve the above objectives, the present invention provides a method for optimizing enzyme dosage in a psicase conversion process based on machine learning, and the specific technical scheme is as follows:

[0006] The method for optimizing enzyme dosage in the psicase conversion process based on machine learning includes:

[0007] Obtain historical data of the psicase conversion process, and train machine learning models based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme dosage calculation model;

[0008] Construct an enzyme dosage optimization model for the allulose enzyme conversion process, introduce the constraints of the conversion process as the constraints of the machine learning model, and use the machine learning model to solve the optimal enzyme dosage for the current conversion process under the constraints of the conversion process;

[0009] Detecting the environmental dependent variable parameters of the allulose enzyme conversion process, designing the threshold value of the environmental dependent variable parameters of the allulose enzyme conversion process, and recalculating the required enzyme dosage when the environmental dependent variable parameters change beyond the threshold range;

[0010] According to the calculated optimal enzyme dosage, a PID algorithm was constructed to control the enzyme dosage process of the allulose enzyme conversion process.

[0011] Preferably, characteristic data of the psicase conversion process are obtained, including enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed , conversion time And conversion rate ;

[0012] Obtaining historical data batches of characteristic data of the allulose enzyme conversion process to construct a dataset ;in represents the enzyme dosage of the i-th batch, represents the concentration of the conversion raw material in the i-th batch, represents the conversion temperature of the i-th batch, represents the pH value of the i-th batch, represents the stirring speed of the i-th batch, represents the transformation time of the i-th batch, represents the conversion rate of the i-th batch;

[0013] Preprocessing the acquired data includes missing value processing, outlier processing and data normalization; the missing value processing includes filling the missing data with an interpolation method; the outlier processing includes identifying and removing outlier data points with a box plot method; the data normalization includes dimensionless processing of the data with a maximum and minimum value normalization method;

[0014] The preprocessed data set is denoted as ,in .

[0015] Preferably, the support vector machine (SVM) model is selected as the machine learning algorithm model, and the preprocessed data set is used. , train the support vector machine SVM model ;

[0016] During the SVM model training process, the enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed and conversion time As input, with conversion rate As output, train the conversion rate prediction model ;

[0017] Enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed And conversion rate As input, the conversion time T is output, and the conversion time prediction model is trained. ;

[0018] Raw material concentration , conversion temperature , pH value And stirring speed As input, enzyme dosage As output, calculate the training enzyme delivery calculation model .

[0019] Preferably, an enzyme dosage optimization model is constructed, and the optimization goal is to minimize the enzyme dosage. and conversion time , the optimization objective is expressed as: ; The constraints of the optimization target include the upper and lower limits of enzyme dosage, the lower limit of conversion rate and the upper limit of conversion time:

[0020]

[0021]

[0022]

[0023] in and are the lower and upper limits of enzyme dosage, is the minimum conversion rate requirement, The maximum conversion time limit; Represents the functional relationship between conversion rate and conversion parameters, It represents the functional relationship between the conversion time and the conversion parameters, which is predicted by the trained SVM model:

[0024]

[0025]

[0026]

[0027] Construct an enzyme dosage optimization model based on constraint conditions.

[0028] Preferably, the enzyme dosage optimization model is solved by using a non-dominated sorting genetic algorithm II to solve the multi-objective optimization problem of the enzyme dosage optimization model to obtain a Pareto optimal solution set; the solution algorithm process is as follows:

[0029] Step a: Initialize the population , each individual represents an enzyme dosage ;

[0030] Step b: For the current population , calculate the objective function value of each individual and , and perform non-dominated sorting and congestion calculation;

[0031] Step c: Generate offspring using binary tournament selection, crossover, and mutation operations ;

[0032] Step d: Merge and get ,right Perform non-dominated sorting and congestion calculation, select the first Excellent individuals form a new generation of population ;

[0033] Step e: Repeat step b to step d until the maximum number of iterations is reached;

[0034] Step f: Output the non-dominated solution set of the final population as the Pareto optimal solution set;

[0035] In the Pareto optimal solution set, select a satisfactory solution according to the needs As the optimal enzyme dosage:

[0036]

[0037] in, is the Pareto optimal solution set, and is the weight coefficient of enzyme dosage and conversion time;

[0038] The optimal enzyme dosage obtained was used as the initial enzyme dosage in the allulose enzyme conversion process.

