A control method, system, equipment, and medium for plant extraction equipment.

Abstract

This application provides a control method, system, device, and medium for plant extraction equipment, belonging to the field of equipment control technology. The method includes: determining an initial particle swarm based on a time-series parameter vector; calculating the fitness value of each particle in the initial particle swarm, taking the initial position of each particle as its historical best position, and taking the position of the particle with the highest fitness value as its global best position; using the global best position and the historical best position of each particle as an initial optimal parameter combination to update each particle in the initial particle swarm, obtaining a new particle swarm, until an iteration condition is met; taking the particle swarm obtained from the last update as the target particle swarm, updating the initial optimal parameter combination based on the target particle swarm to obtain a target optimal parameter combination; and using the target optimal parameter combination to control the plant extraction equipment. This application can improve the control level of plant extraction equipment, thereby enhancing product quality stability.

Description

Technical Field

[0001] This application belongs to the field of equipment control technology, and more specifically, relates to a control method, system, equipment and medium for plant extraction equipment. Background Technology

[0002] In the field of plant extraction, the control level of plant extraction equipment directly affects the yield, purity, and production safety of target plant components. Currently, the plant extraction industry commonly uses proportional-integral-derivative (PID) control. However, this traditional method has significant drawbacks. Existing PID control technology ignores the interrelationships between parameters during the extraction process and cannot handle the coupling relationships between multiple parameters, leading to unpredictable extraction results and difficulty in meeting the requirements for product quality stability. Summary of the Invention

[0003] The purpose of this application is to provide a control method, system, device and medium for plant extraction equipment, so as to improve the control level of plant extraction equipment, reduce abnormal fluctuations in extraction effect, and thus improve product quality stability.

[0004] A first aspect of this application provides a control method for a plant extraction device, comprising:

[0005] A time-series parameter vector is obtained by splicing the control parameters corresponding to each extraction stage in the plant history extraction process, and an initial particle swarm is determined based on the time-series parameter vector; the control parameters include the temperature, pressure and stirring speed in the plant extraction tank of the plant extraction equipment;

[0006] The fitness value of each particle in the initial particle swarm is calculated using the fitness function. The initial position of each particle is taken as the historical best position of that particle, and the position of the particle with the largest fitness value in the initial particle swarm is taken as the global best position.

[0007] The global optimal position and the historical optimal position of each particle are used as the initial optimal parameter combination. Each particle in the initial particle swarm is updated based on the particle update formula to obtain a new particle swarm, until the iteration condition is met. The particle swarm obtained from the last update is used as the target particle swarm. The initial optimal parameter combination is updated based on the target particle swarm to obtain the target optimal parameter combination.

[0008] The plant extraction equipment is controlled using the target optimal parameter combination.

[0009] A second aspect of this application provides a control system for a plant extraction device, comprising:

[0010] The particle swarm construction module is used to splice together the control parameters corresponding to each extraction stage in the plant history extraction process to obtain a time-series parameter vector, and to determine the initial particle swarm based on the time-series parameter vector; the control parameters include the temperature, pressure and stirring speed in the plant extraction tank of the plant extraction equipment.

[0011] The fitness calculation module is used to calculate the fitness value of each particle in the initial particle swarm using the fitness function, take the initial position of each particle as the historical best position of that particle, and take the position of the particle with the largest fitness value in the initial particle swarm as the global best position.

[0012] The parameter optimization module is used to take the global optimal position and the historical optimal position of each particle as the initial optimal parameter combination, update each particle in the initial particle swarm based on the particle update formula to obtain a new particle swarm, until the iteration condition is met; take the particle swarm obtained from the last update as the target particle swarm, and update the initial optimal parameter combination based on the target particle swarm to obtain the target optimal parameter combination;

[0013] The equipment control module is used to control the plant extraction equipment using the optimal combination of target parameters.

[0014] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the control method for a plant extraction device described above.

[0015] In a fourth aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the control method for a plant extraction device described above.

[0016] The beneficial effects of the control method, system, equipment, and medium for plant extraction equipment provided in this application are as follows: This application first splices control parameters from each stage of historical plant extraction to form a time-series parameter vector, thereby determining the initial particle swarm. This fully considers the dynamic changes and mutual influences of multiple parameters such as temperature, pressure, and stirring speed during the extraction process. This application determines the initial optimal parameter combination through a fitness function and continuously iterates and updates the particle swarm to ultimately obtain the target optimal parameter combination for equipment control. This process effectively uncovers the optimal synergistic relationship between multiple parameters, precisely regulates equipment operation, and ensures that the plant extraction process is carried out under the optimal parameter combination. This improves the yield and purity of the target plant components while ensuring production safety, significantly enhancing product quality stability, and meeting the industry's demand for high-quality plant extracts. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic flowchart illustrating a control method for a plant extraction device according to an embodiment of this application;

[0019] Figure 2 A structural block diagram of a control system for a plant extraction device provided in one embodiment of this application;

[0020] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0023] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a control method for a plant extraction device according to an embodiment of this application. The method can be executed by an electronic device, and specifically, the method may include S101 to S104.

[0024] S101: The control parameters corresponding to each extraction stage in the plant history extraction process are spliced ​​to obtain a time-series parameter vector, and the initial particle swarm is determined based on the time-series parameter vector; the control parameters include the temperature, pressure and stirring speed in the plant extraction tank of the plant extraction equipment.

[0025] In this embodiment, the plant history extraction process refers to multiple plant component extraction tasks performed by the plant extraction equipment in the past, including complete time series data, which records the changes in temperature, pressure, and stirring speed in the extraction tank over time, as well as the detection results of the target component concentration.

[0026] The extraction stage refers to a continuous period of time defined based on extraction kinetics, such as the immersion stage, rapid mass transfer stage, and equilibrium stage. Within each extraction stage, the influence of control parameters on extraction efficiency is relatively stable, and the division of extraction stages can be determined by the abrupt change in the slope of the concentration-time curve.

