An Optimization and Intelligent Decision-making Method and System for Abalone Peptide Production Process

Abstract

The present invention relates to the technical field of abalone peptide production, and discloses a method and system for optimizing the production process of abalone peptides and intelligent decision-making. Production data is collected by multi-source sensors, and features are extracted by a convolutional neural network. The genetic algorithm optimization model combines an adaptive mechanism and a multi-objective fitness function to generate an optimized process parameter scheme; the multi-objective production optimization model outputs the optimal production process parameters by means of an improved particle swarm algorithm. In the hierarchical intelligent decision-making model, the planning layer realizes global resource allocation, the scheduling layer adjusts the process in real time based on a Bayesian network, etc., and the operation layer controls the equipment through fuzzy logic and constructs a quality closed-loop control. The invention overcomes problems in traditional production such as parameter control relying on experience, difficult data processing, one-sided optimization, and non-intelligent decision-making, can accurately process data, optimize the process, and make intelligent decisions, effectively improving production efficiency, raw material utilization rate, and product quality, reducing costs, and having broad application prospects.

Description

Technical Field

[0001] The present invention relates to the technical field of abalone peptide production, and specifically to an optimization method and system for abalone peptide production process and intelligent decision-making. Background Art

[0002] Abalone peptide is a short peptide with various biological activities extracted from abalone, and has broad application prospects in the fields of food, medicine, health products, etc. With the continuous increase in market demand for abalone peptide, how to optimize its production process, improve production efficiency, raw material utilization rate and product quality has become a key research direction in this field.

[0003] In the traditional abalone peptide production process, the control of process parameters mainly relies on manual experience, lacking accuracy and scientificity. For example, in the enzymatic hydrolysis reaction link, the setting of parameters such as reaction temperature, pH value, enzyme addition amount and reaction time is often determined according to the long-term accumulated experience of operators. The experience differences of different operators may lead to inconsistencies in the control of process parameters, thus affecting the production efficiency of abalone peptide and the stability of product quality. Moreover, it is difficult for manual experience to comprehensively consider the interaction of various complex factors in the production process and cannot achieve the optimal configuration of process parameters.

[0004] At the same time, there are also great problems in the processing of multi-source sensor data. Although various sensors such as temperature sensors, pH value sensors, and enzyme activity sensors are used to collect data during the production process, due to the characteristics of high-dimensionality, multi-modal and dynamic changes of these sensor data, traditional data processing methods are difficult to effectively extract the key information. A large amount of redundant data not only increases the burden of data storage and transmission, but also may interfere with the accurate judgment of the production process, making potential problems in the production process unable to be discovered and solved in time.

[0005] In terms of production process optimization, most of the existing technologies adopt single-objective optimization methods, such as simply pursuing the improvement of production efficiency or the increase of raw material utilization rate, while ignoring other important factors. This one-sided optimization method may lead to the sacrifice of other indicators while improving a certain indicator, and cannot achieve the maximization of production benefits. For example, in order to improve production efficiency, the amount of enzyme used is increased excessively, which may accelerate the reaction speed, but will lead to a decrease in raw material utilization rate, increase production costs at the same time, and may also have an adverse impact on product quality.

[0006] In addition, the application of intelligent decision-making in abalone peptide production is not mature enough. Current production decisions are often based on post-event data analysis and experience judgment, lacking real-time and foresight. When faced with emergencies or equipment failures in the production process, it is impossible to make effective decisions quickly, which can easily lead to production interruptions or product quality degradation. Moreover, traditional production control systems lack the ability to optimize and coordinate the production process globally, and information transmission between various production links is not smooth, making it difficult to achieve overall optimization of the production process. Summary of the invention

[0007] The object of the present invention is to provide a method and system for optimizing the production process of abalone peptides and intelligent decision-making, so as to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: an abalone peptide production process optimization and intelligent decision-making method, the method comprising:

[0009] Collecting real-time process data in the abalone peptide production process through multi-source sensors, wherein the multi-source sensors include a temperature sensor, a pH sensor, an enzyme activity sensor, and a pressure sensor; performing multi-scale feature extraction on the real-time process data based on a convolutional neural network to obtain process feature data;

[0010] The process characteristic data is input into a pre-trained genetic algorithm optimization model, wherein the genetic algorithm optimization model adopts an adaptive crossover rate and mutation rate mechanism, iteratively optimizes the process parameter combination based on a multi-objective fitness function, and generates a process parameter optimization plan;

[0011] Constructing a multi-objective production optimization model according to the process parameter optimization scheme, wherein the multi-objective production optimization model takes maximizing production efficiency and maximizing raw material utilization as optimization objectives, and adopts an improved particle swarm algorithm to globally optimize the production process, wherein the improved particle swarm algorithm introduces an inertia weight dynamic adjustment mechanism and a neighborhood search strategy; outputting optimal production process parameters based on the multi-objective production optimization model;

[0012] A hierarchical intelligent decision-making model is established according to the optimal production process parameters, and the hierarchical intelligent decision-making model includes a planning layer, a scheduling layer and an operation layer, wherein the planning layer performs global resource allocation planning based on the process parameter optimization plan, the scheduling layer performs dynamic parameter scheduling based on the optimal production process parameters, and the operation layer realizes the execution control of the production equipment based on the fuzzy logic control algorithm; the production control instructions are output through the hierarchical intelligent decision-making model to realize the optimization and intelligent decision-making of the abalone peptide production process.

[0013] Preferably, input the process feature data into a pre-trained genetic algorithm optimization model. The genetic algorithm optimization model adopts an adaptive crossover rate and mutation rate mechanism, and iteratively optimizes the process parameter combination based on a multi-objective fitness function. The generated process parameter optimization scheme includes:

[0014] Obtain real-time process data, which includes the reactor temperature, the pH value of the enzymatic hydrolysate, the enzyme activity concentration, the raw material input rate, and the reaction pressure; construct a parameter space based on the real-time process data, and construct a decision space based on the temperature change amount, pH adjustment amount, enzyme addition amount, and pressure adjustment amount that can be adjusted in the process;

[0015] Construct a multi-objective fitness function based on the parameter space and the decision space. The multi-objective fitness function includes a production efficiency term, a raw material utilization rate term, an energy consumption term, and a stability term. Among them, the production efficiency term is calculated by the ratio of the product output per unit time to the target output, the raw material utilization rate term is calculated by the ratio of the actual product quality to the theoretical maximum product quality, the energy consumption term is calculated by the weighted sum of squares of the temperature change amount, pH adjustment amount, and pressure adjustment amount, and the stability term is calculated by the Euclidean distance between the current process parameters and the historical optimal parameters;

[0016] Construct an adaptive crossover rate and mutation rate mechanism. The adaptive crossover rate is dynamically adjusted according to the population diversity index, and the mutation rate decays exponentially with the increase of the iteration times; select parent individuals based on the elitist retention strategy, generate offspring individuals through crossover operations, and perform mutation operations on the offspring individuals;

[0017] Use the non-dominated sorting method to stratify the population individuals, screen the Pareto front solution set based on the crowding distance, and use the optimal solution that satisfies the production efficiency-raw material utilization rate trade-off in the Pareto front solution set as the process parameter optimization scheme.