[0039] Preferably, the environmental dependent variable parameters of the psicase conversion process are detected, and the environmental dependent variable parameters include temperature and pH value;

[0040] Set the temperature and pH value change thresholds during the conversion process, and set the first temperature change threshold to , the second temperature change threshold is , ; Let the first pH change threshold be , let the second pH change threshold be , , each interval during the conversion process Duration of testing of temperature and pH value;

[0041] The temperature during the detection conversion process is recorded as ,when When , the conversion process is judged to be normal, where Indicates the standard temperature of the conversion process;

[0042] when When the temperature of the conversion process is abnormal, the temperature of the conversion process is adjusted to ;

[0043] when When the temperature of the conversion process is out of control, the temperature of the conversion process is adjusted to When the current conversion parameters are obtained, including the raw material concentration , conversion temperature , pH value And stirring speed , input the conversion parameters into the enzyme delivery calculation model In the process, the amount of enzyme that remains active during the conversion process is detected, and the amount of enzyme that needs to be added is calculated;

[0044] The pH value during the conversion process is detected and recorded as ,when When the pH value of the conversion process is normal, is the standard pH value for the conversion process;

[0045] when When the pH value of the conversion process is abnormal, the conversion process is judged to be acid-base neutralization treatment;

[0046] when When the acid-base conversion process is judged to be out of control, the acid-base conversion process is first neutralized and the acid-base conversion process is neutralized to When the current conversion parameters are obtained, including the raw material concentration , conversion temperature , pH value And stirring speed , input the conversion parameters into the enzyme delivery calculation model In the process, the amount of enzyme that remains active during the conversion process is detected at the same time, and the amount of enzyme that needs to be added is calculated.

[0047] Preferably, the optimal enzyme dosage calculated is , construct a PID control algorithm to adjust the enzyme delivery rate of the allulose enzyme conversion process in real time ;make for The enzyme dosage error at the moment is:

[0048]

[0049] in, for The actual cumulative enzyme delivery at any moment is obtained through flow meter integration:

[0050]

[0051] PID control algorithm is based on Calculate the control amount , regulating the enzyme delivery rate :

[0052]

[0053]

[0054] in, is the proportionality coefficient; is the integration coefficient; is the differential coefficient; is the initial enzyme delivery rate; is the rate conversion factor.

[0055] An enzyme dosage optimization system in a psicase conversion process based on machine learning, which is used to implement the enzyme dosage optimization method in a psicase conversion process based on machine learning, comprising: a data acquisition module, an enzyme dosage optimization module, a process adjustment module, and a delivery control module;

[0056] The data acquisition module is used to acquire historical data of the psicase conversion process, and train a machine learning model based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme dosage calculation model;

[0057] The enzyme dosage optimization module is used to construct an enzyme dosage optimization model for the psicase conversion process, introduce the constraints of the conversion process as the limiting conditions of the machine learning model, and use the machine learning model to solve the optimal enzyme dosage of the current conversion process under the constraints of the conversion process;

[0058] The process adjustment module is used to detect the environmental dependent variable parameters of the psicase conversion process, design the threshold value of the environmental dependent variable parameters of the psicase conversion process, and recalculate the required enzyme dosage when the environmental dependent variable parameters change beyond the threshold range;

[0059] The delivery control module constructs a PID algorithm to control the enzyme delivery process of the psicase conversion process according to the calculated optimal enzyme delivery amount.

[0060] An electronic device comprises: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the method for optimizing enzyme dosage in a psicase conversion process based on machine learning by calling the computer program stored in the memory.

