Dynamic temperature control plant extraction method based on adaptive algorithm

Abstract

The invention discloses a dynamic temperature control plant extraction method based on an adaptive algorithm, and the method mainly comprises the following steps: S1, initializing extraction equipment, and setting a target temperature interval; s2, collecting and extracting real-time temperature data in the device and transmitting the real-time temperature data to a control system; s3, analyzing temperature change by using a self-adaptive algorithm, and automatically adjusting control parameters; s4, power distribution of heating and cooling equipment is optimized through a genetic algorithm, and an optimal temperature control path is found; s5, dynamically adjusting the operation of heating and cooling equipment; s6, optimizing a temperature control strategy in real time according to the extraction effect by using a reinforcement learning algorithm; and S7, after the extraction process is completed, storing the temperature control data. Through the combination of the adaptive algorithm, the genetic algorithm and the reinforcement learning algorithm, accurate temperature control in the plant extraction process is realized, the extraction efficiency and the stability of active components are ensured, the production efficiency is effectively improved, and the energy consumption is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control of plant extraction, and particularly relates to a dynamic temperature control plant extraction method based on an adaptive algorithm. BACKGROUND

[0002] In the field of plant extraction technology, temperature control is a key factor that determines the extraction efficiency and the retention of active ingredients. Traditional plant extraction methods usually rely on manual or fixed program set temperature control schemes, which have many limitations in practical applications. The temperature control systems of the prior art mostly use preset heating or cooling parameters, lack real-time response capability to temperature changes during the extraction process, and cannot adaptively adjust according to the different characteristics of the extract and the dynamic changes of the environmental conditions during the extraction process. This makes it difficult for traditional temperature control systems to achieve efficient and accurate temperature control when facing different plant extracts and complex extraction conditions, resulting in reduced extraction efficiency.

[0003] First, the temperature control method of the prior art mainly relies on fixed parameter settings, which is too mechanical in actual application and cannot flexibly respond to temperature changes during the extraction process; for example, environmental temperature, equipment status, extract characteristics and other factors during the extraction process will cause temperature fluctuations, and the existing temperature control system lacks the ability of dynamic adjustment, which is prone to problems of excessively high or low temperature. In this case, excessively high temperature may cause the active ingredients in the extract to be destroyed, and excessively low temperature will reduce the extraction efficiency, prolong the extraction time and increase the energy consumption.

[0004] Second, the traditional temperature control system lacks intelligent optimization means for the temperature control process; the prior art usually uses a simple linear feedback control system, which cannot accurately analyze and adjust the temperature changes during the extraction process. In the face of complex extraction processes, the response speed of the existing system is slow, the adjustment precision is insufficient, and it is difficult to respond to rapid changes in extraction requirements. At the same time, the traditional temperature control method usually only considers a single temperature dimension, lacks adaptive adjustment capability for different extraction conditions, and cannot achieve dynamic optimization and distribution of heating and cooling power, thereby limiting the optimization of extraction efficiency and energy consumption.

[0005] Third, the existing temperature control technology has obvious limitations in large-scale industrial production; in large-scale plant extraction processes, temperature control has a significant impact on energy consumption. Due to the response lag of the traditional temperature control system, it often needs to maintain high power for a long time to maintain the temperature range required for extraction, which not only increases energy consumption, but also increases the burden on the equipment and shortens the service life of the equipment. In addition, the temperature control system lacking intelligent adjustment cannot optimize the heating and cooling strategy according to the changes in the extraction process, resulting in low energy efficiency.

[0006] Therefore, how to provide a dynamic temperature control plant extraction method based on an adaptive algorithm is a problem that those skilled in the art urgently need to solve. SUMMARY

[0007] One object of the present application is to provide a dynamic temperature control plant extraction method based on an adaptive algorithm. The present application makes full use of adaptive algorithms, genetic algorithms and reinforcement learning algorithms, and describes in detail the technical scheme of intelligent temperature control in the plant extraction process. The method automatically adjusts the power distribution of heating and cooling equipment by collecting real-time temperature data during the extraction process, and optimizes the temperature control strategy combined with a self-learning mechanism. It has the advantages of high temperature control precision, high extraction efficiency and low energy consumption, effectively guarantees the stability of active ingredients in plant extracts and extraction effect, and is suitable for large-scale industrial extraction production.