[0018] Preferably, the method for outputting the optimal production process parameters based on the multi-objective production optimization model includes:

[0019] Construct a multi-objective optimization function, which includes a production efficiency objective function and a raw material utilization rate objective function. Among them, the production efficiency objective function is calculated by the ratio of the reaction time to the product output, and the raw material utilization rate objective function is calculated by the ratio of the raw material consumption to the product quality;

[0020] Construct constraint conditions based on the multi-objective optimization function. The constraint conditions include temperature range constraints, pH range constraints, enzyme activity threshold constraints, and pressure safety threshold constraints;

[0021] The particle swarm optimization algorithm is used to encode the production process parameters. Each particle represents a set of process parameter combinations. The parameter values are adjusted based on the particle position update formula, where the inertia weight dynamic adjustment mechanism adjusts the weight value according to the convergence degree of the particles, and the neighborhood search strategy guides the population update through the local optimal particles;

[0022] A dynamic learning factor is introduced. The dynamic learning factor changes in a piecewise linear manner as the number of iterations increases, and the contribution weights of the global optimal particle and the local optimal particle to the velocity update are adjusted through the dynamic learning factor;

[0023] Based on the improved particle swarm optimization algorithm, iterative optimization is carried out. The production efficiency and raw material utilization rate of the parameter combinations generated in each iteration are evaluated, and the non-dominated solutions are stored in the optimization solution set. Finally, the parameter combination with the highest comprehensive score on the Pareto front is selected as the optimal production process parameters.

[0024] Preferably, the planning layer performs global resource allocation planning based on the process parameter optimization scheme, including:

[0025] The digital twin technology is used to construct a virtual production environment model. The virtual production environment model is driven by real-time process data and equipment status data to simulate the dynamic behavior of the reaction kettle and the material flow process;

[0026] Based on the mixed integer programming method, a resource allocation model is constructed. The resource allocation model aims to maximize the equipment utilization rate and minimize the energy consumption. The constraint conditions include the raw material supply cycle, equipment capacity limit, and production batch continuity;

[0027] The production task is decomposed into multiple subtasks, and a resource allocation matrix is generated based on the task priority and equipment load status. The resource allocation problem is solved through the Lagrangian relaxation algorithm to generate a global equipment scheduling scheme and a raw material feeding plan.

[0028] Preferably, the scheduling layer performs real-time process adjustment based on the optimal production process parameters, including:

[0029] The Bayesian network is used to construct a dynamic process adjustment model. The Bayesian network takes the real-time sensor data as the observation nodes and the process parameter adjustment amount as the decision nodes, and describes the dependency relationship between the parameters through the conditional probability table;

[0030] Based on the Markov decision process, a state transition model is constructed. The state space is defined as the current process parameters and equipment status, and the action space is the adjustment amounts of temperature, pH value, and pressure. The optimal adjustment strategy is solved through the value iteration algorithm;

[0031] A sliding time window mechanism is introduced to perform weighted average processing on the historical process data to eliminate instantaneous noise interference and generate a smoothed process parameter correction instruction.

[0032] Preferably, the precise control of the production equipment by the operation layer based on the fuzzy logic control algorithm includes:

[0033] Establish a multi-input multi-output fuzzy rule base for the production equipment. The input variables of the fuzzy rule base include temperature deviation, pH deviation, and pressure deviation, and the output variables of the fuzzy rule base include heating power, flow rate of acid-base pumps, and opening degree of pressure valves;

[0034] Perform fuzzy processing on the input and output variables using triangular membership functions, define the fuzzy rules in the form of IF-THEN, and obtain the precise control quantity through defuzzification by the centroid method;

[0035] Construct an anti-saturation compensation mechanism. When the control quantity exceeds the physical limit of the equipment, use a proportional-integral-derivative auxiliary controller for error compensation to ensure the stable output of the actuator.

[0036] Preferably, the above optimization of abalone peptide production process and intelligent decision-making method further includes:

[0037] Construct a quality prediction model based on the long short-term memory (LSTM) network. The quality prediction model takes historical process data and real-time sensor data as inputs and outputs the predicted values of the product peptide chain length distribution and bioactivity index;

[0038] Adopt an attention mechanism to enhance the feature weights of key process parameters, extract local temporal patterns through a time series sliding window, and combine the global process state to generate quality prediction results;

[0039] Feed back the quality prediction results to the scheduling layer, dynamically correct the process adjustment strategy, and achieve quality closed-loop control.

[0040] Preferably, the virtual production environment model is constructed through the following steps:

[0041] Use the finite element method to model the fluid dynamics in the reactor, solve the temperature field and concentration field distributions through the Navier-Stokes equation; simulate the collision and dissolution processes of raw material particles based on the discrete element method, and calculate the product generation rate in combination with the enzymatic reaction kinetic equation;

[0042] Update the model parameters through real-time data assimilation technology, and use the Kalman filter algorithm to fuse sensor data and simulation results.

[0043] Preferably, the training of the Bayesian network includes:

[0044] Use the expectation maximization (EM) algorithm to fill in the missing process data, and estimate the posterior distribution of the latent variables through the Gibbs sampling method;

[0045] Construct a network structure learning objective function based on KL divergence, adopt a greedy search strategy to optimize the node connection relationship, and generate a network topology with the minimum description length.

[0046] Evaluate the network uncertainty through the Monte Carlo dropout method and dynamically adjust the confidence threshold of the process adjustment strategy.

