Polysaccharide extraction result prediction method, and multi-variable control method and device in component extraction process

Abstract

The invention discloses a polysaccharide extraction result prediction method and a multi-variable control method and device in a component extraction process, and relates to the technical field of industrial control. According to the method, the polysaccharide extraction prediction can be performed according to the parameter values of different target process parameters at each time point in the current time period to obtain the polysaccharide extraction rate, and the process parameters in the extraction process can be adaptively adjusted according to the polysaccharide extraction rate; the polysaccharide extraction rate obtained through prediction provides a basis for control in the polysaccharide extraction process, stable control over the polysaccharide extraction process can be achieved, the polysaccharide extraction efficiency is effectively improved, and material losses are reduced.

Description

Technical Field

[0001] The present application relates to the field of industrial control technology, and particularly to a method for predicting the results of polysaccharide extraction, a method for controlling various variables in the component extraction process, and a device therefor. Background Art

[0002] In the existing control process of polysaccharide extraction, the commonly adopted method is to perform real-time control on the control quantity during the control process. This control method often exhibits overshoot phenomena, which affects the quality of polysaccharide extraction. Moreover, in the existing control, there is no prediction of the results of polysaccharide extraction, and simply relying on the collected values and target values of process parameters for real-time control often results in large fluctuations in both concentration and purity.

[0003] The extraction process of polysaccharides is usually affected by various factors such as the ratio of material to liquid, extraction temperature, ultrasonic duration, pH value, ultrasonic intensity, etc. These factors are difficult to precisely control in actual production, resulting in fluctuations in the extraction rate, which in turn affects the extraction efficiency and the quality of the final product. Traditional polysaccharide extraction control methods mainly rely on empirical adjustment or manual monitoring to adjust process parameters, and have disadvantages such as low precision, poor stability, and unstable extraction results, making it difficult to achieve precise control in large-scale production. Therefore, how to achieve efficient and stable control during the extraction process has become a key issue for improving the extraction efficiency of polysaccharides and reducing resource waste. Summary of the Invention

[0004] The purpose of the present application is to provide a method for predicting the results of polysaccharide extraction, a method for controlling various variables in the component extraction process, and a device therefor, so as to achieve stable control of the polysaccharide extraction process and improve the quality of polysaccharide extraction.

[0005] To achieve the above object, the present application provides the following solutions.

[0006] In the first aspect, the present application provides a method for predicting the results of polysaccharide extraction, including: Obtaining the parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process; the target process parameters are obtained by means of calculating the degree of association; Inputting the parameter values of different target process parameters at each time point within the current time period into a trained multi-layer perceptron network model for polysaccharide extraction prediction to obtain the polysaccharide extraction rate.

[0007] In the second aspect, the present application provides a method for controlling various variables in the component extraction process, including: Using the above-mentioned method for predicting the results of polysaccharide extraction to predict the polysaccharide extraction rate during the component extraction process to obtain a predicted result of the polysaccharide extraction rate; When the absolute value of the difference between the predicted result of the polysaccharide extraction rate and the target value of the polysaccharide extraction rate is greater than the absolute value threshold, determine the control amount of each target process parameter according to the parameter values of different target process parameters at the current moment; Control the component extraction process according to the control amount of each target process parameter.

[0008] In a third aspect, the present application provides a device for controlling multiple variables in the component extraction process, including: a sensing signal acquisition system and a polysaccharide content prediction and control module; The sensing signal acquisition system is used to collect the parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process; The polysaccharide content prediction and control module is used to adopt the method for controlling multiple variables in the component extraction process described above to obtain the control amount of each target process parameter and control the polysaccharide extraction process.

[0009] According to the specific embodiments provided by the present application, the present application has the following technical effects.

[0010] The present application provides a method for predicting the polysaccharide extraction result, a method and device for controlling multiple variables in the component extraction process. The method for predicting the polysaccharide extraction result of the present application can predict the polysaccharide extraction according to the parameter values of different target process parameters at each time point within the current time period, obtain the polysaccharide extraction rate, and can adaptively adjust the process parameters of the extraction process according to the polysaccharide extraction rate. The predicted polysaccharide extraction rate of the present application provides a basis for the control in the polysaccharide extraction process, can achieve stable control of the polysaccharide extraction process, effectively improve the polysaccharide extraction efficiency and reduce material loss. Description of the Drawings

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0012] Figure 1 It is a schematic flowchart of a method for predicting the polysaccharide extraction result provided by an embodiment of the present application.