[0008] According to the dynamic temperature control plant extraction method based on an adaptive algorithm, the method comprises the following steps:

[0009] S1, initialization of the extraction equipment, setting the temperature range and extraction conditions of the target plant extract;

[0010] S2, collecting real-time temperature data inside the extraction device using a sensor, and transmitting the temperature data to a control system;

[0011] S3, analyzing real-time temperature data through an adaptive algorithm, automatically adjusting control parameters according to the temperature change trend during the extraction process, and gradually optimizing the control strategy according to historical temperature data and real-time feedback;

[0012] S4, introducing a global optimization module based on a genetic algorithm, optimizing the power distribution of heating and cooling equipment through a genetic algorithm, and finding the optimal temperature regulation path during the extraction process;

[0013] S5, dynamically adjusting the operation of heating and cooling equipment according to the output results of the adaptive algorithm and the genetic algorithm;

[0014] S6, during the extraction process, continuously self-learning using a reinforcement learning algorithm, and adjusting the temperature control strategy in real time according to the extraction effect of active ingredients during the extraction process;

[0015] S7, after completing the plant extraction process, automatically turning off the heating and cooling equipment, and saving the temperature control data during the extraction process.

[0016] Optionally, the S3 specifically comprises:

[0017] S31, obtaining real-time temperature data during the extraction process and historical temperature data , wherein is a time parameter, temperature at current time, temperature data at historical time;

[0018] S32, establishing a temperature change model using an adaptive algorithm , calculating temperature change trend by weighted smoothing temperature change data :

[0019] ;

[0020] wherein, is a weighted smoothing coefficient, used to balance the influence of real-time temperature and historical temperature on trend calculation;

[0021] S33, based on the calculated temperature change trend , the adaptive algorithm adjusts control parameters , control parameters include heating power and cooling power , and performs dynamic allocation:

[0022] ;

[0023] wherein, is the maximum heating or cooling power;

[0024] S34, the adaptive algorithm continuously optimizes the control strategy according to the feedback data, and updates the allocation of heating and cooling power in real time, so that the temperature in the extraction process is kept within the predetermined target interval.

[0025] Optionally, the S33 specifically includes:

[0026] S331, calculating the temperature change trend , after that, the adaptive algorithm dynamically adjusts control parameters according to , control parameters include heating power and cooling power ;

[0027] S332, using a nonlinear adjustment formula, calculating control power according to the temperature change trend ;

[0028] S333, according to the size of real-time temperature change trend , the adaptive algorithm dynamically adjusts the coefficient used to control the adjustment smoothness:

[0029] ;

[0030] wherein, is the set target temperature;

[0031] S334, the adaptive algorithm dynamically corrects the control power through a feedback mechanism. Monitor and update in real time throughout the extraction process. and The allocation.

[0032] Optionally, S4 specifically includes:

[0033] S41. Based on the current temperature status of the extraction equipment and the target temperature range, initialize the genetic algorithm, and set the search range for heating power and cooling power, denoted as... and ,in For minimum power, This is the highest power output;

[0034] S42. The genetic algorithm represents the current temperature control state of the system as an individual through encoding. ,in and These represent the allocation values ​​for heating power and cooling power, respectively.

[0035] S43. Construct the improved fitness function The fitness function is used to evaluate the effectiveness of the current combination of temperature control parameters:

[0036] ;

[0037] in, For the target temperature, This is the current real-time temperature. The fitness adjustment coefficient is used to adjust the sensitivity of the fitness function to adapt to different temperature ranges.

[0038] S44. Through selection, crossover and mutation operations in the genetic algorithm, the combination of temperature control power allocation is continuously optimized, and the individual with the highest fitness value is selected as the basis for the next step of control parameters.

[0039] S45. In each generation iteration, new individuals are generated using a genetic algorithm. And according to the fitness function The evaluation process is conducted, and when the fitness function reaches a preset threshold, the optimal combination of heating and cooling power is output. and ;

[0040] S46. Optimal heating power output by the genetic algorithm. and optimal cooling power The operating status of heating and cooling equipment is dynamically adjusted.

[0041] Optionally, S4 specifically includes:

[0042] S431. Constructing the fitness function The fitness function is used to evaluate each individual. The combination of temperature control parameters;

[0043] S432, Adjustment coefficient in fitness function Used to control the system's response sensitivity under different temperature differences, among which The larger the value, the more sensitive the fitness function is to small temperature differences;

[0044] S433. In each algebraic iteration, through the fitness function Calculate the fitness value for each individual; a higher fitness value indicates a better current power combination. The closer the temperature gets to the target temperature, the more the fitness function output is used as the basis for selecting the optimal individual.

[0045] S434. To accelerate algorithm convergence, the fitness values ​​are ranked based on the fitness function, and the top... The individual with the highest fitness value enters the next generation for crossover and variation genetic operations;

[0046] S435. During the iterative process of the genetic algorithm, a dynamic fitness update mechanism is used. When the real-time temperature fluctuation is less than a specific value, the fitness level is adaptively reduced. The value optimizes the system's response to small temperature variations.