[0027] Control parameters refer to the adjustable parameters inside the plant extraction tank of plant extraction equipment that directly affect the extraction effect. Among them, temperature affects the solubility of components and the rate of molecular diffusion. High temperature accelerates extraction but can easily damage heat-sensitive components. High pressure promotes solvent penetration into the cell wall. The pressure value needs to be coupled with temperature. If the temperature rises, the pressure needs to rise synchronously to avoid solvent boiling. Appropriate stirring speed can improve the uniformity of solid-liquid mixing. Excessive stirring speed will cause mechanical shearing and damage the raw material structure. Insufficient stirring speed will affect the plant extraction efficiency.

[0028] The time-series parameter vector refers to a one-dimensional vector formed by concatenating the control parameters of each extraction stage in chronological order. The data structure can be [...]. ], where j is the total number of extraction stages, This refers to the temperature value during the first extraction stage. This represents the pressure value for the first extraction stage. This represents the stirring speed value for the first extraction stage, with the subscript of the letter indicating the index of the extraction stage.

[0029] The initial particle swarm refers to the initial solution set of the Particle Swarm Optimization (PSO) algorithm. The initial particle swarm contains multiple initialized particles, each representing a combination of control parameters, namely the temperature, pressure, and stirring speed values ​​corresponding to all extraction stages.

[0030] In this embodiment, the optimal parameter combination for each extraction stage can be obtained as prior knowledge by analyzing historical extraction data of the plant's extraction process, i.e., a time-series parameter vector. Random particles are then introduced to expand the search range. In this embodiment, the control parameters are discretized into staged vectors, with each particle representing a complete extraction process route. The particle swarm iteration process optimizes the control parameters for each extraction stage of this route to maximize the fitness function.

[0031] S102: Calculate the fitness value of each particle in the initial particle swarm using the fitness function, take the initial position of each particle as the historical best position of that particle, and take the position of the particle with the largest fitness value in the initial particle swarm as the global best position.

[0032] In this embodiment, the fitness function is:

[0033] ;

[0034] in, ;

[0035] in, Each particle represents a control parameter value corresponding to all extraction stages. Let X be the fitness value of particle X. Here, j represents the index of the extraction stage, and j represents the total number of extraction stages. For the first Weighting coefficients for each extraction stage For the first The local fitness function of the stage. For the first Particles in each extraction stage The corresponding fitness value, Here, g is the constraint penalty coefficient, and g() is the dynamic constraint function for adjacent extraction stages. For the first Control parameters for each extraction stage For the first -1 control parameters for the extraction stage, where Y() is the extraction rate function for the plant target components. To preset the target extraction rate, For plant extraction purity function, To achieve the preset target purity, Let the energy consumption function be the energy consumption function of the plant extraction equipment. For plant extraction time function, , , and These are the weighting coefficients. .

[0036] In this embodiment, Used to reflect the The importance of each stage can be adjusted according to the process objectives. Weighting coefficients. , , and It can be used to prioritize extraction rate, purity, energy consumption, and extraction time. (Constraint penalty coefficient) It can be used to control the penalty for sudden changes in parameters between adjacent stages. The larger the value, the smoother the transition of control parameters between different extraction stages. The fitness function should not be too large, as excessive constraints can lead to optimization stagnation. The purpose of the fitness function is to quantify the merits of each combination of control parameters; therefore, all parameters in the fitness function are numerical and dimensionless.

[0037] In this embodiment, the target component extraction rate refers to the ratio of the amount of the target component actually extracted from the raw material to the total content of that component in the raw material during the plant extraction process. The target component may include effective medicinal ingredients, active substances, etc. The target component extraction rate is a core indicator for measuring the efficiency of the extraction process, directly reflecting the enrichment capacity of the extraction process for the target component.

[0038] For example, the target component extraction rate function can be:

[0039]

[0040] in, , These are the fitting coefficients obtained through regression analysis of historical data. for Temperature parameters in for The pressure parameter S is... The stirring speed parameter in the text.

[0041] The fitting coefficients in the target component extraction rate function determined based on historical extraction data , This method quantifies the effects of temperature, pressure, and stirring speed on the extraction rate of the target component. In plant extraction, increased temperature accelerates molecular motion and increases component solubility, pressure promotes solvent penetration, and stirring enhances mass transfer. These three factors interact to affect extraction efficiency. The coefficients obtained through regression analysis of historical data in this embodiment reflect the contribution of each parameter under specific plant and extraction processes, thereby estimating the extraction rate of the target component under different parameter combinations and providing a basis for optimizing the extraction process. The purpose of the target component extraction rate function is to quantitatively estimate the extraction effect of each combination of control parameters; therefore, all parameters in the target component extraction rate function are numerical values ​​and dimensionless.

[0042] For example, the plant extract purity function can be:

[0043]

[0044] in, This is the impurity dissolution coefficient. This is used to simulate the relationship between high temperature and high stirring promoting the dissolution of impurities. Let T be a fixed constant, and T be the particle size. The temperature parameter in the text, S represents the particle... The stirring speed parameter in the text.

[0045] Plant extract purity functions primarily measure extraction purity by simulating the relationship between high temperature, high stirring, and impurity dissolution. Under conditions of high temperature and high stirring, impurities in the raw material are more easily dissolved, thus reducing extraction purity. Plant extract purity functions are based on... The degree of impurity dissolution is quantified by this value. A higher value indicates a higher proportion of impurities relative to the target component in the extract, resulting in lower purity. Therefore, its reciprocal is used as the method for calculating purity, intuitively reflecting the impact of each parameter on extraction purity and helping to control the extraction process to improve product purity. The purpose of the plant extract purity function is to quantitatively estimate the extraction purity of each combination of control parameters; therefore, all parameters in the plant extract purity function are numerical values ​​and dimensionless.