[0013] Figure 2 It is a schematic diagram of the principle of a polysaccharide extraction prediction method provided by an embodiment of the present application and its application in the control process.

[0014] Figure 3 It is a schematic diagram of the principle of a method for controlling multiple variables in the component extraction process provided by an embodiment of the present application.

[0015] Figure 4 The structural schematic diagram of a multi-variable control device in the process of ingredient extraction provided by an embodiment of the present application. Specific embodiments

[0016] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0017] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0018] In an exemplary embodiment, a method for predicting the polysaccharide extraction result is provided. As Figure 1 and Figure 2 shown, it includes the following steps 101 and 102.

[0019] Step 101, obtaining the parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process; the target process parameters are obtained by means of correlation degree calculation.

[0020] Step 102, inputting the parameter values of different target process parameters at each time point within the current time period into the trained multi-layer perceptron network model for polysaccharide extraction prediction to obtain the polysaccharide extraction rate.

[0021] Implementing the above steps 101 and 102 can realize the prediction of the polysaccharide extraction rate, determine whether the target process parameters within the current time period have an impact on the polysaccharide extraction result, and thus provide a basis for the control in the polysaccharide extraction process. The present application can achieve precise control of various target process parameters during the extraction process, and dynamically adjust the target process parameters according to real-time data and the empirical rules of the expert knowledge base, thereby improving the extraction efficiency, ensuring the product quality, reducing energy consumption, and reducing the need for manual intervention. It solves the technical problem that the results cannot be predicted in advance during the existing ingredient extraction process, resulting in large fluctuations in concentration and purity during the extraction process.

[0022] In another exemplary embodiment, the process parameters of the present application include temperature, pH value, ultrasonic intensity, solid-liquid ratio, ultrasonic duration, etc. However, the degree of influence of each process parameter on the extraction result (i.e., the polysaccharide extraction rate) is different during the extraction process. Understanding and quantifying the degree of influence of each process parameter on the polysaccharide extraction rate can not only help optimize the extraction process, improve efficiency, but also save resources, reduce costs, and enhance the accuracy and stability of the model. Therefore, the present application proposes a method for calculating the correlation degree between the polysaccharide extraction rate and process parameters, and effectively analyzes each process parameter as an influencing factor. The specific implementation method is as follows: ; wherein, 、 、 、 、 respectively represent the parameter sequences of extraction temperature, pH, ultrasonic intensity, solid-liquid ratio, and ultrasonic duration. The subscripts 、 、 、 、 respectively represent temperature, pH, solid-liquid ratio, ultrasonic intensity, and ultrasonic time. represents the number of historical time points. Let represent the comparison sequence composed of the parameter values of process parameter i from the 1st to the Kth historical time points. . When , they respectively correspond to which are respectively 、 、 、 、 , respectively correspond to 、 、 、 、 .

[0023] In order to unify the influencing factors of the parameters, the present application performs dimensionality unification processing. The specific processing method is as follows: ; wherein, is the parameter value of process parameter i at the kth historical time point, is the normalized parameter value of process parameter i at the th historical time point, is the parameter value of process parameter i at the th historical time point.

[0024] Calculate the correlation coefficient between each process parameter and the polysaccharide extraction rate.

[0025] Set the polysaccharide extraction rate in the polysaccharide extraction process as the reference sequence, then the reference sequence is: ; The comparison sequence after dimension unification processing is: ; Then the resolution coefficient is: ; Where: is the resolution coefficient between the process parameter i at the kth historical time point and the polysaccharide extraction rate; is the polysaccharide extraction rate at the th historical time point, is the normalized parameter value of the process parameter i at the th historical time point, K is the number of historical time points, is the polysaccharide extraction rate at the th historical time point, is the resolution constant, with a value between 0 and 1, indicating the similarity between sequences. The closer the value is to 1, the more similar the reference sequence and the comparison sequence are; the closer the value is to 0, the greater the difference between the reference sequence and the comparison sequence. Usually, = 0.5 represents a medium degree of resolution.

[0026] Calculate the correlation coefficient as: ; Where, is the correlation coefficient between the process parameter i and the polysaccharide extraction rate. The larger the value, the more similar the reference sequence and the comparison sequence are.

[0027] In another exemplary embodiment, step 101 above specifically includes the following steps: Obtain the sensor signal of the target process parameter in the current time period; Perform noise reduction on the sensor signal to obtain the noise-reduced sensor signal; Perform sampling and analog-to-digital conversion on the noise-reduced sensor signal to obtain the parameter values of the target process parameter at each time point in the current time period.