[0047] Optionally, S6 specifically includes:

[0048] S61. During the extraction process, the temperature data of the extract is acquired in real time. and extraction efficiency of active ingredients The extraction environment inside the extraction equipment is monitored by sensors, and the relationship between the extraction efficiency of active ingredients and temperature is recorded at each time point;

[0049] S62. Construct an optimization model for extraction efficiency and temperature control using reinforcement learning algorithms, and set a reward function. :

[0050] ;

[0051] in, To improve real-time extraction efficiency, For real-time temperature, For the target temperature, and This is the adjustment coefficient;

[0052] S63, Based on reward function Based on the feedback, the reinforcement learning algorithm dynamically adjusts the heating power. and cooling power Optimize temperature control strategy;

[0053] S64. The reinforcement learning algorithm continuously learns from historical data and real-time feedback within each extraction cycle. Through accumulated feedback, it gradually optimizes the relationship between extraction efficiency and temperature control parameters, specifically by adjusting coefficients. and To adapt to the temperature control requirements under different extraction conditions, and to continuously optimize the operation strategies of heating and cooling equipment;

[0054] S65. The reinforcement learning algorithm generates the optimal temperature control strategy curve based on real-time feedback and historical extraction data throughout the extraction process. .

[0055] Optionally, S62 specifically includes:

[0056] S621, Based on real-time temperature data during the extraction process and extraction efficiency of active ingredients Establish a reward function The nonlinear model, with a reward function used to evaluate the impact of the current temperature control strategy on extraction efficiency;

[0057] S622. In the reward function model, use the adjustment coefficient. and To control the system's sensitivity and penalty for temperature deviations;

[0058] S623. When the temperature deviation is less than a specific value, the system adjusts... Enhance the sensitivity of the reward function to extraction efficiency and optimize the response of the temperature control system to small temperature fluctuations;

[0059] S624. When the temperature deviation exceeds a specific value, the adjustment coefficient... This is used to increase the penalty for temperature deviation, enabling the system to automatically adjust heating and cooling power when faced with large temperature fluctuations;

[0060] S625, Reinforcement learning algorithm based on improved reward function Dynamically adjust heating power and cooling power ;

[0061] S626. The system continuously optimizes the adjustment coefficient through self-learning. and This allows it to adapt to the temperature requirements of different plant extracts and extraction processes, generating the optimal temperature control strategy curve.

[0062] Optionally, S64 specifically includes:

[0063] S641. Obtain real-time temperature data during the extraction process using a reinforcement learning algorithm. and extraction efficiency of active ingredients Simultaneously record historical extracted data ,in This includes historical temperature, extraction efficiency, and power allocation data;

[0064] S642. The reinforcement learning algorithm performs self-learning based on real-time feedback and historical data, constructs a non-linear relationship model between temperature control parameters and extraction efficiency, and sets the optimization objective of self-learning as maximizing extraction efficiency. Minimize temperature deviation ;

[0065] S643. Define the loss function Used to evaluate the effectiveness of the current temperature control strategy:

[0066] ;

[0067] in, For real-time temperature, For the target temperature, To improve real-time extraction efficiency, To achieve the expected maximum extraction efficiency, For balance coefficient, and This is an adjustment coefficient used to control the degree to which temperature deviation affects extraction efficiency. To avoid dividing by zero for small constants;

[0068] S644, Based on Loss Function Based on feedback, the reinforcement learning algorithm continuously optimizes the heating power through self-learning. and cooling power Adjust the temperature control strategy;

[0069] S645. The system dynamically updates the learning rate of the reinforcement learning algorithm based on historical data and real-time feedback. ;

[0070] S646, Reinforcement learning algorithm generates optimal temperature control strategy curve .

[0071] The beneficial effects of this invention are:

[0072] (1) This invention utilizes an adaptive algorithm to collect and analyze temperature data in real time, and automatically adjusts control parameters to ensure that the temperature can be timely and accurately regulated according to changes in the extraction process. By combining historical data and real-time feedback, the adaptive algorithm continuously optimizes the temperature control strategy, significantly improving the flexibility and accuracy of the temperature control system, and avoiding the problems of temperature fluctuation and excessive energy consumption caused by fixed parameter settings in traditional temperature control methods.

[0073] (2) This invention introduces a genetic algorithm to globally optimize the power allocation of heating and cooling equipment, ensuring that the temperature control system can use power more efficiently. By optimizing the temperature control path, the optimal temperature control scheme is found, reducing unnecessary energy consumption. This optimization scheme improves the efficiency of temperature control equipment, extends the service life of the equipment, and reduces energy consumption during operation.