[0046] For example, the energy consumption function of a plant extraction device can be:

[0047]

[0048] in, The preset standard temperature is such as room temperature. The heating / cooling power consumption is required for the part of the temperature that exceeds the preset standard temperature. For the first The stirring time for each extraction stage. The energy consumption function of the plant extraction equipment divides energy consumption into two parts. This indicates the energy consumption for heating or cooling due to temperature differences. The greater the difference between the temperature and the preset standard temperature, and the higher the pressure, the higher this energy consumption. The energy consumption for stirring is determined by the stirring speed and time; the function uses 10 units as a speed unit. The energy consumption function for plant extraction equipment comprehensively considers the main energy-consuming factors during equipment operation, helping to assess energy consumption under different extraction parameters and thus optimize the process to reduce production costs. The purpose of the energy consumption function for plant extraction equipment is to quantitatively estimate the extraction energy consumption for each combination of control parameters; therefore, all parameters in the energy consumption function for plant extraction equipment are numerical values ​​and dimensionless.

[0049] In this embodiment, the fitness function refers to the core evaluation function of multi-objective optimization, which is used to quantify the quality of particles. The input of the fitness function is particle X, which contains the control parameters of all stages. The output of the fitness function is the comprehensive score. The higher the score, the better the process. For example, particles with high extraction rate, high purity, low energy consumption, short time and smooth parameter transition.

[0050] In this embodiment, the local fitness function is an evaluation function for a single extraction stage, primarily used to calculate the extraction efficiency, purity, energy consumption, and extraction time for that stage. The contribution of each stage is determined by weighted coefficients. Adjustments, such as assigning higher weighting coefficients to the rapid extraction stage, can be made.

[0051] In this embodiment, the dynamic constraint function is used to measure the degree of abrupt change in control parameters between adjacent stages, such as step changes in temperature and pressure, to avoid drastic fluctuations in equipment operating parameters, such as a sudden rise in temperature that may lead to damage to the heat-sensitive components or overload of the equipment.

[0052] In this embodiment, the core objective of the multi-objective optimization of the fitness function is to transform the four conflicting objectives of plant extraction rate, extraction purity, energy consumption, and extraction time into a single-objective optimization problem through weighted summation. Extraction rate and purity are positive indicators; higher values ​​result in higher scores. Energy consumption and extraction time are negative indicators; lower values ​​result in higher scores. When α > β, the algorithm prioritizes optimizing the extraction rate; when γ > α, the algorithm prioritizes optimizing energy consumption.

[0053] In this embodiment, the core objectives differ at different extraction stages; for example, rapid infiltration is required in the initial stage. The focus is on time; high purity is required in the mid-to-late stages. Focusing on purity, this embodiment can be achieved through... Dynamically adjust contributions at each stage to avoid a one-size-fits-all approach to optimization.

[0054] In actual production, sudden changes in parameters between adjacent stages can lead to equipment malfunctions and fluctuations in extraction efficiency. Therefore, by... Punish parameter abrupt changes and force a smooth transition of process parameters.

[0055] For example, the dynamic constraint function can be:

[0056]

[0057] in, for The penalty value for changes in the value between intervals. For the first The extraction stage and the first Temperature difference in each extraction stage For the first The extraction stage and the first Pressure difference in each extraction stage For the first The extraction stage and the first The difference in stirring speed between extraction stages. The purpose of the dynamic constraint function is to quantify the degree of abrupt change between combinations of control parameters in adjacent extraction stages; therefore, all parameters in the dynamic constraint function are only numerical values ​​and are dimensionless.

[0058] S103: Using the global optimal position and the historical optimal position of each particle as the initial optimal parameter combination, update each particle in the initial particle swarm based on the particle update formula to obtain a new particle swarm, until the iteration condition is met; use the particle swarm obtained from the last update as the target particle swarm, and update the initial optimal parameter combination based on the target particle swarm to obtain the target optimal parameter combination.

[0059] In this embodiment, the global optimal position refers to the position of the particle with the highest fitness value in the initial particle swarm. It represents the optimal combination of extraction process parameters among all particles, including temperature, pressure, and stirring speed at each extraction stage, and is the optimal experience shared by the group. The historical optimal position refers to the optimal position that each particle has reached during the iteration process. It is the initial position of the particle and is used to record the individual optimal experience of that particle.

[0060] In this embodiment, the target optimal parameter combination is the final solution determined by the particle swarm optimization algorithm after iterative optimization. It is the sequence of control parameters that satisfies all constraints and has the highest fitness value, and is directly used to guide the actual operation of the plant extraction equipment.

[0061] In this embodiment, the particle update formula is used to guide particles to search for better positions in the solution space by integrating individual historical bests, global bests, and adjacent stage parameter constraints. Iteration conditions refer to the rules for terminating particle swarm iteration, which may include the maximum number of iterations or fitness convergence accuracy, such as the global optimal solution remaining unchanged for N consecutive iterations.

[0062] For example, each particle can adjust its search direction through its own historical best and the swarm's global best. Its own historical best allows the particle to return to the parameter region where it previously performed well, avoiding blind searching; the swarm's global best allows the particle to move closer to the current optimal solution, accelerating swarm convergence. The particle with the highest fitness from the particle swarm of the last iteration is selected as the optimal parameter sequence to guide device control, ensuring that the parameters at each stage both optimize the extraction effect and meet engineering constraints.

[0063] For example, a company uses plant extraction equipment to extract chlorogenic acid from honeysuckle. First, historical data from multiple batches of honeysuckle extraction are collected to determine the initial particle swarm. In the initial particle swarm, the position of the particle with the highest fitness value is found as the global optimum, and the initial position of each particle is set as its historical optimum.

[0064] Then, the particles are updated based on the particle update formula. For example, after multiple iterations, when the maximum number of iterations (200) is reached, the target particle swarm is obtained. From the target particle swarm, the particle with the highest fitness is selected, and its parameters such as temperature, pressure, and stirring speed at each stage are considered to be the optimal combination of parameters.

[0065] In actual production, the optimal combination of parameters is applied to the honeysuckle extraction equipment. For example, in the initial stage of extraction, the extraction tank temperature is set to 55℃, the pressure to 1.2MPa, and the stirring speed to 150rpm; in subsequent stages, the parameters are precisely adjusted based on the optimized parameters.

[0066] S104: Control the plant extraction equipment using the target optimal parameter combination.