[0047] Preferably, the present invention further includes an abalone peptide production process optimization and intelligent decision-making system for implementing the above-mentioned method for optimizing the abalone peptide production process and intelligent decision-making, including:

[0048] A data acquisition and feature extraction module for collecting real-time process data through multi-source sensors, extracting process feature data based on a convolutional neural network, and generating a process parameter optimization scheme by optimizing the model through a genetic algorithm;

[0049] A production process optimization module for constructing a multi-objective production optimization model according to the process parameter optimization scheme and outputting the optimal production process parameters by using an improved particle swarm algorithm;

[0050] An intelligent decision-making execution module for realizing global resource allocation, real-time process adjustment and precise equipment control based on a hierarchical intelligent decision-making model, and outputting production control instructions.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] By using multi-source sensors to collect real-time process data in the abalone peptide production process and performing multi-scale feature extraction with the help of a convolutional neural network, key information can be accurately extracted from high-dimensional and multi-modal dynamic data. Compared with the problem that traditional data processing methods are easily interfered by redundant data, the present invention not only reduces the data storage and transmission burden, but also provides accurate and reliable data support for subsequent process optimization and decision-making, greatly improving the judgment accuracy of the production process.

[0053] The genetic algorithm optimization model adopts an adaptive crossover rate and mutation rate mechanism, combined with a multi-objective fitness function, which can comprehensively consider various factors such as production efficiency, raw material utilization rate, energy consumption and stability, and iteratively optimize the process parameter combination. Different from the traditional method of setting process parameters relying on manual experience and unable to comprehensively consider the interaction of complex factors, the present invention can generate a better process parameter optimization scheme, achieve the balance of various production indicators, significantly improve the production efficiency and raw material utilization rate, while reducing energy consumption and production costs.

[0054] Construct a multi-objective production optimization model aiming at maximizing production efficiency and raw material utilization rate, and use an improved particle swarm optimization algorithm for global optimization. This algorithm introduces an inertia weight dynamic adjustment mechanism, a neighborhood search strategy, and dynamic learning factors, etc., and can output the optimal production process parameters under various constraint conditions such as temperature, pH value, enzyme activity, and pressure. This changes the limitation of single-objective optimization in the existing technology, realizes the overall optimization of the production process, effectively avoids the problem of index imbalance caused by one-sided optimization, and further improves the production efficiency.

[0055] The planning layer of the hierarchical intelligent decision-making model uses digital twin technology and mixed integer programming methods to realize the reasonable allocation and planning of global resources, generate a scientific equipment scheduling plan and raw material feeding plan, and ensure the orderly progress of production; the scheduling layer is based on Bayesian networks and Markov decision-making processes, combined with a sliding time window mechanism, and can dynamically and accurately adjust process parameters according to real-time data to timely respond to uncertainties in the production process; the operation layer realizes the precise control of production equipment through a fuzzy logic control algorithm and constructs an anti-saturation compensation mechanism to ensure the stable operation of the equipment. In addition, the quality prediction model combines the attention mechanism and time series analysis, and feeds back the quality prediction results to the scheduling layer to form a quality closed-loop control. The entire intelligent decision-making and control system realizes the real-time and intelligent regulation of the production process, improves the stability of production and product quality, and has stronger real-time performance and foresight compared with the traditional decision-making method based on post-event analysis and empirical judgment. Brief Description of the Drawings

[0056] Figure 1 It is the working principle diagram of the abalone peptide production process optimization and intelligent decision-making method described in the present invention;

[0057] Figure 2 It is the working principle diagram of the genetic algorithm optimization model generating the process parameter optimization plan;

[0058] Figure 3 It is the working principle diagram of the multi-objective production optimization model outputting the optimal production process parameters. Detailed Embodiments

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0060] Please refer to Figures 1-3 , the present invention provides a technical solution: a method for optimizing the abalone peptide production process and intelligent decision-making, the method includes:

[0061] Using multi-source sensors such as temperature sensors, pH sensors, enzyme activity sensors, and pressure sensors, the process data during the production of abalone peptides is collected in real time, such as the reactor temperature, the pH value of the enzymatic hydrolysate, the enzyme activity concentration, the raw material input rate, and the reaction pressure. The collected real-time process data is input into a convolutional neural network, and through components such as the convolutional layer and pooling layer of the convolutional neural network, multi-scale feature extraction is performed on the data to obtain process feature data that can reflect the key information of the production process.

[0062] The extracted process feature data is input into a pre-trained genetic algorithm optimization model. This model adopts an adaptive crossover rate and mutation rate mechanism, dynamically adjusts the crossover rate according to the population diversity index, and makes the mutation rate decay exponentially with the increase of the iteration number. Based on a multi-objective fitness function that includes production efficiency terms, raw material utilization terms, energy consumption terms, and stability terms, the process parameter combinations are iteratively optimized. After multiple rounds of selection, crossover, and mutation operations, the non-dominated sorting method is used to stratify the population individuals, and the Pareto front solution set is screened based on the crowding distance. The optimal solution that satisfies the production efficiency-raw material utilization trade-off is selected as the process parameter optimization plan.

[0063] According to the generated process parameter optimization plan, a multi-objective production optimization model with the maximization of production efficiency and raw material utilization as the optimization objectives is constructed. An improved particle swarm algorithm is used to globally optimize the production process. This algorithm introduces an inertia weight dynamic adjustment mechanism and a neighborhood search strategy. By constructing a multi-objective optimization function that includes a production efficiency objective function and a raw material utilization objective function, and setting conditions such as temperature range constraints, pH range constraints, enzyme activity threshold constraints, and pressure safety threshold constraints, the production process parameters are encoded and iteratively optimized, and finally the optimal production process parameters are output.

[0064] A hierarchical intelligent decision-making model is established according to the optimal production process parameters. The model includes a planning layer, a scheduling layer, and an operation layer. The planning layer uses digital twin technology to construct a virtual production environment model, constructs a resource allocation model based on the mixed-integer programming method, decomposes the production tasks into multiple subtasks, generates a global equipment scheduling plan and a raw material feeding plan, and realizes the global resource allocation planning. The scheduling layer uses a Bayesian network to construct a dynamic process adjustment model, constructs a state transition model based on the Markov decision process, and introduces a sliding time window mechanism to realize real-time process adjustment. The operation layer establishes a multi-input multi-output fuzzy rule base for production equipment, uses a triangular membership function to fuzzify the input and output variables, obtains an accurate control quantity through the centroid method for defuzzification, and constructs an anti-saturation compensation mechanism to realize the accurate control of production equipment. Production control instructions are output through the hierarchical intelligent decision-making model to realize the optimization and intelligent decision-making of the abalone peptide production process. In addition, a quality prediction model is constructed based on the long short-term memory (LSTM) network, the attention mechanism is used to enhance the feature weights of key process parameters, and the quality prediction results are fed back to the scheduling layer to dynamically correct the process adjustment strategy and realize quality closed-loop control.