[0074] (3) This invention utilizes a reinforcement learning algorithm for self-learning, automatically adjusting the operating strategies of the heating and cooling equipment based on real-time extraction efficiency and temperature control feedback during the extraction process. The reinforcement learning algorithm continuously optimizes the relationship between extraction efficiency and temperature control parameters based on historical extraction data and real-time feedback, ensuring that the temperature is always maintained within the optimal extraction range. This process not only improves extraction efficiency but also reduces the loss of active ingredients due to improper temperature.

[0075] (4) This invention combines adaptive algorithms, genetic algorithms, and reinforcement learning algorithms to adapt to different extraction conditions in different plant extraction processes and dynamically adjust the temperature requirements of various plant extracts. The system has high-efficiency adaptive capabilities and can optimize and control according to actual needs, thereby improving production efficiency and reducing energy consumption. It is particularly suitable for large-scale industrial plant extraction production. Attached Figure Description

[0076] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0077] Figure 1 This is a flowchart of a dynamic temperature-controlled plant extraction method based on an adaptive algorithm proposed in this invention;

[0078] Figure 2 This is a flowchart illustrating the steps of adjusting temperature control parameters using an adaptive algorithm in a dynamic temperature-controlled plant extraction method based on an adaptive algorithm proposed in this invention. Detailed Implementation

[0079] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0080] refer to Figures 1-2 A dynamic temperature-controlled plant extraction method based on an adaptive algorithm includes the following steps:

[0081] S1. Initialize the extraction equipment and set the temperature range and extraction conditions for the target plant extract;

[0082] S2. Use sensors to collect and extract temperature data inside the device in real time, and transmit the temperature data to the control system.

[0083] S3. Real-time temperature data is analyzed through an adaptive algorithm, and control parameters are automatically adjusted according to the temperature change trend during the extraction process. The adaptive algorithm also optimizes the control strategy step by step based on historical temperature data and real-time feedback.

[0084] S4. Introduce a global optimization module based on genetic algorithm to optimize the power allocation of heating and cooling equipment and find the optimal temperature control path in the extraction process.

[0085] S5. Based on the output results of the adaptive algorithm and the genetic algorithm, dynamically adjust the operation of the heating and cooling equipment;

[0086] S6. During the extraction process, reinforcement learning algorithms are used for continuous self-learning, and the temperature control strategy is adjusted in real time according to the extraction effect of active ingredients.

[0087] S7. After the plant extraction process is completed, the heating and cooling equipment will be automatically turned off, and the temperature control data during the extraction process will be saved.

[0088] In this embodiment, S3 specifically includes:

[0089] S31. Obtain real-time temperature data during the extraction process. and historical temperature data ,in For time parameters, This indicates the temperature at the current moment. This represents temperature data at historical moments.

[0090] S32. Establish a temperature change model using an adaptive algorithm. By using weighted smoothing of temperature change data, the temperature change trend is calculated. :

[0091] ;

[0092] in, This is a weighted smoothing coefficient used to balance the impact of real-time temperature and historical temperature on trend calculation;

[0093] S33, Calculation-based temperature change trend Adaptive algorithm adjusts control parameters Control parameters include heating power and cooling power And dynamically allocate:

[0094] ;

[0095] in, This represents the maximum heating or cooling power.

[0096] S34. The adaptive algorithm continuously optimizes the control strategy based on feedback data and updates the distribution of heating and cooling power in real time, so that the temperature during the extraction process is kept within the predetermined target range.

[0097] In this embodiment, S33 specifically includes:

[0098] S331. Calculate the temperature change trend Then, the adaptive algorithm is based on Dynamically adjust control parameters, including heating power. and cooling power ;

[0099] S332. A non-linear adjustment formula is adopted, based on the temperature change trend. Calculate control power ;

[0100] S333, Based on real-time temperature change trends The size of the coefficients is dynamically adjusted by the adaptive algorithm. Used to control the smoothness of adjustment:

[0101] ;

[0102] in, The set target temperature;

[0103] S334, the adaptive algorithm dynamically corrects the control power through a feedback mechanism. Monitor and update in real time throughout the extraction process. and The allocation.

[0104] In this embodiment, S4 specifically includes:

[0105] S41. Based on the current temperature status of the extraction equipment and the target temperature range, initialize the genetic algorithm, and set the search range for heating power and cooling power, denoted as... and ,in For minimum power, This is the highest power output;

[0106] S42. The genetic algorithm represents the current temperature control state of the system as an individual through encoding. ,in and These represent the allocation values ​​for heating power and cooling power, respectively.

[0107] S43. Construct the improved fitness function The fitness function is used to evaluate the effectiveness of the current combination of temperature control parameters:

[0108] ;

[0109] in, For the target temperature, This is the current real-time temperature. The fitness adjustment coefficient is used to adjust the sensitivity of the fitness function to adapt to different temperature ranges.

[0110] S44. Through selection, crossover and mutation operations in the genetic algorithm, the combination of temperature control power allocation is continuously optimized, and the individual with the highest fitness value is selected as the basis for the next step of control parameters.