[0067] For example, the optimal combination of parameters is input into the control system of the plant extraction equipment. The system analyzes the parameters for each stage and sets the initial states of the temperature control device, pressure regulation device, and stirring device based on the parameters of the first stage. During the extraction process, sensors collect data once per second, and the control system compares the actual values ​​with the target values. If the deviation exceeds ±2%, the corresponding device is adjusted. When the set time or other switching conditions for the current stage are reached, the control system automatically adjusts the equipment's operating state according to the parameters for the next stage. After extraction is completed, the control system records the actual operating data for subsequent analysis and optimization, and simultaneously shuts down all components of the equipment.

[0068] For example, taking the extraction of ginsenosides from ginseng by a pharmaceutical factory as an example, the factory analyzed and processed extraction data from multiple batches of ginseng and used particle swarm optimization to obtain the optimal combination of parameters. In actual production, the factory controlled the plant extraction equipment according to this combination. In the initial stage of extraction, the equipment precisely controlled the temperature at 60℃, the pressure at 1.5MPa, and the stirring speed at 180rpm to promote full infiltration of ginseng cells. During the rapid extraction stage, the temperature was increased to 70℃, the pressure to 2.0MPa, and the stirring speed to 220rpm to accelerate the dissolution of ginsenosides. In the equilibrium stage, the temperature was reduced to 55℃, the pressure was adjusted back to 1.2MPa, and the stirring speed was reduced to 120rpm to make the extraction process more stable.

[0069] As can be seen from the above, this embodiment determines the initial particle swarm by analyzing historical data and uses a fitness function to comprehensively consider factors such as extraction rate, purity, energy consumption, and time. Through iterative optimization of the particle swarm, the optimal parameter combination is obtained, which can significantly improve extraction efficiency and product quality. The dynamic constraint function penalizes abrupt changes in parameters between adjacent stages, forcing a smooth transition of process parameters, effectively avoiding drastic fluctuations in equipment operating parameters, reducing the risk of equipment failure, and extending equipment lifespan. The entire control process is highly automated, from parameter analysis and equipment status setting to real-time monitoring and adjustment, and then to data recording, analysis, and optimization, achieving intelligent production, reducing manual intervention, lowering labor costs, and ensuring the accuracy and stability of the production process.

[0070] In one embodiment of this application, for each historical extraction process of a plant, before splicing the control parameters corresponding to each extraction stage in the historical extraction process of the plant to obtain the time-series parameter vector, the method further includes: based on the dynamic curve of the target component concentration changing with time in the historical extraction process of the target plant, using a clustering algorithm to divide the plant extraction process into multiple extraction stages;

[0071] The target plant is the plant from which the target component is extracted using plant extraction equipment.

[0072] In this embodiment, the target plant refers to a plant species from which specific target components are extracted using plant extraction equipment, such as ginkgo for extracting flavonoids and yew for extracting paclitaxel. Different target plants have different characteristics, and therefore require different extraction parameters.

[0073] A kinetic curve depicting the concentration of a target component over time is a curve plotted with time on the x-axis and the concentration change of the target component in the target plant during the extraction process on the y-axis. This kinetic curve reflects the dynamic change in the concentration of the target component during extraction, and its parameters include time and the concentration of the target component. The slope of the curve reflects the change in the extraction rate.

[0074] In plant extraction, the concentration change patterns of target components vary at different stages due to differences in their dissolution and diffusion mechanisms. Clustering algorithms can identify these patterns, grouping stages with similar patterns together, thus rationally dividing the extraction process and providing a basis for determining the optimal control parameters for each stage. For example, in this embodiment, clustering algorithms such as K-Means can be used to group the extraction process based on the concentration change patterns of the target components, resulting in multiple extraction stages.

[0075] For example, during the plant history extraction process, this embodiment periodically collects target component concentration data and records the corresponding time to form a time-concentration dataset. This embodiment first determines the number of clusters K; for example, for most plant extraction processes, K is set to 3, corresponding to the infiltration stage, rapid extraction stage, equilibration stage, etc. Taking the K-Means algorithm as an example, this embodiment uses the time-concentration data as input and runs the algorithm. The algorithm randomly selects K initial cluster centers and iteratively adjusts them to assign each data point to the cluster most similar to it, until the cluster centers no longer change or change very little. Based on the clustering results, this embodiment divides the time intervals corresponding to data belonging to the same cluster into one extraction stage.

[0076] The extraction stage division method based on kinetic curves and clustering algorithms provided in this embodiment can accurately grasp the concentration change pattern of target components during plant extraction. By rationally dividing the stages, this embodiment lays the foundation for subsequently determining the optimal control parameters for each stage, thereby improving extraction efficiency and product quality, reducing ineffective extraction time, lowering energy consumption, and enhancing the scientific and economic aspects of the entire plant extraction process.

[0077] In one embodiment of this application, determining the initial particle swarm based on a temporal parameter vector includes:

[0078] The initial and target particle counts are determined based on the time-series parameter vector.

[0079] If the initial number of particles is less than the target number of particles, then the target number of particles is used as the number of particles in the initial particle space, and the temporal parameter vector is used as the first particle swarm; the initial particle space includes the first particle swarm and multiple zero-vector particles.

[0080] A second particle swarm is obtained by initializing multiple zero-vector particles based on the initial parameter range and parameter coupling constraints; the first and second particle swarms are used as the initial particle swarm.

[0081] The sum of the number of particles in the first particle group and the number of particles in the second particle group is the target number of particles.

[0082] Parameter coupling constraints include the relationship constraints between various control parameters; the initial parameter range includes the range of each control parameter.

[0083] In this embodiment, the initial parameter range includes the temperature range, pressure range, and stirring speed range; the parameter coupling constraints include temperature-pressure constraints and stirring speed-pressure constraints.

[0084] The second particle swarm is obtained by initializing multiple zero-vector particles based on the initial parameter range and parameter coupling constraints, including:

[0085] Perform the first initialization operation for each zero-vector particle;

[0086] The first initialization operation includes:

[0087] Within the pressure range, a pressure value is randomly initialized for the particle. Based on the pressure value, temperature range, and temperature-pressure constraint, the feasible temperature domain of the particle is determined. Within the feasible temperature domain, a temperature value is randomly initialized for the particle.