[0065] The present invention will be further described below in conjunction with Embodiments 1 to 5: Embodiment 1:

[0066] This embodiment details the specific process of generating a process parameter optimization plan using a genetic algorithm optimization model. By constructing a reasonable parameter space, decision space, and multi-objective fitness function, and adopting an adaptive crossover rate and mutation rate mechanism and an elite retention strategy, it is ensured that a process parameter optimization plan that meets production requirements can be obtained. Specifically, it includes:

[0067] (1) Constructing the parameter space and decision space: Real-time collection of the reactor temperature , the pH value of the enzymatic hydrolysate , the enzyme activity concentration , the raw material input rate , and the reaction pressure during the abalone peptide production process. These data constitute real-time process data. Based on these real-time process data, a parameter space is constructed. According to the adjustable temperature change , pH adjustment amount , enzyme addition amount , and pressure adjustment amount during the process, a decision space is constructed. For example, at a certain moment, the reactor temperature . According to production experience and process requirements, the adjustable temperature range is . Then the value range of the temperature change is [-5, 5]. The change ranges of other parameters are determined similarly to construct a complete decision space.

[0068] (2) Constructing a multi-objective fitness function:

[0069] Production efficiency item: Production efficiency item Product output per unit time With target output The ratio of is calculated as follows: For example, over a period of time The actual product output is 50 grams and the target output is 80 grams. The production efficiency item in this time period is .

[0070] Raw material utilization item: Raw material utilization item The actual product quality Theoretical maximum product mass The ratio of is calculated as follows: Assuming that the theoretical maximum product mass is 100 grams according to the composition of the raw materials and the chemical reaction formula, and the actual product mass is 70 grams, then the raw material utilization rate term .

[0071] Energy consumption items: Energy consumption items Adjust the amount by temperature , pH adjustment amount And pressure adjustment The weighted square sum of is calculated as follows: ,in , , is the weight coefficient, which is determined according to the degree of influence of each parameter on energy consumption in the production process. For example, after experiments and data analysis, it is determined , , , during an adjustment , , , then the energy consumption item .

[0072] Stability term: Stability term It is calculated by the Euclidean distance between the current process parameters and the historical optimal parameters. Assume that the current process parameter vector is , the historical optimal parameter vector is , then the stability term For example, the current process parameters are temperature , pH value is 7.2, and the temperature is the best parameter in history. , pH value is 7.0, then the stability term . Multi-objective fitness function Taking all the above into consideration, it can be expressed as ,in , , , are the weights of each item, which are determined according to the actual production requirements to balance the relationship between different objectives.

[0073] (3) Adaptive crossover rate and mutation rate mechanism and genetic operations:

[0074] Adaptive crossover rate: The adaptive crossover rate is dynamically adjusted according to the population diversity index. The population diversity index can be measured by calculating the degree of difference between individuals in the population. For example, methods such as the Hamming distance of individual coding can be used. When the population diversity is high, the crossover rate is appropriately reduced to avoid excessive destruction of excellent individuals; when the population diversity is low, the crossover rate is increased to promote the generation of new individuals. Assume that the population diversity index takes values between 0 and 1, the lower limit of the crossover rate is , and the upper limit is , then the calculation formula of the adaptive crossover rate can be . For example, , , when the population diversity index , .

[0075] Mutation rate: The mutation rate exponentially decays with the increase of the iteration number , and the formula is , where is the initial mutation rate is the decay coefficient, which is determined according to experiments. For example, , , when the iteration number , .

[0076] Genetic operations: Select parent individuals based on the elitist retention strategy, that is, retain a part of the individuals with higher fitness in the current population directly into the next generation. Then, perform crossover operations on the selected parent individuals according to the adaptive crossover rate to generate offspring individuals. For example, using the single-point crossover method, randomly select two parent individuals, randomly select a crossover point on their coding strings, and exchange the parts after the crossover point to generate two offspring individuals. Perform mutation operations on the offspring individuals according to the mutation rate. For example, for individuals with binary coding, take the mutation rate as the probability to invert each bit in the individual coding.

[0077] (4) Screening process parameter optimization scheme: Use the non-dominated sorting method to stratify the population individuals. Non-dominated sorting divides the individuals in the population into different levels according to the dominance relationship. If individual is not inferior to individual and is superior to the individual in at least one objective , then the individual is said to dominate the individual . The individuals that are not dominated by any other individual are classified into the first layer. After removing the first-layer individuals from the population, the above operations are repeated for the remaining individuals to obtain the second layer, and so on. The Pareto front solution set is screened based on the crowding distance, and the crowding distance is used to measure the degree of crowding of an individual in its corresponding layer. When calculating the crowding distance of each individual, first calculate the sum of the distances of the individual from its adjacent individuals in each objective dimension, and then normalize the sum of the distances of all objective dimensions. Select the individuals with larger crowding distances to form the Pareto front solution set because these individuals are more evenly distributed in the objective space and have better diversity. Select the optimal solution that satisfies the production efficiency - raw material utilization trade-off from the Pareto front solution set as the process parameter optimization scheme. For example, according to the actual production requirements, a comprehensive evaluation index of production efficiency and raw material utilization can be set, such as , and select the solution with the largest value in the Pareto front solution set as the final process parameter optimization scheme. Example 2:

[0078] This example details how to iteratively optimize the production process parameters based on a multi-objective production optimization model by constructing an optimization function, constraint conditions, using an improved particle swarm algorithm, combined with an inertia weight dynamic adjustment mechanism, a neighborhood search strategy, and a dynamic learning factor, and output the optimal production process parameters that can simultaneously maximize production efficiency and raw material utilization, providing accurate parameter guidance for actual production. The specific methods include:

[0079] ① Construct a multi-objective optimization function:

[0080] Production efficiency objective function: The production efficiency objective function is calculated through the ratio of the reaction time to the product yield , and the formula is . For example, in a production process, the reaction time is 5 hours and the product yield is 200 grams, then the value of the production efficiency objective function is 40 grams / hour.