[0111] S45. In each generation iteration, new individuals are generated using a genetic algorithm. And according to the fitness function The evaluation process is conducted, and when the fitness function reaches a preset threshold, the optimal combination of heating and cooling power is output. and ;

[0112] S46. Optimal heating power output by the genetic algorithm. and optimal cooling power The operating status of heating and cooling equipment is dynamically adjusted.

[0113] In this embodiment, S4 specifically includes:

[0114] S431. Constructing the fitness function The fitness function is used to evaluate each individual. The combination of temperature control parameters;

[0115] S432, Adjustment coefficient in fitness function Used to control the system's response sensitivity under different temperature differences, among which The larger the value, the more sensitive the fitness function is to small temperature differences;

[0116] S433. In each algebraic iteration, through the fitness function Calculate the fitness value for each individual; a higher fitness value indicates a better current power combination. The closer the temperature gets to the target temperature, the more the fitness function output is used as the basis for selecting the optimal individual.

[0117] S434. To accelerate algorithm convergence, the fitness values ​​are ranked based on the fitness function, and the top... The individual with the highest fitness value enters the next generation for crossover and variation genetic operations;

[0118] S435. During the iterative process of the genetic algorithm, a dynamic fitness update mechanism is used. When the real-time temperature fluctuation is less than a specific value, the fitness level is adaptively reduced. The value optimizes the system's response to small temperature variations.

[0119] In this embodiment, S6 specifically includes:

[0120] S61. During the extraction process, the temperature data of the extract is acquired in real time. and extraction efficiency of active ingredients The extraction environment inside the extraction equipment is monitored by sensors, and the relationship between the extraction efficiency of active ingredients and temperature is recorded at each time point;

[0121] S62. Construct an optimization model for extraction efficiency and temperature control using reinforcement learning algorithms, and set a reward function. :

[0122] ;

[0123] in, To improve real-time extraction efficiency, For real-time temperature, For the target temperature, and This is the adjustment coefficient;

[0124] S63, Based on reward function Based on the feedback, the reinforcement learning algorithm dynamically adjusts the heating power. and cooling power Optimize temperature control strategy;

[0125] S64. The reinforcement learning algorithm continuously learns from historical data and real-time feedback within each extraction cycle. Through accumulated feedback, it gradually optimizes the relationship between extraction efficiency and temperature control parameters, specifically by adjusting coefficients. and To adapt to the temperature control requirements under different extraction conditions, and to continuously optimize the operation strategies of heating and cooling equipment;

[0126] S65. The reinforcement learning algorithm generates the optimal temperature control strategy curve based on real-time feedback and historical extraction data throughout the extraction process. .

[0127] In this embodiment, S62 specifically includes:

[0128] S621, Based on real-time temperature data during the extraction process and extraction efficiency of active ingredients Establish a reward function The nonlinear model, with a reward function used to evaluate the impact of the current temperature control strategy on extraction efficiency;

[0129] S622. In the reward function model, use the adjustment coefficient. and To control the system's sensitivity and penalty for temperature deviations;

[0130] S623. When the temperature deviation is less than a specific value, the system adjusts... Enhance the sensitivity of the reward function to extraction efficiency and optimize the response of the temperature control system to small temperature fluctuations;

[0131] S624. When the temperature deviation exceeds a specific value, the adjustment coefficient... This is used to increase the penalty for temperature deviation, enabling the system to automatically adjust heating and cooling power when faced with large temperature fluctuations;

[0132] S625, Reinforcement learning algorithm based on improved reward function Dynamically adjust heating power and cooling power ;

[0133] S626. The system continuously optimizes the adjustment coefficient through self-learning. and This allows it to adapt to the temperature requirements of different plant extracts and extraction processes, generating the optimal temperature control strategy curve.

[0134] In this embodiment, S64 specifically includes:

[0135] S641. Obtain real-time temperature data during the extraction process using a reinforcement learning algorithm. and extraction efficiency of active ingredients Simultaneously record historical extracted data ,in This includes historical temperature, extraction efficiency, and power allocation data;

[0136] S642. The reinforcement learning algorithm performs self-learning based on real-time feedback and historical data, constructs a non-linear relationship model between temperature control parameters and extraction efficiency, and sets the optimization objective of self-learning as maximizing extraction efficiency. Minimize temperature deviation ;

[0137] S643. Define the loss function Used to evaluate the effectiveness of the current temperature control strategy:

[0138] ;

[0139] in, For real-time temperature, For the target temperature, To improve real-time extraction efficiency, To achieve the expected maximum extraction efficiency, For balance coefficient, and This is an adjustment coefficient used to control the degree to which temperature deviation affects extraction efficiency. To avoid dividing by zero for small constants;

[0140] S644, Based on Loss Function Based on feedback, the reinforcement learning algorithm continuously optimizes the heating power through self-learning. and cooling power Adjust the temperature control strategy;

[0141] S645. The system dynamically updates the learning rate of the reinforcement learning algorithm based on historical data and real-time feedback. ;

[0142] S646, Reinforcement learning algorithm generates optimal temperature control strategy curve .