[0061] A computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the method for optimizing enzyme dosage in a psicase conversion process based on machine learning.

[0062] Beneficial effects of the present invention: The present invention can accurately predict the conversion rate, conversion time and optimize the enzyme dosage through analysis and modeling of historical data, providing an important basis for subsequent production optimization and improving production efficiency and product quality.

[0063] By introducing the constraints of the conversion process, the optimization model can calculate the optimal enzyme dosage while meeting the process requirements, effectively reducing production costs and improving economic benefits.

[0064] The present invention can detect abnormal situations in time by real-time monitoring of key environmental parameters of the conversion process, and dynamically adjust the amount of enzyme added according to the set threshold value, to ensure that the conversion process is always carried out under optimal conditions and to ensure stable product quality.

[0065] The present invention realizes automatic control of the enzyme delivery process based on the calculated optimal enzyme delivery amount and utilizes a PID control algorithm, which can accurately adjust the enzyme delivery rate, make the conversion process more stable and controllable, reduce the uncertainty of manual operation, and improve production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A flow chart of the method for optimizing enzyme dosage in the psicase conversion process based on machine learning provided by the present invention;

[0067] Figure 2 This is a structural diagram of the enzyme dosage optimization system in the psicase conversion process based on machine learning provided by the present invention. DETAILED DESCRIPTION

[0068] In order to better understand the present invention, a more detailed description will be made of various aspects of the present invention with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0069] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The accompanying drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in the present invention, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0070] It should also be understood that expressions such as "comprises", "including", "having", "includes" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present invention, "may" is used to mean "one or more embodiments of the present invention". And, the term "exemplary" is intended to refer to an example or illustration.

[0071] Unless otherwise defined, all terms (including engineering terms and scientific and technological terms) used in this document have the same meaning as commonly understood by ordinary technicians in the field to which the present invention belongs. It should also be understood that unless otherwise clearly stated in the present invention, words defined in commonly used dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0072] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0073] Example 1

[0074] Reference Figure 1 , which is the first embodiment of the present invention, provides a method for optimizing enzyme dosage in the psicase conversion process based on machine learning.

[0075] S1: Obtain historical data of the allulase conversion process, and train machine learning models based on the historical data of the allulase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme delivery amount calculation model.

[0076] Obtain characteristic data of the allulose enzyme conversion process, including: enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed , conversion time And conversion rate .

[0077] Obtaining historical data batches of characteristic data of the allulose enzyme conversion process to construct a dataset ;in represents the enzyme dosage of the i-th batch, represents the concentration of the conversion raw material in the i-th batch, represents the conversion temperature of the i-th batch, represents the pH value of the i-th batch, represents the stirring speed of the i-th batch, represents the transformation time of the i-th batch, represents the conversion rate of the i-th batch.

[0078] The acquired data is preprocessed, including missing value processing, outlier processing and data normalization; the missing value processing includes filling the missing data with an interpolation method; the outlier processing includes identifying and removing abnormal data points using a box plot method; the data normalization includes dimensionlessly processing the data using a maximum and minimum value normalization method.

[0079] The preprocessed data set is denoted as ,in .

[0080] In step S1, by acquiring and preprocessing the historical data of the allulase conversion process, a machine learning model was constructed, including conversion rate prediction, conversion time prediction, and enzyme dosage calculation. This laid the data and model foundation for the subsequent enzyme dosage optimization and helped to improve the efficiency and economy of the allulase conversion process.

[0081] S2: Construct an enzyme dosage optimization model for the allulose enzyme conversion process, introduce the constraints of the conversion process as the constraints of the machine learning model, and use the machine learning model to solve the optimal enzyme dosage for the current conversion process under the constraints of the conversion process.

[0082] Select the support vector machine SVM model as the machine learning algorithm model and use the preprocessed data set , train the support vector machine SVM model .

[0083] The dataset Divide into training set and test set;

[0084] During the SVM model training process, the enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed and conversion time As input, with conversion rate As output, train the conversion rate prediction model .