[0081] Raw material utilization objective function: The raw material utilization objective function is calculated through the ratio of the raw material consumption to the product quality , and the formula is . Assuming that the mass of the consumed raw material is 300 grams and the mass of the obtained product is 200 grams, then the value of the raw material utilization objective function

[0082] ②Construct constraint conditions:

[0083] Temperature range constraint: According to the requirements of the abalone peptide production process, the temperature in the reaction kettle needs to be within a certain range. Let the lowest temperature be and the highest temperature be , then the temperature range constraint is . For example, if the production process requires the temperature to be between , then ,

[0084] pH range constraint: The enzymatic hydrolysis reaction has specific requirements for the pH value. Let the lower limit of the pH value of the enzymatic hydrolysis solution be and the upper limit be , then the pH range constraint is . For example, if the suitable pH value range is 6.5 - 7.5, then ,

[0085] Enzyme activity threshold constraint: The enzyme activity concentration needs to reach a certain threshold to ensure the normal progress of the reaction. Let the enzyme activity threshold be , then the enzyme activity threshold constraint is . Suppose the enzyme activity threshold is determined to be through experiments, then the enzyme activity concentration during the production process must be greater than or equal to this value.

[0086] Pressure safety threshold constraint: The reaction pressure cannot exceed the safe bearing range of the equipment. Let the pressure safety threshold be , then the pressure safety threshold constraint is . For example, if the pressure safety threshold of the equipment is 5 MPa, then the reaction pressure needs to satisfy .

[0087] ③Particle swarm algorithm encoding and parameter adjustment:

[0088] Encoding: Use the particle swarm algorithm to encode the production process parameters. Each particle represents a set of process parameter combinations, such as temperature, pH value, enzyme addition amount, raw material input rate, etc. Suppose the position vector of each particle is , where represents the th process parameter.

[0089] Particle position update formula: The particle position update formula is , where is the particle velocity, and the calculation formula is . is the inertia weight, and are the learning factors, and are random numbers between [0, 1]. is the historical best position of particle , is the global best position.

[0090] Inertia weight dynamic adjustment mechanism: The inertia weight adjusts the weight value according to the convergence degree of particles. When the convergence speed of particles is fast and approaching the local optimal solution, the inertia weight is reduced to make the particles pay more attention to local search; when the convergence speed of particles is slow, the inertia weight is increased to promote the global search of particles. Suppose the convergence degree of particles can be measured by the distance variance between particles. The lower limit of the inertia weight is , and the upper limit is . Then the inertia weight adjustment formula can be , where and are the pre-set maximum and minimum values of the distance variance. For example, , , , . When the calculated distance variance of the current particle is , .

[0091] Neighborhood search strategy: The neighborhood search strategy guides the population update through the local optimal particle. The entire population is divided into multiple neighborhoods, and each particle only exchanges information with other particles within the neighborhood. The local optimal particle is found in each neighborhood. When updating the particle velocity, in addition to considering the global optimal particle , the local optimal particle is also considered. That is, the velocity update formula becomes , where is the neighborhood search learning factor, is a random number between [0, 1].

[0092] Introducing dynamic learning factors: The dynamic learning factors and change in a piecewise linear manner with the increase of the iteration number . In the early stage of iteration, it is hoped that the particles are more inclined to global search, so the value of is increased and the value of is decreased; in the later stage of iteration, it is hoped that the particles pay more attention to local search, so the value of is increased and the value of is decreased. Assuming that the total number of iterations is , the iteration process can be divided into segment, and the length of each segment is . Taking the case of dividing into three segments as an example:

[0093] In the first segment when , t;

[0094] In the second segment when , ;

[0095] In the third segment when , .

[0096] Among them , , , are the minimum and maximum values of the learning factor set in advance. For example , , , .

[0097] Iterative optimization and optimal solution selection: Based on the improved particle swarm optimization algorithm for iterative optimization. In each iteration, the generated parameter combinations are used to calculate the production efficiency and raw material utilization rate according to the multi-objective optimization function. The non-dominated solutions that satisfy all constraint conditions and are not dominated by other solutions are stored in the optimization solution set. After multiple iterations, the parameter combination with the highest comprehensive score is selected as the optimal production process parameters on the Pareto front. The comprehensive score can assign different weights to the production efficiency and raw material utilization rate according to the actual production requirements. For example , among which , are the weights, which are determined according to the actual production situation. The parameter combination with the largest Score value is selected as the final optimal production process parameters. Example 3:

[0098] This example elaborates in detail how the planning layer conducts global resource allocation planning based on the process parameter optimization scheme. By constructing a virtual production environment model to simulate the production process, using the mixed integer programming method to construct a resource allocation model, decomposing the production tasks and generating a resource allocation matrix, and finally obtaining a global equipment scheduling plan and raw material feeding plan, realizing the reasonable allocation and efficient utilization of production resources and ensuring the smooth progress of the production process.

[0099] Constructing a virtual production environment model:

[0100] Modeling inside the reactor: Using the finite element method to model the fluid dynamics inside the reactor. According to the Navier - Stokes equation , where is the fluid density, is the flow velocity vector, is the time, is the pressure, is the dynamic viscosity, is the external force. By solving this equation, the temperature field and concentration field distributions in the reactor are obtained, and the flow and reaction conditions of the materials in the reactor are simulated. For example, after specifying the geometric shape of the reactor, the initial conditions of the materials, and the boundary conditions, the reactor is meshed using finite element software, the Navier - Stokes equation is discretized, and the distributions of the temperature field and concentration field at different times are obtained through numerical calculations.

[0101] Raw material particle simulation: The collision and dissolution processes of raw material particles are simulated based on the discrete element method. The raw material particles are regarded as discrete individuals, and the mutual forces between the particles, such as gravity, friction, collision force, etc., are considered to establish the particle motion equation. Combining with the enzymatic reaction kinetic equation, such as the Michaelis - Menten equation (where is the reaction rate, is the maximum reaction rate, is the substrate concentration, is the Michaelis constant), the product formation rate is calculated. By simulating the movement, dissolution, and reaction processes of raw material particles in the reactor, the material changes during the production process can be predicted more accurately.

[0102] Model parameter update: The model parameters are updated through real - time data assimilation technology. The Kalman filter algorithm is used to fuse sensor data and simulation results. Assume the state equation is , and the observation equation is , where is the state vector, is the state transition matrix, is the process noise, is the observation vector, is the observation matrix, is the observation noise. Using the Kalman filter algorithm, the prediction results of the model are continuously corrected according to the data collected by the sensors in real - time, making the virtual production environment model more consistent with the actual production situation.