[0143] Example 1:

[0144] The dynamic temperature-controlled plant extraction method based on an adaptive algorithm, as described in this invention, has been applied in a large-scale plant extraction factory located in China. This factory primarily extracts medicinal plants, and the extracts are widely used in the pharmaceutical, health product, and cosmetic industries. The total investment for the project is estimated at 300 million RMB. The plant extraction process involved is complex, and the active ingredients in the extracts are temperature-sensitive; therefore, traditional temperature control methods are insufficient to meet the demands for high-efficiency production and energy conservation. This invention aims to solve the problems of inaccurate temperature control, high energy consumption, and low extraction efficiency in existing technologies.

[0145] In this project, the factory adopted the intelligent temperature control system of this invention, which comprehensively optimized the temperature control requirements during plant extraction using adaptive algorithms, genetic algorithms, and reinforcement learning algorithms. During daily extraction production, the system continuously collects and analyzes temperature data through adaptive algorithms, adjusting the operating status of heating and cooling equipment in real time to ensure that the temperature during extraction is controlled within the predetermined optimal range.

[0146] In the extraction process of honeysuckle, a medicinal plant, precise temperature control is crucial because the active ingredients can only be efficiently extracted at specific temperatures. Traditional temperature control methods typically use fixed temperature settings, resulting in large temperature fluctuations that can easily lead to decreased extraction efficiency or even damage to the active ingredients. By applying the adaptive algorithm of this invention, the system can automatically adjust based on real-time and historical temperature data. For example, if the temperature rises from the initial setting of 80°C to 85°C within the first hour of extraction, the system detects that this temperature exceeds the target temperature range and immediately adjusts the heating power through the adaptive algorithm to restore the temperature to the optimal level of approximately 82°C. This precise temperature control not only improves extraction efficiency but also reduces energy consumption increases caused by temperature fluctuations. In the middle stage of extraction, the system further optimizes the power allocation of heating and cooling equipment through a genetic algorithm. Taking the honeysuckle extraction process as an example, the extraction process lasts for 8 hours, and temperature stability needs to be maintained every hour. The genetic algorithm dynamically optimizes the power allocation of the equipment, ensuring that the temperature control system achieves optimal temperature control performance with minimal energy consumption. In this process, the energy consumption of the extraction equipment was reduced by approximately 15% compared to traditional temperature control methods, while the extraction efficiency was improved by approximately 10%. This is because the genetic algorithm can globally optimize the power allocation for heating and cooling, avoiding the frequent start-stop phenomena common in traditional methods, thereby improving the stability and energy efficiency of equipment operation. In the final stage of the extraction process, the system introduced a reinforcement learning algorithm to achieve intelligent optimization of the extraction process. Taking the extraction of another medicinal plant, perilla, as an example, the effective components of perilla are extracted most efficiently when the temperature fluctuates slightly. The reinforcement learning algorithm continuously adjusts the temperature control strategy based on real-time feedback on the extraction effect. For example, in the fourth hour of extraction, the system detected that the extraction efficiency reached 90%, but the temperature was slightly higher than the set value. At this point, the reinforcement learning algorithm, based on historical data, allowed for slight temperature fluctuations, further improving the extraction efficiency. Ultimately, the extraction efficiency of perilla was increased to 94%, approximately 8% higher than the traditional method.

[0147] Through the intelligent temperature control system of this invention, the factory has achieved significant energy savings and efficiency improvements in actual production. The following is a display of key data from the extraction process:

[0148] Table 1. Data on temperature control effectiveness and energy consumption optimization during plant extraction.

[0149] Time period Extracted plant Initial temperature (°C) Target temperature interval (°C) Actual temperature interval (°C) Extraction efficiency (%) Energy consumption reduction (%) 1st hour Honeysuckle 80 82-83 82-83 85 10 2nd hour Honeysuckle 83 82-83 82-83 88 12 4th hour Purple perilla 85 83-85 84-85 90 15 6th hour Purple perilla 86 84-85 85-86 94 18 8th hour Purple perilla 87 85-86 85-86 94 20

[0150] As shown in Table 1 above, this invention achieves precise temperature control and energy consumption optimization at different stages of plant extraction by combining adaptive algorithms, genetic algorithms, and reinforcement learning algorithms. Taking honeysuckle extraction as an example, the extraction efficiency increased to 85% and energy consumption decreased by 10% within the first hour. In the 6th and 8th hours of perilla extraction, the extraction efficiency reached 94% respectively, while energy consumption decreased by 18% and 20%. This intelligent temperature control system not only significantly improves extraction efficiency but also significantly reduces energy consumption and optimizes production costs.