[0028] The accuracy and precision of the sensor data acquisition will directly affect the prediction accuracy. Therefore, this application proposes a signal processing method for the sensing acquisition process to avoid interference such as noise and drift during the signal acquisition process by the sensor. The above-mentioned acquired sensor signal is: ; Among them, , , , , are the parameter values of the process parameters actually collected at time t, , , , , respectively, is the sensor signal representing the parameter values of each process parameter, is the output characteristic coefficient of the sensor, is the corresponding function between the process parameter and the sensor signal, which can be obtained by model training or function fitting.

[0029] The extracted sensor signal is: ; Among them, as a whole represents the signal impulse response collected by the sensor, is the sensor signal after noise reduction, is the sensor signal, is the integration variable.

[0030] Through the above signal processing method, the collected sensing information is used for the prediction of the extraction rate, which can effectively improve the accuracy.

[0031] In another exemplary embodiment, the above target process parameters include but are not limited to temperature, pH value, ultrasonic intensity, liquid-to-material ratio, and ultrasonic duration. The above target process parameters are all collected in real time by the sensor module. The parameter sequences of the temperature, pH value, ultrasonic intensity, liquid-to-material ratio, and ultrasonic duration collected by the sensor module are respectively . Through the above multi-layer perceptron network model, the polysaccharide extraction rate is obtained by performing perceptual calculations on the target process parameters, and its formal definition is: ; Among them, , , , , respectively represent the parameter sequences of the extraction temperature, pH, ultrasonic intensity, liquid-to-material ratio, and ultrasonic duration.

[0032] In another exemplary embodiment, the specific implementation manner of the above step 102 is: Taking each target process parameter as an input parameter, these input parameters are collected in real time by the sensor module and used as the input of the multi-layer perceptron network model.

[0033] Therefore, the input vector for constructing the multi-layer perceptron network model is: ; Create a multi-layer perceptron network model to achieve the prediction of the polysaccharide extraction rate. This multi-layer perceptron network model includes an input layer, an output layer, and a hidden layer.

[0034] Among them, the output of the th neuron in the th hidden layer is expressed as: ; Among them, is the weight from the th neuron in the th hidden layer to the th neuron in the th hidden layer, is the output of the th neuron in the th hidden layer, is the bias of the th neuron in the th hidden layer; is the activation function, is the number of neurons in the th hidden layer.

[0035] The finally obtained extraction rate is calculated by the output layer, and its calculation method is: ; Among them, is the weight from the th neuron in the th hidden layer to the output layer; is the output of the th neuron in the th hidden layer, is the bias of the output layer, is the total number of hidden layers, is the number of neurons in the th hidden layer.

[0036] When the target process parameters (temperature, pH value, ultrasonic intensity, solid-liquid ratio, ultrasonic duration) are obtained in real time through the sensor module, input these parameters into the trained multi-layer perceptron network model, and the predicted polysaccharide extraction rate can be obtained.

[0037] In another exemplary embodiment, the present application provides a method for controlling multiple variables in the component extraction process, such asFigure 3 The steps shown include the following steps 201 - step 203.

[0038] Step 201, using the above - mentioned polysaccharide extraction result prediction method, predict the polysaccharide extraction rate during the component extraction process to obtain the polysaccharide extraction rate prediction result.

[0039] Step 202, when the absolute value of the difference between the polysaccharide extraction rate prediction result and the polysaccharide extraction rate target value is greater than the absolute value threshold, determine the control quantity of each target process parameter according to the parameter value of each target process parameter at the current moment.

[0040] Step 203, control the component extraction process according to the control quantity of each target process parameter.

[0041] Implementing the above steps 201 - step 203 realizes the stable control of the target process parameters during the polysaccharide extraction process.

[0042] In another exemplary embodiment, the above - mentioned control process can adopt a fuzzy control algorithm to simultaneously achieve the global control of temperature, pH value, ultrasonic intensity, solid - liquid ratio, and ultrasonic duration. The set control variables are temperature ( ), pH value ( ), ultrasonic intensity ( ), solid - liquid ratio ( ), and ultrasonic duration ( ). For each control variable , set its fuzzy rule set , and the output of the control system is the concentration of the target component , that is, the polysaccharide extraction rate.