[0103] Construct a resource allocation model: A resource allocation model is constructed based on the mixed - integer programming method, with the goal of maximizing equipment utilization and minimizing energy consumption. Let the equipment set be , the task set be , the utilization rate of equipment be , and the energy consumption be , then the objective function is .

[0104] The constraints include:

[0105] Raw material supply cycle constraint: Let the raw material supply cycle be , the amount of raw material supplied each time is , the consumption rate of raw material during the production process is , then it is necessary to satisfy , where is the number of supply times, is the raw material amount required for task .

[0106] Equipment capacity limit constraint: The capacity of equipment is , and the capacity occupied by task when processed on equipment is , then .

[0107] Production batch continuity constraint: If the production task needs to be carried out in batches, let the minimum interval time between adjacent batches be , then it is necessary to ensure that the production arrangements of adjacent batches meet the requirements of this time interval.

[0108] Generate a resource allocation plan: Decompose the production task into multiple subtasks, and generate a resource allocation matrix according to the task priority and equipment load status. The task priority can be determined according to factors such as order urgency and product profit, and the equipment load status can be measured by indicators such as the current working time and remaining capacity of the equipment. For example, for task and equipment , the element in the resource allocation matrix can be defined as: If task is allocated to equipment for production, then ; otherwise . Solve the resource allocation problem through the Lagrangian relaxation algorithm. This algorithm incorporates the constraint conditions into the objective function by introducing Lagrange multipliers, and transforms the original problem into a series of unconstrained subproblems for solution. Finally, generate a global equipment scheduling plan, determine the production task arrangements of each equipment at different times, and the raw material feeding plan, and clarify the raw material feeding amount and feeding time at each production stage. Example 4:

[0109] This example details how the scheduling layer performs real-time process adjustment based on the optimal production process parameters. By constructing a Bayesian network dynamic process adjustment model and a Markov decision process state transition model, combined with a sliding time window mechanism, it is possible to adjust the process parameters in a timely and accurate manner according to real-time sensor data, adapt to the uncertainties in the production process, and ensure the stability of the production process and the product quality.

[0110] Construct a Bayesian network dynamic process adjustment model

[0111] Network structure setting: Use a Bayesian network to construct a dynamic process adjustment model, with real-time sensor data, such as temperature sensor data , pH sensor data , enzyme activity sensor data etc. as observation nodes, and process parameter adjustment amounts, such as temperature adjustment amount , pH adjustment amount , pressure adjustment amount etc. as decision nodes. Describe the dependence relationship between parameters through a conditional probability table. For example, for the temperature adjustment amount and temperature sensor data , the conditional probability table can represent the probability of different temperature adjustment amounts under different temperature sensor measurement values.

[0112] Network training: Use the Expectation-Maximization (EM) algorithm to fill in the missing process data. Assume that there is some missing temperature sensor data. The EM algorithm iteratively calculates. In the E-step (expectation step), it estimates the expected value of the missing data based on the current model parameters. In the M-step (maximization step), it uses the estimated complete data to re-estimate the model parameters, and continuously iterates until convergence. Estimate the posterior distribution of latent variables through the Gibbs sampling method to handle the latent variables in the network. Construct a network structure learning objective function based on the KL divergence. The KL divergence , where is the true distribution, is the model distribution. Use a greedy search strategy to optimize the node connection relationship. Starting from an initial network structure, by continuously adding, deleting, or modifying edges, try to find the network topology that minimizes the KL divergence and generates the network topology with the minimum description length. Evaluate the network uncertainty through the Monte Carlo dropout method. Randomly discard some nodes during the training process, and obtain different network structures through multiple trainings. Evaluate the network uncertainty based on the differences in the prediction results of these network structures, and dynamically adjust the confidence threshold of the process adjustment strategy.

[0113] Construct a Markov decision process state transition model: Based on the Markov decision process, construct a state transition model. Define the state space as the current process parameters and equipment status. Let the state vector , where is the temperature, is the pH value, is the pressure, is the equipment status (such as whether the equipment is operating normally, the current load of the equipment, etc.). The action space is the adjustment amounts of temperature, pH value, and pressure, that is The optimal adjustment strategy is solved by the value iteration algorithm. The core formula of the value iteration algorithm is:

[0114]

[0115] where is the optimal value function of state , is the probability of transferring from state to state by executing action , is the reward obtained by transferring from state to state by executing action , and is the discount factor, with a value between 0 and 1, reflecting the degree of emphasis on future rewards. By continuously iterating and updating the value function, the optimal action in each state, that is, the optimal adjustment strategy, is finally obtained.

[0116] Sliding time window mechanism: The sliding time window mechanism is introduced to perform weighted average processing on historical process data. Let the size of the time window be , the current time be , the historical process data be , and the weight be ( , and the weight closer to the current time is larger, such as ), then the smoothed process parameter value . In this way, instantaneous noise interference is eliminated, and a smoothed process parameter correction instruction is generated to avoid frequent and unreasonable process adjustments caused by noise. For example, for temperature data, a more stable temperature correction value is obtained after weighted average processing by the sliding time window, providing more reliable data support for subsequent process adjustments. Example 5:

[0117] This example details how the operation layer realizes the precise control of production equipment based on the fuzzy logic control algorithm and how to achieve quality closed-loop control through the quality prediction model. By establishing a fuzzy rule base, performing fuzzification and defuzzification processing, and constructing an anti-saturation compensation mechanism, it is ensured that the production equipment can accurately execute control instructions. At the same time, the process is dynamically adjusted using the results fed back by the quality prediction model to improve the stability and consistency of product quality.

[0118] ① Equipment control based on the fuzzy logic control algorithm:

[0119] Establish a fuzzy rule base: Establish a multi-input multi-output fuzzy rule base for the production equipment. The input variables include the temperature deviation ( is the set temperature, is the actually measured temperature), pH deviation ( is the set pH value, is the actually measured pH value) and pressure deviation ([[]] is the set pressure, is the actually measured pressure), the output variables include heating power , the flow rate of acid-base pumps and the opening degree of the pressure valve .