[0088] Based on the pressure value, stirring speed range, and stirring speed-pressure constraint, the feasible region of stirring speed for the particle is determined, and a stirring speed value is randomly initialized for the particle within the feasible region of stirring speed.

[0089] The pressure, temperature, and stirring speed values ​​are used as the initial values ​​for this particle.

[0090] In this embodiment, the initial parameter range includes the temperature range, pressure range, and stirring speed range; the parameter coupling constraints include temperature-pressure constraints and stirring speed-pressure constraints.

[0091] The second particle swarm is obtained by initializing multiple zero-vector particles based on the initial parameter range and parameter coupling constraints, including:

[0092] Perform the second initialization operation for each zero-vector particle;

[0093] The second initialization operation includes:

[0094] Within the temperature range, a temperature value is randomly initialized for the particle. Based on the temperature value, pressure range, and temperature-pressure constraints, the pressure feasible region of the particle is determined. Within the pressure feasible region, a pressure value is randomly initialized for the particle.

[0095] The feasible region of stirring speed for the particle is determined based on the pressure value, stirring speed range, and stirring speed-pressure constraint. Within the feasible region of stirring speed, a stirring speed value is randomly initialized for the particle.

[0096] The pressure, temperature, and stirring speed values ​​are used as the initial values ​​for this particle.

[0097] In this embodiment, the initial particle count is determined based on the temporal parameter vector and is related to the amount of historical plant data extracted. The target particle count is the particle swarm size set according to algorithm optimization requirements and computing resources, representing the desired total number of particles. When the initial particle count is less than the target particle count, data augmentation is needed to obtain a second particle swarm, so that the sum of the first and second particle swarm counts reaches the target particle count. When the initial particle count is greater than the target particle count, data downsampling is required to select the target particle count to form the initial particle swarm.

[0098] The first particle swarm is a collection of particles directly composed of temporal parameter vectors. These particles contain extracted parameters from actual historical data, providing a basic feasible solution for optimization. Zero-vector particles are particles with initial values ​​of zero, used to supplement the particle swarm size and subsequently initialized according to constraints. The second particle swarm refers to the collection of particles obtained by initializing the zero-vector particles based on constraints, which can increase the diversity of solutions.

[0099] The initial parameter range can include the allowable value range of control parameters such as temperature, pressure, and stirring speed, for example, temperature [30℃, 80℃], pressure [0.5MPa, 2.0MPa], and stirring speed [50rpm, 300rpm]. Parameter coupling constraints refer to the constraint relationships between control parameters. For example, a temperature-pressure constraint means that the pressure must increase accordingly when the temperature increases to prevent the solvent from boiling; a stirring speed-pressure constraint means that changes in stirring speed will affect the pressure.

[0100] This embodiment constructs a first particle swarm using historical data to ensure the feasibility of the initial solution. A second particle swarm is generated using zero-vector particles to increase solution diversity and prevent the algorithm from getting trapped in local optima. When initializing the zero-vector particles, the feasible region of each parameter is determined based on parameter coupling constraints and the initial parameter range. Random values ​​are selected to ensure that the generated particles meet actual production requirements, providing a rich initial solution space for subsequent particle swarm optimization.

[0101] For example, in this embodiment, the initial number of particles is determined based on the amount of data contained in the time-series parameter vector, and compared with the target number of particles. If the initial number of particles is insufficient, the number of particles to be supplemented is determined. In this embodiment, the time-series parameter vector is converted into particles, and each particle corresponds to a set of parameter combinations from the historical extraction stages, forming a first particle swarm.

[0102] In this embodiment, for each zero-vector particle, a pressure value is first randomly selected within the pressure range, such as 1.2 MPa within the range of [0.5 MPa, 2.0 MPa]. The feasible temperature region is then determined based on the temperature-pressure constraint; for example, when the pressure is 1.2 MPa, the feasible temperature region is [40℃, 60℃], and a temperature value is randomly selected within this range. Next, the feasible stirring speed region is determined based on the pressure value and the stirring speed-pressure constraint; for example, when the pressure is 1.2 MPa, the feasible stirring speed region is [100 rpm, 200 rpm], and a stirring speed value is randomly selected to complete particle initialization.

[0103] In this embodiment, the temperature value can be initialized randomly first, the feasible pressure domain can be determined and the value can be selected, and finally the feasible stirring speed domain can be determined and the value can be selected to complete the particle initialization.

[0104] In this embodiment, the first and second particle swarms are merged to obtain an initial particle swarm that meets the target particle number requirement, providing an initial solution for subsequent particle swarm optimization algorithms.

[0105] This embodiment effectively constructs an initial particle swarm by combining historical data with random initialization. The first particle swarm constructed using historical data ensures the feasibility of the initial solution, while the second particle swarm, obtained by initializing zero-vector particles based on constraints, increases the diversity of solutions. This embodiment avoids the algorithm getting trapped in local optima, providing a rich solution space for subsequent particle swarm optimization, and helps improve optimization performance and extraction process quality.

[0106] In one embodiment of this application, the particle update formula includes a position update formula and a velocity update formula;

[0107] The initial particle swarm is updated based on the particle update formula, including:

[0108] The velocity of each particle in the initial particle swarm is updated based on the velocity update formula;

[0109] The speed update formula is:

[0110] ;

[0111] Where t is the initial node of the current iteration process, and t+1 is the completion node of the current iteration process. For the first The updated velocity of the i-th particle in each extraction stage For inertial weights, For the first The velocity of the i-th particle before the update in each extraction stage and As a learning factor, and Uniformly distributed random numbers, For the first The historical best position of the i-th particle in each extraction stage. For the first The global optimal position in each extraction stage. For the first The control parameter values ​​for the i-th particle in each extraction stage. For time series smoothing coefficients, For the first -1 control parameter values ​​for the i-th particle in the extraction stage;

[0112] The position of each particle in the initial particle swarm is updated based on the position update formula.

[0113] In this embodiment, the position update formula is:

[0114] ∈[1,j];

[0115] in, For the first The updated position of the i-th particle in each extraction stage. For the first The position of the i-th particle before the update in each extraction stage. For the first The velocity of the i-th particle in each extraction stage.