[0085] Enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed And conversion rate As input, to convert time For output, train the conversion time prediction model .

[0086] Raw material concentration , conversion temperature , pH value And stirring speed As input, enzyme dosage As output, calculate the training enzyme delivery calculation model .

[0087] Given the preprocessed training dataset ,in For the The input feature vector of samples, is the corresponding target value, and n is the number of training samples.

[0088] Using the training dataset Train the support vector regression SVM model. The optimization goal of SVM is to find the optimal and :

[0089]

[0090] The constraints are:

[0091]

[0092]

[0093]

[0094] in, is the normal vector of the hyperplane, is the bias term; and is a slack variable, indicating the The error allowed for each sample; is the penalty coefficient, which controls the balance between model complexity and error; It is the threshold of the insensitive loss function and controls the tolerance of the model.

[0095] By solving the above optimization problem, the decision function of SVM can be obtained:

[0096]

[0097] in, and is the Lagrange multiplier of the optimization problem, is the Gaussian kernel function.

[0098] During the training process, the optimal hyperparameters are selected through methods such as cross-validation. and As well as the parameters of the kernel function, hyperparameters are tuned using methods such as grid search or Bayesian optimization to obtain the SVM model with the best performance.

[0099] The k-fold cross-validation method was used to evaluate the generalization performance of the model on the test set; the mean square error (MSE) and the coefficient of determination (CDR) were used to evaluate the generalization performance of the model on the test set. The indicator measures the regression effect of the model:

[0100]

[0101]

[0102] in is the number of test set samples, is the mean conversion rate of the test set; The smaller, The closer it is to 1, the better the model fit is.

[0103] Finally, the trained SVM model is used to predict the new input data to obtain the corresponding predicted values ​​of conversion rate, conversion time or enzyme dosage.

[0104] Construct an enzyme dosage optimization model, with the optimization goal of minimizing enzyme dosage and conversion time , where enzyme dosage It is a key factor affecting cost and time; the optimization goal can be expressed as:

[0105]

[0106] The constraints of the optimization objective include the upper and lower limits of enzyme dosage, the lower limit of conversion rate, and the upper limit of conversion time:

[0107]

[0108]

[0109]

[0110] in and are the lower and upper limits of enzyme dosage, is the minimum conversion rate requirement, The maximum conversion time limit; Represents the functional relationship between conversion rate and conversion parameters, It represents the functional relationship between the conversion time and the conversion parameters, which is predicted by the trained SVM model:

[0111]

[0112]

[0113]

[0114] Construct an enzyme dosage optimization model based on constraint conditions.

[0115] Solve the enzyme dosage optimization model, use the non-dominated sorting genetic algorithm II to solve the multi-objective optimization problem of the enzyme dosage optimization model, and obtain the Pareto optimal solution set; the solution algorithm process is as follows:

[0116] Step a: Initialize the population , each individual represents an enzyme dosage ;

[0117] Step b: For the current population , calculate the objective function value of each individual and , and perform non-dominated sorting and congestion calculation;

[0118] Step c: Generate offspring using binary tournament selection, crossover, and mutation operations ;

[0119] Step d: Merge and get ,right Perform non-dominated sorting and congestion calculation, select the first Excellent individuals form a new generation of population ;

[0120] Step e: Repeat step b to step d until the maximum number of iterations is reached;

[0121] Step f: Output the non-dominated solution set of the final population as the Pareto optimal solution set.

[0122] In the Pareto optimal solution set, select a satisfactory solution according to the needs As the optimal enzyme dosage:

[0123]

[0124] in, is the Pareto optimal solution set, and is the weight coefficient of enzyme dosage and conversion time;

[0125] The optimal enzyme dosage obtained was used as the initial enzyme dosage in the allulose enzyme conversion process.

[0126] In step S2, the support vector machine algorithm was used to train the conversion rate prediction model, the conversion time prediction model, and the enzyme dosage calculation model, and an enzyme dosage optimization model was constructed with the goal of minimizing enzyme dosage and conversion time. By solving the optimization model, the optimal enzyme dosage for the allulose enzyme conversion process was obtained, which can effectively reduce production costs and shorten conversion time.