[0120] Fuzzification: The input and output variables are fuzzified using triangular membership functions. Taking the temperature deviation as an example, assuming the temperature deviation range is [-5, 5], fuzzy sets such as "negative large", "negative small", "zero", "positive small", "positive large" are defined, and the vertex coordinates of the triangular membership functions are determined according to the actual situation. For example, the vertex coordinates of the triangular membership function of "negative large" can be (-5, -5, -3), and the vertex coordinates of "negative small" can be (-3, -1, 1), etc. The actual temperature deviation value is mapped to the corresponding fuzzy set through the membership function to obtain the fuzzified result.

[0121] Defuzzification and calculation of control quantity: The precise control quantity is obtained through defuzzification by the centroid method. For the output variable, such as heating power, assuming the multiple fuzzy output quantities obtained from fuzzy inference are , and their membership degrees are , then the precise control quantity . In this way, the fuzzy control decision is converted into an actual executable control signal to control the operation of the production equipment.

[0122] Anti-saturation compensation mechanism: An anti-saturation compensation mechanism is constructed. When the control quantity exceeds the physical limit of the equipment, a proportional-integral-derivative (PID) auxiliary controller is used for error compensation. The physical limit of is , when the calculated heating power is , the PID controller calculates the compensation quantity according to the error where ,

[0123] ②Quality prediction and feedback control:

[0124] Building a Quality Prediction Model: Build a quality prediction model based on the Long Short-Term Memory (LSTM) network, using historical process data such as temperature, pH value, enzyme activity, raw material input rate, etc. over a certain period of time, as well as real-time sensor data as inputs. The memory cells in the LSTM network can effectively capture long-term dependencies in time series data. Assuming the input data sequence is After being processed layer by layer through the LSTM network, the output is the distribution of the peptide chain length and the predicted values of the bioactivity indicators .

[0125] Application of Attention Mechanism: Use the attention mechanism to enhance the feature weights of key process parameters. During the data processing of the LSTM network, calculate the attention weights , , where is calculated through an attention scoring function, which can be based on the correlation between the input features and the hidden state of the network. For example: , where is the hidden state at the previous time step. Weight the input features with the attention weights to highlight the process parameter features that have a greater impact on the product quality. Then extract local temporal patterns through a time series sliding window. Assume the size of the sliding window is, at each time step, take data from the current time step backwards as a window of data, analyze the change trends and features of the data within the window, and combine with the global process state, such as the current equipment operating state, overall production progress, etc., to generate more accurate quality prediction results.

[0126] Quality Closed-Loop Control: Feed back the quality prediction results to the scheduling layer to dynamically correct the process adjustment strategy. If the quality prediction model predicts that the distribution of the peptide chain length or the bioactivity indicators of the product deviate from the expected target, the scheduling layer re-evaluates the current process parameters based on the feedback information, and combines with the Bayesian network dynamic process adjustment model and the Markov decision process state transition model to adjust the adjustment amounts of process parameters such as temperature, pH value, pressure, etc., so that the production process moves in a direction conducive to improving product quality. For example, if it is predicted that the bioactivity indicator of the product is low and it is found through analysis that it is related to the low current reaction temperature, the scheduling layer will appropriately increase the temperature adjustment amount, thereby achieving quality closed-loop control, continuously optimizing the production process, and ensuring stable product quality.

[0127] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0128] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An abalone peptide production process optimization and intelligent decision-making method, characterized in that: The method comprises: Collecting real-time process data in the abalone peptide production process through multi-source sensors, wherein the multi-source sensors include a temperature sensor, a pH sensor, an enzyme activity sensor, and a pressure sensor; performing multi-scale feature extraction on the real-time process data based on a convolutional neural network to obtain process feature data; The process characteristic data is input into a pre-trained genetic algorithm optimization model, wherein the genetic algorithm optimization model adopts an adaptive crossover rate and mutation rate mechanism, iteratively optimizes the process parameter combination based on a multi-objective fitness function, and generates a process parameter optimization plan; Constructing a multi-objective production optimization model according to the process parameter optimization scheme, wherein the multi-objective production optimization model takes maximizing production efficiency and maximizing raw material utilization as optimization objectives, and adopts an improved particle swarm algorithm to globally optimize the production process, wherein the improved particle swarm algorithm introduces an inertia weight dynamic adjustment mechanism and a neighborhood search strategy; outputting optimal production process parameters based on the multi-objective production optimization model; A hierarchical intelligent decision-making model is established according to the optimal production process parameters, and the hierarchical intelligent decision-making model includes a planning layer, a scheduling layer and an operation layer, wherein the planning layer performs global resource allocation planning based on the process parameter optimization plan, the scheduling layer performs dynamic parameter scheduling based on the optimal production process parameters, and the operation layer realizes the execution control of the production equipment based on the fuzzy logic control algorithm; the production control instructions are output through the hierarchical intelligent decision-making model to realize the optimization and intelligent decision-making of the abalone peptide production process.

2. The method for optimizing the production process and intelligent decision-making of abalone peptide according to claim 1, characterized in that: The process characteristic data is input into a pre-trained genetic algorithm optimization model. The genetic algorithm optimization model adopts an adaptive crossover rate and mutation rate mechanism to iteratively optimize the process parameter combination based on a multi-objective fitness function to generate a process parameter optimization solution including: Acquire real-time process data, wherein the real-time process data includes reactor temperature, enzymatic solution pH value, enzyme activity concentration, raw material input rate and reaction pressure; construct a parameter space based on the real-time process data, and construct a decision space based on the process adjustable temperature change, pH adjustment amount, enzyme addition amount and pressure adjustment amount; A multi-objective fitness function is constructed based on the parameter space and the decision space, wherein the multi-objective fitness function includes a production efficiency item, a raw material utilization item, an energy consumption item, and a stability item, wherein the production efficiency item is calculated by the ratio of the product output per unit time to the target output, the raw material utilization item is calculated by the ratio of the actual product quality to the theoretical maximum product quality, the energy consumption item is calculated by the weighted square sum of the temperature adjustment amount, the pH adjustment amount, and the pressure adjustment amount, and the stability item is calculated by the Euclidean distance between the current process parameter and the historical optimal parameter; Construct an adaptive crossover rate and mutation rate mechanism, wherein the adaptive crossover rate is dynamically adjusted with the population diversity index, and the mutation rate changes exponentially with the increase in the number of iterations; select parent individuals based on the elite retention strategy, generate offspring individuals through crossover operations, and perform mutation operations on the offspring individuals; The non-dominated sorting method is used to stratify the population individuals, and the Pareto frontier solution set is screened based on the crowding distance. The optimal solution in the Pareto frontier solution set that satisfies the trade-off between production efficiency and raw material utilization is taken as the process parameter optimization solution.