[0116] In this embodiment, the inertia weight can be between 0.4 and 0.9, controlling the degree to which particles inherit previous velocities. A larger inertia weight makes particles tend to maintain their original direction of motion, facilitating global search; a smaller inertia weight makes particles focus more on local information, allowing for fine-tuning. The learning factor can be 2. Used to control the degree to which a particle learns from its own best historical position. The degree to which particles learn towards the global optimal position is controlled, and these two factors determine the adjustment strength of the particle search direction. Uniformly distributed random numbers can be randomly generated within the interval [0,1], introducing randomness into the particle search and preventing the algorithm from getting trapped in local optima. The temporal smoothing coefficient can be between 0 and 1, used to constrain the parameter variation amplitude between adjacent extraction stages, preventing sudden changes in control parameters and ensuring stable equipment operation.

[0117] In this embodiment, each particle has different control parameters, corresponding to different positions in the particle space. The particle update formula simulates the foraging behavior of a flock of birds, guiding the particles to search for the optimal solution in the solution space through velocity and position updates. The velocity update formula integrates the particle's current velocity, learning from its own historical best and global best positions, and the parameter constraints of adjacent stages to determine a new velocity for the particle. The position update formula adjusts the particle's position based on the updated velocity, allowing the particle to gradually approach the optimal solution.

[0118] For example, this embodiment first initializes the particle swarm, determining the initial position and velocity of each particle. This embodiment calculates the fitness value of each particle, determining the global optimal position and the historical optimal position for each particle. For each particle, this embodiment calculates the updated velocity according to the velocity update formula. This embodiment calculates the updated position according to the position update formula. This embodiment checks whether the iteration termination condition is met; if not, the above steps are repeated.

[0119] For example, a company uses plant extraction equipment to extract paclitaxel from yew trees. Before starting extraction, an initial particle swarm is determined based on historical extraction data and process requirements. Each particle contains control parameters such as temperature, pressure, and stirring speed for different extraction stages.

[0120] During iterative optimization, the particle update formula is used to update the particles. For example, in the 5th iteration, a particle is in the 2nd extraction stage, l=2, and i is the particle number. At this time, the inertia weight ω is set to 0.6. The inertia weight allows the particle to maintain its previous motion trend to a certain extent, searching for possible solutions within a larger range. (Learning factor) and All are 2, random numbers and They are randomly generated in the interval [0,1], assuming =0.3, =0.7.

[0121] According to the velocity update formula, the particle determines its updated velocity by referring to its historical best position, global best position, and parameters from the previous stage. Once the updated velocity is calculated, the particle's position in the solution space is updated according to the position update formula.

[0122] After multiple iterations, when the iteration conditions are met, such as the fitness convergence accuracy reaching the required level, the optimal parameter combination is obtained. The company applies this combination to actual extraction equipment, strictly controlling the temperature, pressure, and stirring speed according to the optimized parameters during the extraction process.

[0123] This embodiment balances global and local searches, enabling the algorithm to converge to the optimal solution faster, reducing computation time. The timing smoothing coefficient ensures a smooth transition of parameters between adjacent stages, preventing sudden changes in equipment parameters and extending the equipment's service life.

[0124] A control method for a plant extraction device corresponding to the above embodiment, Figure 2 This is a structural block diagram of a control system for a plant extraction device according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The control system 20 of the plant extraction equipment includes: a particle swarm construction module 21, a fitness calculation module 22, a parameter optimization module 23, and an equipment control module 24.

[0125] Among them, the particle swarm construction module 21 is used to splice the control parameters corresponding to each extraction stage in the plant history extraction process to obtain a time-series parameter vector, and determine the initial particle swarm based on the time-series parameter vector; the control parameters include the temperature, pressure and stirring speed in the plant extraction tank of the plant extraction equipment.

[0126] The fitness calculation module 22 is used to calculate the fitness value of each particle in the initial particle swarm using the fitness function, take the initial position of each particle as the historical best position of that particle, and take the position of the particle with the largest fitness value in the initial particle swarm as the global best position.

[0127] The parameter optimization module 23 is used to take the global optimal position and the historical optimal position of each particle as the initial optimal parameter combination, update each particle in the initial particle swarm based on the particle update formula to obtain a new particle swarm, until the iteration condition is met; take the particle swarm obtained from the last update as the target particle swarm, and update the initial optimal parameter combination based on the target particle swarm to obtain the target optimal parameter combination;

[0128] Equipment control module 24 is used to control the plant extraction equipment using the target optimal parameter combination.

[0129] In one embodiment of this application, for each historical extraction process of a plant, before splicing the control parameters corresponding to each extraction stage in the historical extraction process of the plant to obtain the time-series parameter vector, the process further includes: a stage division module, which is used to divide the plant extraction process into multiple extraction stages based on the dynamic curve of the target component concentration changing with time in the historical extraction process of the target plant using a clustering algorithm;

[0130] The target plant is the plant from which the target component is extracted using plant extraction equipment.

[0131] In one embodiment of this application, the particle swarm construction module 21 is specifically used to determine the initial number of particles and the target number of particles based on the time-series parameter vector;

[0132] If the initial number of particles is less than the target number of particles, then the target number of particles is used as the number of particles in the initial particle space, and the temporal parameter vector is used as the first particle swarm; the initial particle space includes the first particle swarm and multiple zero-vector particles.

[0133] A second particle swarm is obtained by initializing multiple zero-vector particles based on the initial parameter range and parameter coupling constraints; the first and second particle swarms are used as the initial particle swarm.

[0134] The sum of the number of particles in the first particle group and the number of particles in the second particle group is the target number of particles.

[0135] Parameter coupling constraints include the relationship constraints between various control parameters; the initial parameter range includes the range of each control parameter.