[0127] S3: Detect the environmental dependent variable parameters of the allulose enzyme conversion process, design the threshold value of the environmental dependent variable parameter change of the allulose enzyme conversion process, and recalculate the required enzyme dosage when the environmental dependent variable parameter change exceeds the threshold range.

[0128] The environmental dependent variable parameters of the D-psicase conversion process are detected, wherein the environmental dependent variable parameters include temperature and pH value. Changes in temperature and pH value will affect enzyme activity. Excessive changes in temperature and pH value will lead to enzyme inactivation, thereby terminating the conversion.

[0129] Set the temperature and pH value change thresholds during the conversion process. According to the enzyme activity range of psicase, set the first temperature change threshold to , the second temperature change threshold is , ; Let the first pH change threshold be , let the second pH change threshold be , , each interval during the conversion process Time to detect temperature and pH value.

[0130] The temperature during the detection conversion process is recorded as ,when When , the conversion process is judged to be normal, where Indicates the standard temperature of the conversion process.

[0131] when When the temperature of the conversion process is abnormal, the temperature of the conversion process is adjusted to .

[0132] when When the temperature of the conversion process is out of control, the temperature of the conversion process is adjusted to When the current conversion parameters are obtained, including the raw material concentration , conversion temperature , pH value And stirring speed , input the conversion parameters into the enzyme delivery calculation model In the process, the amount of enzyme that remains active during the conversion process is detected at the same time, and the amount of enzyme that needs to be added is calculated.

[0133] The pH value during the conversion process is detected and recorded as ,when When the pH value of the conversion process is normal, It is the standard pH value for the conversion process.

[0134] when When the pH value of the conversion process is determined to be abnormal, the conversion process is subjected to acid-base neutralization treatment.

[0135] when When the acid-base conversion process is judged to be out of control, the acid-base conversion process is first neutralized and the acid-base conversion process is neutralized to When the current conversion parameters are obtained, including the raw material concentration , conversion temperature , pH value And stirring speed , input the conversion parameters into the enzyme delivery calculation model In the process, the amount of enzyme that remains active during the conversion process is detected at the same time, and the amount of enzyme that needs to be added is calculated.

[0136] Step S3 detects environmental dependent variable parameters such as temperature and pH value in the allulose enzyme conversion process in real time, sets reasonable change thresholds, and promptly adjusts and recalculates the enzyme dosage when the parameters exceed the threshold range. This can ensure that the conversion process is carried out under optimal conditions, avoid enzyme inactivation and conversion termination due to abnormal environmental parameters, and improve production efficiency and product quality.

[0137] S4: Based on the calculated optimal enzyme dosage, a PID algorithm is constructed to control the enzyme dosage process of the allulose enzyme conversion process.

[0138] The optimal enzyme dosage calculated , construct a PID control algorithm to adjust the enzyme delivery rate of the allulose enzyme conversion process in real time ;make for The enzyme dosage error at the moment is:

[0139]

[0140] in, for The actual enzyme delivery amount at each moment is obtained by integrating the flow meter:

[0141]

[0142] PID control algorithm is based on Calculate the control amount , regulating the enzyme delivery rate :

[0143]

[0144]

[0145] in, is the proportionality coefficient; is the integration coefficient; is the differential coefficient; is the initial enzyme delivery rate; is the rate conversion coefficient, and the control quantity Converted to actual enzyme delivery rate adjustment.

[0146] Step S4 adjusts the three parameters of the PID controller , , , can realize the precise control of enzyme delivery process, so that the actual enzyme delivery amount Converges to the optimum value quickly and smoothly , ensuring the optimization of the conversion process.

[0147] Example 2

[0148] Reference Figure 2 , which is the second embodiment of the present invention, provides an enzyme dosage optimization system in the psicase conversion process based on machine learning.