[0043] The specific implementation scheme is as follows: When the prediction result is obtained, the feedback control module is started to effectively control the above - mentioned five target process parameters. Let the input variables of the control system be: ; Among them, each control variable in the input vector is a fuzzy linguistic variable. That is, initially, the system is not sure how to adjust the system. For example, when the temperature is 60 °C, the system receives an instruction to heat the temperature in the extraction tank to 70 °C. However, at this time, the heating system does not know how long to heat to reach 70 °C, and the adjustment of the other four indicators is similar. The input variables are sensing signals collected by multiple sensors, and the output quantity is the control instruction calculated by the control system in the feedback adjustment module according to the input variables.

[0044] In an exemplary embodiment, the determination of the control quantity is achieved through a fuzzy control algorithm.

[0045] Through the calculations of the above steps, the control system can effectively achieve the individual and effective control of five control variables, namely temperature, pH value, ultrasonic duration, ultrasonic intensity, and material-liquid ratio, further improving the efficiency of the overall process flow and the product quality. It not only increases the yield and purity of the target product but also reduces energy consumption and production costs, ultimately realizing an efficient and sustainable production process.

[0046] In another exemplary embodiment, the above control process can adopt a single-input single-output model.

[0047] When the prediction result indicates that the extraction result cannot meet the target, the feedback adjustment module operates to adjust its process parameters. However, it is an ideal state that each process parameter is adjusted and there is no coupling interaction effect between the process parameters. In actual work, when the feedback adjustment module adjusts the five process parameters of temperature, pH value, ultrasonic duration, ultrasonic intensity, and material-liquid ratio, a change in one of these five process parameters will affect one or more of the others to change. For example, when adding acid or base solution to adjust the pH, heat will be generated, affecting the change in temperature. Therefore, this application proposes a multivariable decoupling method for polysaccharide extraction process control, which can control a single process parameter without affecting the changes of other parameters, that is, in the feedback adjustment module, the static values of the corresponding control variables can effectively control the static values of the individual output control variables alone.

[0048] Among them, the construction method of the single-input single-output model is as follows: In the component extraction process of this application, its multivariable coupling control model is expressed as: ; Among them: is the input vector, is the transfer function matrix, is the input quantity.

[0049] ; ; Among them, , , are respectively the parameter values of the 1st, 2nd, and nth process parameters at the current moment s, , , are respectively the control quantities of the 1st, 2nd, and nth process parameters at the current moment s, is the correlation relationship between the parameter value of the process parameter at the current moment s and the control quantity of the process parameter at the current moment s, , , is the number of process parameters. In the embodiments of the present application, .

[0050] In the next step, the present application converts the five-dimensional multivariable coupling control model into 5 multi-input single-output models, where the multi-input single-output model of process parameter i is as follows: ; In the present application, 5 inputs will affect the output of the system. Therefore, the present application selects a single process parameter as the main research object. In this way, the remaining four process parameters become the interference of this process parameter. By using this method, each multi-input single-output model is decomposed into 5 single-input single-output models.

[0051] For the decomposed single-input single-output models, there is no longer any coupling cross-influence between them. A separate control method is used to control the process parameters. If the temperature is cold, the power of heater 11 is increased; if the pH value is acidic, an alkaline substance is added; if the ultrasonic intensity is weak, the ultrasonic intensity is increased; if the liquid ratio is low, the liquid ratio is increased; if the ultrasonic duration is short, the ultrasonic time is extended. Otherwise, the opposite operation is performed.

[0052] In another exemplary embodiment, a method for scoring a control method is provided, which can realize the evaluation of the above control method.

[0053] Based on multiple indicators, a comprehensive judgment and a judgment on the superiority and inferiority of conditions are made on the control results of various variables in the polysaccharide extraction process. This method mainly considers four aspects: extraction rate, purity, economy, and low carbon. By determining the influence coefficients of each indicator, the control results of various variables are scored, and finally a comprehensive score of the control results of various variables is obtained. Among them, the comprehensive score is represented by S as: ; Among them, , , , are the extraction rate, purity, economy, and low carbon respectively, are the influence coefficients of the extraction rate, purity, economy, and low carbon respectively.

[0054] The embodiments of the present application consider the problem of inconsistent units of the four indicators (extraction rate , purity , economy and low carbon ), and perform standardization processing on each indicator. The specific processing method is as follows: The extraction rate of the polysaccharide is expressed as a percentage, and its maximum value cannot exceed the preset maximum value. : ; is the percentage representation of the extraction rate.

[0055] The value range of the purity is usually [0, 1], and after processing, it is expressed as: ; Among them, is the processed purity, is the maximum purity.