[0136] In one embodiment of this application, the fitness function is:

[0137] ;

[0138] in, ;

[0139] in, Each particle represents a control parameter value corresponding to all extraction stages. Let X be the fitness value of particle X. Here, j represents the index of the extraction stage, and j represents the total number of extraction stages. For the first Weighting coefficients for each extraction stage For the first The local fitness function of the stage. For the first Particles in each extraction stage The corresponding fitness value, Here, g is the constraint penalty coefficient, and g() is the dynamic constraint function for adjacent extraction stages. For the first Control parameters for each extraction stage For the first -1 control parameters for the extraction stage, where Y() is the extraction rate function for the plant target components. To preset the target extraction rate, For plant extraction purity function, To achieve the preset target purity, Let the energy consumption function be the energy consumption function of the plant extraction equipment. For plant extraction time function, , , and These are the weighting coefficients. .

[0140] In one embodiment of this application, the particle update formula includes a position update formula and a velocity update formula; the parameter optimization module 23 is specifically used to update the velocity of each particle in the initial particle swarm based on the velocity update formula;

[0141] The speed update formula is:

[0142] ;

[0143] Where t is the initial node of the current iteration process, and t+1 is the completion node of the current iteration process. For the first The updated velocity of the i-th particle in each extraction stage For inertial weights, For the first The velocity of the i-th particle before the update in each extraction stage and As a learning factor, and Uniformly distributed random numbers, For the first The historical best position of the i-th particle in each extraction stage. For the first The global optimal position in each extraction stage. For the first The control parameter values ​​for the i-th particle in each extraction stage. For time series smoothing coefficients, For the first -1 control parameter values ​​for the i-th particle in the extraction stage;

[0144] The position of each particle in the initial particle swarm is updated based on the position update formula.

[0145] In one embodiment of this application, the initial parameter range includes a temperature range, a pressure range, and a stirring speed range; the parameter coupling constraints include temperature-pressure constraints and stirring speed-pressure constraints; the particle swarm construction module 21 is further configured to perform a first initialization operation for each zero vector particle;

[0146] The first initialization operation includes:

[0147] Within the pressure range, a pressure value is randomly initialized for the particle. Based on the pressure value, temperature range, and temperature-pressure constraint, the feasible temperature domain of the particle is determined. Within the feasible temperature domain, a temperature value is randomly initialized for the particle.

[0148] Based on the pressure value, stirring speed range, and stirring speed-pressure constraint, the feasible region of stirring speed for the particle is determined, and a stirring speed value is randomly initialized for the particle within the feasible region of stirring speed.

[0149] The pressure, temperature, and stirring speed values ​​are used as the initial values ​​for this particle.

[0150] In one embodiment of this application, the initial parameter range includes a temperature range, a pressure range, and a stirring speed range; the parameter coupling constraints include temperature-pressure constraints and stirring speed-pressure constraints; the particle swarm construction module 21 is further configured to perform a second initialization operation for each zero vector particle;

[0151] The second initialization operation includes:

[0152] Within the temperature range, a temperature value is randomly initialized for the particle. Based on the temperature value, pressure range, and temperature-pressure constraints, the pressure feasible region of the particle is determined. Within the pressure feasible region, a pressure value is randomly initialized for the particle.

[0153] The feasible region of stirring speed for the particle is determined based on the pressure value, stirring speed range, and stirring speed-pressure constraint. Within the feasible region of stirring speed, a stirring speed value is randomly initialized for the particle.

[0154] The pressure, temperature, and stirring speed values ​​are used as the initial values ​​for this particle.

[0155] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned system embodiments, for example... Figure 2 The functions of the particle swarm construction module 21, fitness calculation module 22, parameter optimization module 23, and device control module 24 are shown.

[0156] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0157] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0158] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store information about device operating parameters.

[0159] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in the embodiments of the control method for a plant extraction device provided in the embodiments of this application, or they can execute the implementation methods of the electronic device 300 described in the embodiments of this application, which will not be repeated here.

[0160] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0161] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0162] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0163] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0164] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0165] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0166] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0167] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control method for a plant extraction device, characterized in that, include: A time-series parameter vector is obtained by splicing the control parameters corresponding to each extraction stage in the plant history extraction process, and an initial particle swarm is determined based on the time-series parameter vector; the control parameters include the temperature, pressure and stirring speed in the plant extraction tank of the plant extraction equipment; The fitness value of each particle in the initial particle swarm is calculated using the fitness function. The initial position of each particle is taken as the historical best position of that particle, and the position of the particle with the largest fitness value in the initial particle swarm is taken as the global best position. The fitness function is: ； in, ; in, Each particle represents a control parameter value corresponding to all extraction stages. Let X be the fitness value of particle X. Here, j represents the index of the extraction stage, and j represents the total number of extraction stages. For the first Weighting coefficients for each extraction stage For the first The local fitness function of the stage. For the first Particles in each extraction stage The corresponding fitness value, Here, g is the constraint penalty coefficient, and g() is the dynamic constraint function for adjacent extraction stages. For the first Control parameters for each extraction stage For the first -1 control parameters for the extraction stage, where Y() is the extraction rate function for the plant target components. To preset the target extraction rate, For plant extraction purity function, To achieve the preset target purity, Let the energy consumption function be the energy consumption function of the plant extraction equipment. For plant extraction time function, , , and These are the weighting coefficients. ; The target component extraction rate function is: in, , These are the fitting coefficients obtained through regression analysis of historical data. for Temperature parameters in for The pressure parameter S is... The stirring speed parameter in the text; The purity function of plant extracts is: in, This is the impurity dissolution coefficient. It is a fixed constant; The energy consumption function of the plant extraction equipment is: in, To preset standard temperature, For the first Stirring time for each extraction stage; The global optimal position and the historical optimal position of each particle are used as the initial optimal parameter combination. Each particle in the initial particle swarm is updated based on the particle update formula to obtain a new particle swarm, until the iteration condition is met. The particle swarm obtained from the last update is used as the target particle swarm. The initial optimal parameter combination is updated based on the target particle swarm to obtain the target optimal parameter combination. The plant extraction equipment is controlled using the target optimal parameter combination.