[0149] The system comprises: a data acquisition module, an enzyme dosage optimization module, a process adjustment module and a delivery control module.

[0150] The data acquisition module is used to obtain historical data of the psicase conversion process, and train a machine learning model based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme dosage calculation model.

[0151] The enzyme dosage optimization module is used to construct an enzyme dosage optimization model for the allulose enzyme conversion process, introduce the constraints of the conversion process as the constraints of the machine learning model, and use the machine learning model to solve the optimal enzyme dosage for the current conversion process under the constraints of the conversion process.

[0152] The process regulation module is used to detect the environmental dependent variable parameters of the psicase conversion process, design the threshold value of the environmental dependent variable parameter change of the psicase conversion process, and recalculate the required enzyme dosage when the environmental dependent variable parameter change exceeds the threshold range.

[0153] The delivery control module constructs a PID algorithm to control the enzyme delivery process of the psicase conversion process according to the calculated optimal enzyme delivery amount.

[0154] Example 3

[0155] The present invention also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the method for optimizing the enzyme dosage in the psicase conversion process based on machine learning as described above can be executed.

[0156] The method or system according to the embodiment of the present invention may also be implemented with the aid of the architecture of the electronic device of the present invention.

[0157] An electronic device may include a bus, one or more CPUs, read-only memory (ROM), random access memory (RAM), a communication port for connecting to a network, input / output components, a hard disk, and the like.

[0158] A storage device in an electronic device, such as a ROM or a hard disk, can store the method for optimizing enzyme dosage in a psicase conversion process based on machine learning provided by the present invention.

[0159] The method for optimizing the amount of enzyme in the psicase conversion process based on machine learning includes: obtaining historical data of the psicase conversion process, training a machine learning model based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme delivery amount calculation model; constructing an enzyme dosage optimization model for the psicase conversion process, introducing the constraints of the conversion process as the constraints of the machine learning model, and solving the optimal enzyme delivery amount of the current conversion process based on the constraints of the conversion process using the machine learning model; detecting the environmental dependent variable parameters of the psicase conversion process, designing the threshold value of the environmental dependent variable parameters of the psicase conversion process, and recalculating the required enzyme delivery amount when the environmental dependent variable parameters change beyond the threshold range; constructing a PID algorithm to control the enzyme delivery process of the psicase conversion process based on the calculated optimal enzyme delivery amount. Further, the electronic device may also include a user interface. Of course, the architecture of the present invention is only exemplary. When implementing different devices, one or more components in the electronic device of the present invention may be omitted according to actual needs.

[0160] Example 4

[0161] One embodiment of the present invention discloses a computer-readable storage medium.

[0162] The computer-readable storage medium has computer-readable instructions stored thereon.

[0163] When the computer-readable instructions are executed by a processor, the method for optimizing enzyme dosage in a psicase conversion process based on machine learning according to an embodiment of the present invention described with reference to the above drawings may be executed.

[0164] The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. In addition, according to an embodiment of the present invention, the process described above with reference to the flowchart may be implemented as a computer software program.

[0165] For example, the present invention provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided by the present invention, for example: obtaining historical data of a psicase conversion process, training a machine learning model based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme dosage calculation model; constructing an enzyme dosage optimization model for the psicase conversion process, introducing the constraints of the conversion process as constraints of the machine learning model, and using the machine learning model to solve the optimal enzyme dosage of the current conversion process under the constraints of the conversion process; detecting the environmental dependent variable parameters of the psicase conversion process, designing the threshold value of the environmental dependent variable parameter change of the psicase conversion process, and recalculating the required enzyme dosage when the environmental dependent variable parameter changes beyond the threshold range; constructing a PID algorithm to control the enzyme dosage process of the psicase conversion process according to the calculated optimal enzyme dosage.

[0166] When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present invention are performed. The method, apparatus, and device of the present invention may be implemented in many ways. For example, the method, apparatus, and device of the present invention may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware.