3. The abalone peptide production process optimization and intelligent decision-making method according to claim 1, characterized in that: The method for outputting optimal production process parameters based on the multi-objective production optimization model includes: Constructing a multi-objective optimization function, wherein the multi-objective optimization function includes a production efficiency objective function and a raw material utilization objective function, wherein the production efficiency objective function is calculated by the ratio of reaction time to product output, and the raw material utilization objective function is calculated by the ratio of raw material consumption to product quality; Constructing constraint conditions based on the multi-objective optimization function, wherein the constraint conditions include temperature range constraint, pH range constraint, enzyme activity threshold constraint and pressure safety threshold constraint; The particle swarm algorithm is used to encode the production process parameters. Each particle represents a set of process parameter combinations. The parameter values ​​are adjusted based on the particle position update formula. The inertia weight dynamic adjustment mechanism adjusts the weight value according to the particle convergence degree. The neighborhood search strategy guides the population update through the local optimal particle. A dynamic learning factor is introduced, and the dynamic learning factor changes piecewise linearly with the increase of the number of iterations, and the contribution weights of the global optimal particle and the local optimal particle to the speed update are adjusted by the dynamic learning factor; Based on the improved particle swarm algorithm, iterative optimization is performed, the production efficiency and raw material utilization rate of the parameter combination generated in each iteration are evaluated, the non-dominated solution is stored in the optimization solution set, and finally the parameter combination with the highest comprehensive score in the Pareto front is selected as the optimal production process parameter.

4. The method for optimizing the production process and intelligent decision-making of abalone peptide according to claim 1, characterized in that: The planning layer performs global resource allocation planning based on the process parameter optimization solution, including: A virtual production environment model is constructed using digital twin technology. The virtual production environment model is driven by real-time process data and equipment status data to simulate the dynamic behavior of the reactor and the material flow process. A resource allocation model is constructed based on a mixed integer programming method, wherein the resource allocation model aims to maximize equipment utilization and minimize energy consumption, and the constraints include raw material supply cycle, equipment capacity limit, and production batch continuity; The production task is decomposed into multiple subtasks, and the resource allocation matrix is ​​generated based on the task priority and equipment load status. The resource allocation problem is solved through the Lagrangian relaxation algorithm to generate a global equipment scheduling plan and raw material delivery plan.

5. The abalone peptide production process optimization and intelligent decision-making method according to claim 1, characterized in that: The scheduling layer performs real-time process adjustments based on the optimal production process parameters, including: A dynamic process adjustment model is constructed using a Bayesian network, which uses real-time sensor data as observation nodes, process parameter adjustment amounts as decision nodes, and describes the dependency relationship between parameters through a conditional probability table; A state transition model is constructed based on the Markov decision process. The state space is defined as the current process parameters and equipment status, and the action space is defined as the adjustment amount of temperature, pH value and pressure. The optimal adjustment strategy is solved through the value iteration algorithm. A sliding time window mechanism is introduced to perform weighted average processing on historical process data, eliminate instantaneous noise interference, and generate smoothed process parameter correction instructions.

6. The abalone peptide production process optimization and intelligent decision-making method according to claim 1, characterized in that: The operation layer realizes precise control of production equipment based on fuzzy logic control algorithm, including: Establishing a multi-input multi-output fuzzy rule base for production equipment, wherein the input variables of the fuzzy rule base include temperature deviation, pH deviation and pressure deviation, and the output variables of the fuzzy rule base include heating power, acid-base pump flow and pressure valve opening; The input and output variables are fuzzified using triangular membership functions, and the fuzzy rules are defined as IF-THEN forms. The precise control quantity is obtained by defuzzification using the centroid method. An anti-saturation compensation mechanism is constructed. When the control quantity exceeds the physical limit of the equipment, a proportional-integral-differential auxiliary controller is used to compensate for the error to ensure the stable output of the actuator.

7. The abalone peptide production process optimization and intelligent decision-making method according to claim 1, characterized in that: Also includes: A quality prediction model is constructed based on a long short-term memory (LSTM) network. The quality prediction model takes historical process data and real-time sensor data as input and outputs the distribution of product peptide chain length and the predicted value of biological activity index. The attention mechanism is used to enhance the feature weights of key process parameters, local time series patterns are extracted through time series sliding windows, and quality prediction results are generated in combination with global process status; The quality prediction results are fed back to the scheduling layer to dynamically correct the process adjustment strategy to achieve closed-loop quality control.

8. The method for optimizing the production process and intelligent decision-making of abalone peptide according to claim 4, characterized in that: The virtual production environment model is constructed by the following steps: The finite element method was used to model the fluid dynamics in the reactor, and the temperature field and concentration field distribution were solved by the Navier-Stokes equation. The collision and dissolution process of the raw material particles was simulated based on the discrete element method, and the product generation rate was calculated in combination with the enzymatic reaction kinetic equation. The model parameters are updated through real-time data assimilation technology, and the Kalman filter algorithm is used to fuse sensor data and simulation results.

9. The method for optimizing the production process and intelligent decision-making of abalone peptide according to claim 5, characterized in that: The training of the Bayesian network includes: The expectation maximization (EM) algorithm is used to fill in the missing process data, and the posterior distribution of latent variables is estimated by the Gibbs sampling method; Based on KL divergence, the network structure learning objective function is constructed, and the node connection relationship is optimized using a greedy search strategy to generate a network topology with the minimum description length. The network uncertainty is evaluated through the Monte Carlo dropout method, and the confidence threshold of the process adjustment strategy is dynamically adjusted.

10. An abalone peptide production process optimization and intelligent decision-making system, used to implement the method according to any one of claims 1 to 9, characterized in that: include: Data acquisition and feature extraction module, which is used to collect real-time process data through multi-source sensors, extract process feature data based on convolutional neural network, and generate process parameter optimization scheme through genetic algorithm optimization model; A production process optimization module is used to construct a multi-objective production optimization model according to the process parameter optimization scheme, and output the optimal production process parameters using an improved particle swarm algorithm; The intelligent decision-making execution module is used to realize global resource allocation, real-time process adjustment and precise equipment control based on the hierarchical intelligent decision-making model, and output production control instructions.

Images