2. The control method for a plant extraction device as described in claim 1, characterized in that, For each plant history extraction process, before concatenating the control parameters corresponding to each extraction stage to obtain the time-series parameter vector, the following steps are also included: Based on the dynamic curve of the target plant's target component concentration changing over time during the historical extraction process, a clustering algorithm was used to divide the plant extraction process into multiple extraction stages; The target plant is the plant from which the target component is extracted by the plant extraction device.

3. The control method for a plant extraction device as described in claim 1, characterized in that, The determination of the initial particle swarm based on the time-series parameter vector includes: The initial number of particles and the target number of particles are determined based on the time-series parameter vector; If the initial number of particles is less than the target number of particles, then the target number of particles is used as the number of particles in the initial particle space, and the timing parameter vector is used as the first particle swarm; the initial particle space includes the first particle swarm and multiple zero-vector particles. The multiple zero-vector particles are initialized based on the initial parameter range and parameter coupling constraints to obtain a second particle group; the first particle group and the second particle group are used as the initial particle group. The sum of the number of particles in the first particle group and the number of particles in the second particle group is the target number of particles. The parameter coupling constraint includes the relationship constraint between various control parameters; the initial parameter range includes the range of each control parameter.

4. The control method for a plant extraction device as described in claim 1, characterized in that, The particle update formula includes a position update formula and a velocity update formula; The process of updating each particle in the initial particle swarm based on the particle update formula includes: The velocity of each particle in the initial particle swarm is updated based on the velocity update formula; The speed update formula is: Where t is the initial node of the current iteration process, and t+1 is the completion node of the current iteration process. For the first The updated velocity of the i-th particle in each extraction stage For inertial weights, For the first The velocity of the i-th particle before the update in each extraction stage and As a learning factor, and Uniformly distributed random numbers, For the first The historical best position of the i-th particle in each extraction stage. For the first The global optimal position in each extraction stage. For the first The control parameter values ​​for the i-th particle in each extraction stage. For time series smoothing coefficients, For the first -1 control parameter values ​​for the i-th particle in the extraction stage; The position of each particle in the initial particle swarm is updated based on the position update formula.

5. The control method for a plant extraction device as described in claim 3, characterized in that, The initial parameter range includes a temperature range, a pressure range, and a stirring speed range; the parameter coupling constraints include temperature-pressure constraints and stirring speed-pressure constraints. The initialization of the multiple zero-vector particles based on the initial parameter range and parameter coupling constraints to obtain the second particle swarm includes: Perform the first initialization operation for each zero-vector particle; The first initialization operation includes: Within the pressure range, a pressure value is randomly initialized for the particle. Based on the pressure value, the temperature range, and the temperature-pressure constraint, the feasible temperature domain of the particle is determined. Within the feasible temperature domain, a temperature value is randomly initialized for the particle. Based on the pressure value, stirring speed range, and stirring speed-pressure constraint, the feasible stirring speed region of the particle is determined, and a stirring speed value is randomly initialized for the particle within the feasible stirring speed region. The pressure, temperature, and stirring speed values ​​are used as the initial values ​​for the particle.

6. The control method for a plant extraction device as described in claim 3, characterized in that, The initial parameter range includes a temperature range, a pressure range, and a stirring speed range; the parameter coupling constraints include temperature-pressure constraints and stirring speed-pressure constraints. The initialization of the multiple zero-vector particles based on the initial parameter range and parameter coupling constraints to obtain the second particle swarm includes: Perform the second initialization operation for each zero-vector particle; The second initialization operation includes: Within the temperature range, a temperature value is randomly initialized for the particle. Based on the temperature value, the pressure range, and the temperature-pressure constraint, the pressure feasible region of the particle is determined. Within the pressure feasible region, a pressure value is randomly initialized for the particle. Based on the pressure value, stirring speed range, and stirring speed-pressure constraint, the feasible stirring speed region of the particle is determined, and a stirring speed value is randomly initialized for the particle within the feasible stirring speed region. The pressure, temperature, and stirring speed values ​​are used as the initial values ​​for the particle.

7. A control system for a plant extraction device, characterized in that, include: The particle swarm construction module is used to splice together the control parameters corresponding to each extraction stage in the plant history extraction process to obtain a time-series parameter vector, and to determine the initial particle swarm based on the time-series parameter vector; the control parameters include the temperature, pressure and stirring speed in the plant extraction tank of the plant extraction equipment. The fitness calculation module is used to calculate the fitness value of each particle in the initial particle swarm using the fitness function, take the initial position of each particle as the historical best position of that particle, and take the position of the particle with the largest fitness value in the initial particle swarm as the global best position. The fitness function is: ; in, ; in, Each particle represents a control parameter value corresponding to all extraction stages. Let X be the fitness value of particle X. Here, j represents the index of the extraction stage, and j represents the total number of extraction stages. For the first Weighting coefficients for each extraction stage For the first The local fitness function of the stage. For the first Particles in each extraction stage The corresponding fitness value, Here, g is the constraint penalty coefficient, and g() is the dynamic constraint function for adjacent extraction stages. For the first Control parameters for each extraction stage For the first -1 control parameters for the extraction stage, where Y() is the extraction rate function for the plant target components. To preset the target extraction rate, For plant extraction purity function, To achieve the preset target purity, Let the energy consumption function be the energy consumption function of the plant extraction equipment. For plant extraction time function, , , and These are the weighting coefficients. ; The target component extraction rate function is: in, , These are the fitting coefficients obtained through regression analysis of historical data. for Temperature parameters in for The pressure parameter S is... The stirring speed parameter in the text; The purity function of plant extracts is: in, This is the impurity dissolution coefficient. It is a fixed constant; The energy consumption function of the plant extraction equipment is: in, To preset standard temperature, For the first Stirring time for each extraction stage; The parameter optimization module is used to take the global optimal position and the historical optimal position of each particle as the initial optimal parameter combination, update each particle in the initial particle swarm based on the particle update formula to obtain a new particle swarm, until the iteration condition is met; take the particle swarm obtained from the last update as the target particle swarm, and update the initial optimal parameter combination based on the target particle swarm to obtain the target optimal parameter combination; The equipment control module is used to control the plant extraction equipment using the optimal combination of target parameters.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6.

Images