[0167] The above sequence for the steps of the method is for illustration only, and the steps of the method of the present invention are not limited to the sequence specifically described above unless otherwise specifically stated.

[0168] In addition, in some embodiments, the present invention can also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present invention. Therefore, the present invention also covers a recording medium storing a program for executing the method according to the present invention.

[0169] In addition, the parts of the above technical solutions provided in the embodiments of the present invention that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0170] The specific implementation modes described above further describe the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing enzyme dosage in the psicase conversion process based on machine learning, characterized in that: include: Obtain historical data of the psicase conversion process, and train machine learning models based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme dosage calculation model; Construct an enzyme dosage optimization model for the allulose enzyme conversion process, introduce the constraints of the conversion process as the constraints of the machine learning model, and use the machine learning model to solve the optimal enzyme dosage for the current conversion process under the constraints of the conversion process; Detecting the environmental dependent variable parameters of the allulose enzyme conversion process, designing the threshold value of the environmental dependent variable parameters of the allulose enzyme conversion process, and recalculating the required enzyme dosage when the environmental dependent variable parameters change beyond the threshold range; According to the calculated optimal enzyme dosage, a PID algorithm was constructed to control the enzyme dosage process of the allulose enzyme conversion process.

2. The method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to claim 1, characterized in that: Obtain characteristic data of the psicase conversion process, including enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed , conversion time And conversion rate ; Obtaining historical data batches of characteristic data of the allulose enzyme conversion process to construct a dataset ;in represents the enzyme dosage of the i-th batch, represents the concentration of the conversion raw material in the i-th batch, represents the conversion temperature of the i-th batch, represents the pH value of the i-th batch, represents the stirring speed of the i-th batch, represents the transformation time of the i-th batch, represents the conversion rate of the i-th batch; Preprocessing the acquired data includes missing value processing, outlier processing and data normalization; the missing value processing includes filling the missing data with an interpolation method; the outlier processing includes identifying and removing outlier data points with a box plot method; the data normalization includes dimensionless processing of the data with a maximum and minimum value normalization method; The preprocessed data set is denoted as ,in .

3. The method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to claim 2, characterized in that: Select the support vector machine SVM model as the machine learning algorithm model and use the preprocessed data set , train the support vector machine SVM model ; During the SVM model training process, the enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed and conversion time As input, with conversion rate As output, train the conversion rate prediction model ; Enzyme dosage , Conversion raw material concentration , conversion temperature , pH value , stirring speed And conversion rate As input, the conversion time T is output, and the conversion time prediction model is trained. ; Raw material concentration , conversion temperature , pH value And stirring speed As input, enzyme dosage As output, calculate the training enzyme delivery calculation model .

4. The method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to claim 3, characterized in that: Construct an enzyme dosage optimization model, with the optimization goal of minimizing enzyme dosage and conversion time , the optimization objective is expressed as: ; The constraints of the optimization target include the upper and lower limits of enzyme dosage, the lower limit of conversion rate and the upper limit of conversion time: ； ； ； in and are the lower and upper limits of enzyme dosage, is the minimum conversion rate requirement, The maximum conversion time limit; Represents the functional relationship between conversion rate and conversion parameters, It represents the functional relationship between the conversion time and the conversion parameters, which is predicted by the trained SVM model: ； ； ； Construct an enzyme dosage optimization model based on constraint conditions.

5. The method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to claim 4, characterized in that: Solve the enzyme dosage optimization model, use the non-dominated sorting genetic algorithm II to solve the multi-objective optimization problem of the enzyme dosage optimization model, and obtain the Pareto optimal solution set; the solution algorithm process is as follows: Step a: Initialize the population , each individual represents an enzyme dosage ; Step b: For the current population , calculate the objective function value of each individual and , and perform non-dominated sorting and congestion calculation; Step c: Generate offspring using binary tournament selection, crossover, and mutation operations ; Step d: Merge and get ,right Perform non-dominated sorting and congestion calculation, select the first Excellent individuals form a new generation of population ; Step e: Repeat step b to step d until the maximum number of iterations is reached; Step f: Output the non-dominated solution set of the final population as the Pareto optimal solution set; In the Pareto optimal solution set, select a satisfactory solution according to the needs As the optimal enzyme dosage: ； in, is the Pareto optimal solution set, and is the weight coefficient of enzyme dosage and conversion time; The optimal enzyme dosage obtained was used as the initial enzyme dosage in the allulose enzyme conversion process.