[0151] Through the implementation of this invention, the factory has significantly improved the efficiency of plant extraction, particularly in real-time temperature control and energy consumption optimization during the extraction process. For example, in the extraction of perilla, the system effectively adjusts the temperature control strategy at different stages of extraction by combining adaptive algorithms, genetic algorithms, and reinforcement learning algorithms, ensuring that the temperature remains within the optimal range and further optimizing energy consumption. At the 6th hour, the extraction efficiency reaches 94%, and energy consumption is reduced by 18% compared to traditional temperature control methods, significantly saving production costs.

[0152] In summary, this invention successfully solves the problems of inaccurate temperature control, high energy consumption, and low extraction efficiency in traditional plant extraction methods through an intelligent temperature control strategy. By applying this invention, factories have not only significantly improved extraction efficiency and optimized energy consumption, but also reduced equipment wear and operating costs, providing reliable technical support for large-scale industrial production.

[0153] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A dynamic temperature-controlled plant extraction method based on an adaptive algorithm, characterized in that, Includes the following steps: S1. Initialize the extraction equipment and set the temperature range and extraction conditions for the target plant extract; S2. Use sensors to collect and extract temperature data inside the device in real time, and transmit the temperature data to the control system. S3. Real-time temperature data is analyzed through an adaptive algorithm, and control parameters are automatically adjusted according to the temperature change trend during the extraction process. The adaptive algorithm also optimizes the control strategy step by step based on historical temperature data and real-time feedback. S4. Introduce a global optimization module based on genetic algorithm to optimize the power allocation of heating and cooling equipment and find the optimal temperature control path in the extraction process. S5. Based on the output results of the adaptive algorithm and the genetic algorithm, dynamically adjust the operation of the heating and cooling equipment; S6. During the extraction process, reinforcement learning algorithms are used for continuous self-learning, and the temperature control strategy is adjusted in real time according to the extraction effect of active ingredients. S7. After the plant extraction process is completed, the heating and cooling equipment will be automatically turned off, and the temperature control data during the extraction process will be saved.

2. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 1, characterized in that, S3 specifically includes: S31. Obtain real-time temperature data during the extraction process. and historical temperature data ,in For time parameters, This indicates the temperature at the current moment. This represents temperature data at historical moments. S32. Establish a temperature change model using an adaptive algorithm. By using weighted smoothing of temperature change data, the temperature change trend is calculated. : ； in, This is a weighted smoothing coefficient used to balance the impact of real-time temperature and historical temperature on trend calculation; S33, Calculation-based temperature change trend Adaptive algorithm adjusts control parameters Control parameters include heating power and cooling power And dynamically allocate: ； in, This represents the maximum heating or cooling power. S34. The adaptive algorithm continuously optimizes the control strategy based on feedback data and updates the distribution of heating and cooling power in real time, so that the temperature during the extraction process is kept within the predetermined target range.

3. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 2, characterized in that, Specifically, S33 includes: S331. Calculate the temperature change trend Then, the adaptive algorithm is based on Dynamically adjust control parameters, including heating power. and cooling power ; S332. A non-linear adjustment formula is adopted, based on the temperature change trend. Calculate control power ; S333, Based on real-time temperature change trends The size of the coefficients is dynamically adjusted by the adaptive algorithm. Used to control the smoothness of adjustment: ； in, The set target temperature; S334, the adaptive algorithm dynamically corrects the control power through a feedback mechanism. Monitor and update in real time throughout the extraction process. and The allocation.

4. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 1, characterized in that, S4 specifically includes: S41. Based on the current temperature status of the extraction equipment and the target temperature range, initialize the genetic algorithm, and set the search range for heating power and cooling power, denoted as... and ,in For minimum power, This is the highest power output; S42. The genetic algorithm represents the current temperature control state of the system as an individual through encoding. ,in and These represent the allocation values ​​for heating power and cooling power, respectively. S43. Construct the improved fitness function The fitness function is used to evaluate the effectiveness of the current combination of temperature control parameters: ； in, For the target temperature, This is the current real-time temperature. The fitness adjustment coefficient is used to adjust the sensitivity of the fitness function to adapt to different temperature ranges. S44. Through selection, crossover and mutation operations in the genetic algorithm, the combination of temperature control power allocation is continuously optimized, and the individual with the highest fitness value is selected as the basis for the next step of control parameters. S45. In each generation iteration, new individuals are generated using a genetic algorithm. And according to the fitness function The evaluation process is conducted, and when the fitness function reaches a preset threshold, the optimal combination of heating and cooling power is output. and ; S46. Optimal heating power output by the genetic algorithm. and optimal cooling power The operating status of heating and cooling equipment is dynamically adjusted.

5. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 1, characterized in that, S4 specifically includes: S431. Constructing the fitness function The fitness function is used to evaluate each individual. The combination of temperature control parameters; S432, Adjustment coefficient in fitness function Used to control the system's response sensitivity under different temperature differences, among which The larger the value, the more sensitive the fitness function is to small temperature differences; S433. In each algebraic iteration, through the fitness function Calculate the fitness value for each individual; a higher fitness value indicates a better current power combination. The closer the temperature gets to the target temperature, the more the fitness function output is used as the basis for selecting the optimal individual. S434. To accelerate algorithm convergence, the fitness values ​​are ranked based on the fitness function, and the top... The individual with the highest fitness value enters the next generation for crossover and variation genetic operations; S435. During the iterative process of the genetic algorithm, a dynamic fitness update mechanism is used. When the real-time temperature fluctuation is less than a specific value, the fitness level is adaptively reduced. The value optimizes the system's response to small temperature variations.

6. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 1, characterized in that, S6 specifically includes: S61. During the extraction process, the temperature data of the extract is acquired in real time. and extraction efficiency of active ingredients The extraction environment inside the extraction equipment is monitored by sensors, and the relationship between the extraction efficiency of active ingredients and temperature is recorded at each time point; S62. Construct an optimization model for extraction efficiency and temperature control using reinforcement learning algorithms, and set a reward function. : ； in, To improve real-time extraction efficiency, For real-time temperature, For the target temperature, and This is the adjustment coefficient; S63, Based on reward function Based on the feedback, the reinforcement learning algorithm dynamically adjusts the heating power. and cooling power Optimize temperature control strategy; S64. The reinforcement learning algorithm continuously learns from historical data and real-time feedback within each extraction cycle. Through accumulated feedback, it gradually optimizes the relationship between extraction efficiency and temperature control parameters, specifically by adjusting coefficients. and To adapt to the temperature control requirements under different extraction conditions, and to continuously optimize the operation strategies of heating and cooling equipment; S65. The reinforcement learning algorithm generates the optimal temperature control strategy curve based on real-time feedback and historical extraction data throughout the extraction process. .

7. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 6, characterized in that, S62 specifically includes: S621, Based on real-time temperature data during the extraction process and extraction efficiency of active ingredients Establish a reward function The nonlinear model, with a reward function used to evaluate the impact of the current temperature control strategy on extraction efficiency; S622. In the reward function model, use the adjustment coefficient. and To control the system's sensitivity and penalty for temperature deviations; S623. When the temperature deviation is less than a specific value, the system adjusts... Enhance the sensitivity of the reward function to extraction efficiency and optimize the response of the temperature control system to small temperature fluctuations; S624. When the temperature deviation exceeds a specific value, the adjustment coefficient... This is used to increase the penalty for temperature deviation, enabling the system to automatically adjust heating and cooling power when faced with large temperature fluctuations; S625, Reinforcement learning algorithm based on improved reward function Dynamically adjust heating power and cooling power ; S626. The system continuously optimizes the adjustment coefficient through self-learning. and This allows it to adapt to the temperature requirements of different plant extracts and extraction processes, generating the optimal temperature control strategy curve.

8. The method for dynamic temperature-controlled plant extraction based on an adaptive algorithm according to claim 6, characterized in that, Specifically, S64 includes: S641. Obtain real-time temperature data during the extraction process using a reinforcement learning algorithm. and extraction efficiency of active ingredients Simultaneously record historical extracted data ,in This includes historical temperature, extraction efficiency, and power allocation data; S642. The reinforcement learning algorithm performs self-learning based on real-time feedback and historical data, constructs a non-linear relationship model between temperature control parameters and extraction efficiency, and sets the optimization objective of self-learning as maximizing extraction efficiency. Minimize temperature deviation ; S643. Define the loss function Used to evaluate the effectiveness of the current temperature control strategy: ； in, For real-time temperature, For the target temperature, To improve real-time extraction efficiency, To achieve the expected maximum extraction efficiency, For balance coefficient, and This is an adjustment coefficient used to control the degree to which temperature deviation affects extraction efficiency. To avoid dividing by zero for small constants; S644, Based on Loss Function Based on feedback, the reinforcement learning algorithm continuously optimizes the heating power through self-learning. and cooling power Adjust the temperature control strategy; S645. The system dynamically updates the learning rate of the reinforcement learning algorithm based on historical data and real-time feedback. ; S646, Reinforcement learning algorithm generates optimal temperature control strategy curve .

Images