6. The method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to claim 5, characterized in that: detecting environmental dependent variable parameters of the psicase conversion process, wherein the environmental dependent variable parameters include temperature and pH value; Set the temperature and pH value change thresholds during the conversion process, and set the first temperature change threshold to , the second temperature change threshold is , ; Let the first pH change threshold be , let the second pH change threshold be , , each interval during the conversion process Duration of testing of temperature and pH value; The temperature during the detection conversion process is recorded as ,when When , the conversion process is judged to be normal, where Indicates the standard temperature of the conversion process; when When the temperature of the conversion process is abnormal, the temperature of the conversion process is adjusted to ; when When the temperature of the conversion process is out of control, the temperature of the conversion process is adjusted to When the current conversion parameters are obtained, including the raw material concentration , conversion temperature , pH value And stirring speed , input the conversion parameters into the enzyme delivery calculation model In the process, the amount of enzyme that remains active during the conversion process is detected, and the amount of enzyme that needs to be added is calculated; The pH value during the conversion process is detected and recorded as ,when When the pH value of the conversion process is normal, is the standard pH value for the conversion process; when When the pH value of the conversion process is abnormal, the conversion process is judged to be acid-base neutralization treatment; when When the acid-base conversion process is judged to be out of control, the acid-base conversion process is first neutralized and the acid-base conversion process is neutralized to When the current conversion parameters are obtained, including the raw material concentration , conversion temperature , pH value And stirring speed , input the conversion parameters into the enzyme delivery calculation model In the process, the amount of enzyme that remains active during the conversion process is detected at the same time, and the amount of enzyme that needs to be added is calculated.

7. The method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to claim 6, characterized in that: The optimal enzyme dosage calculated , construct a PID control algorithm to adjust the enzyme delivery rate of the allulose enzyme conversion process in real time ;make for The enzyme dosage error at the moment is: ； in, for The actual cumulative enzyme delivery at any moment is obtained through flow meter integration: ； PID control algorithm is based on Calculate the control amount , regulating the enzyme delivery rate : ； ； in, is the proportionality coefficient; is the integration coefficient; is the differential coefficient; is the initial enzyme delivery rate; is the rate conversion factor.

8. A system for optimizing enzyme dosage in a psicase conversion process based on machine learning, which is used to implement the method for optimizing enzyme dosage in a psicase conversion process based on machine learning according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, enzyme dosage optimization module, process adjustment module and delivery control module; The data acquisition module is used to acquire historical data of the psicase conversion process, and train a machine learning model based on the historical data of the psicase conversion process, including a conversion rate prediction model, a conversion time prediction model, and an enzyme dosage calculation model; The enzyme dosage optimization module is used to construct an enzyme dosage optimization model for the psicase conversion process, introduce the constraints of the conversion process as the limiting conditions of the machine learning model, and use the machine learning model to solve the optimal enzyme dosage of the current conversion process under the constraints of the conversion process; The process adjustment module is used to detect the environmental dependent variable parameters of the psicase conversion process, design the threshold value of the environmental dependent variable parameters of the psicase conversion process, and recalculate the required enzyme dosage when the environmental dependent variable parameters change beyond the threshold range; The delivery control module constructs a PID algorithm to control the enzyme delivery process of the psicase conversion process according to the calculated optimal enzyme delivery amount.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the method for optimizing enzyme dosage in the psicase conversion process based on machine learning according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer executes the method for optimizing enzyme dosage in a psicase conversion process based on machine learning as described in any one of claims 1 to 7.

Images