[0056] The economy and low carbon are evaluated by experts. After averaging the scores of the experts, the standardization method is: ; Among them, and are the economy and low carbon after averaging respectively, and are the minimum and maximum values of the economy respectively, and are the minimum and maximum values of the low carbon respectively.

[0057] The representation of the final standardized comprehensive score is: .

[0058] Finally, according to the grading standard, the standardized comprehensive score is divided into four grades, and the grading standard is as follows: , indicating poor (low efficiency, high cost, poor low carbon); , indicating medium (relatively low efficiency, relatively high cost, relatively poor low carbon); , indicating good (relatively high efficiency, relatively low composition, relatively good low carbon); , indicating very good (high efficiency, low composition, good low carbon).

[0059] According to the above evaluation method, this application effectively evaluates the prediction result.

[0060] In another exemplary embodiment of this application, a multi-variable control device during the component extraction process is provided, including: a sensing signal acquisition system and a polysaccharide content prediction and control module.

[0061] The sensing signal acquisition system is used to acquire the parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process.

[0062] The polysaccharide content prediction and control module is used to adopt the multiple variable control methods in the above-mentioned component extraction process to obtain the control quantity of each target process parameter and control the polysaccharide extraction process.

[0063] As Figure 4 shown, in another exemplary embodiment, the above-mentioned sensing signal acquisition system includes: a material-liquid ratio measurement module 2, a pH measurement module 3, an ultrasonic intensity measurement module 4, an ultrasonic time measurement module 5, and a temperature measurement module 6. The above-mentioned polysaccharide content prediction and control module includes a signal processing module 1, an intelligent calculation module, a feedback adjustment module, and an expert knowledge base.

[0064] The above-mentioned multiple variable control device is applied to an automatic extraction device, which includes a tank body 7, a bracket 8, a discharge port 9, a ladder 10, a heater 11, and a handle 12.

[0065] The sensing signal acquisition system mainly includes a material-liquid ratio measurement module 2, a pH measurement module 3, an ultrasonic intensity measurement module 4, an ultrasonic time measurement module 5, and a temperature measurement module 6. Through these sensors, important process parameters such as temperature, pH value, ultrasonic intensity, material-liquid ratio, and ultrasonic duration during the extraction process, that is, the target process parameters, can be monitored in real time to ensure that the extraction environment always remains in an ideal state. The above-mentioned sensing signal acquisition system is installed in an automated extraction device. Among them, the temperature measurement module 6 is used to detect the temperature based on a temperature sensor. The temperature sensor is installed in the extraction tank, and a heating system and a cooling system are installed in the automated extraction device. The heating or cooling is adjusted according to the temperature information feedback by the temperature sensor to ensure a stable extraction environment in the extraction tank; the pH measurement module 3 is used to detect the pH value based on a pH sensor. The pH sensor is installed in the flow path of the extraction liquid to monitor the change of the pH value of the extraction liquid in real time and maintain the acidity and alkalinity by automatically adjusting the acid-base adjustment module; the ultrasonic intensity measurement module 4 is used to detect the ultrasonic intensity based on a power meter. The power meter is installed on the ultrasonic generator. By measuring the output power of the ultrasonic wave, the ultrasonic intensity is indirectly measured, and the power of the ultrasonic generator is automatically adjusted to achieve the adjustment of the ultrasonic intensity; the material-liquid ratio measurement module 2 is used to detect the material-liquid ratio based on a material weight sensor and a liquid level sensor. The material weight sensor and the liquid level sensor are installed on the automated extraction device. By monitoring the weight of the raw material and the liquid level of the solvent respectively, the calculation of the material-liquid ratio is realized to ensure the consistency and accuracy of the material-liquid ratio; the ultrasonic time measurement module 5 is used to detect the ultrasonic duration based on an ultrasonic treatment timer. The ultrasonic treatment timer is installed on the automated extraction device, and the opening and closing times of the ultrasonic generator are automatically adjusted according to the duration to ensure its best extraction effect; the signal processing module 1 stores data preprocessing methods to filter, remove outliers, suppress noise, and correct data for the collected sensor signals, providing accurate and reliable data for the subsequent multi-layer perceptron network model; the expert knowledge base contains prior model data during the polysaccharide extraction process. When the polysaccharide extraction process starts, the prior model data is transmitted to the feedback adjustment module. The feedback adjustment module adjusts and controls the material-liquid ratio measurement module 2, the pH measurement module 3, the ultrasonic intensity measurement module 4, the ultrasonic time measurement module 5, and the temperature measurement module 6, specifically including: first, obtaining the prior model data in the expert knowledge base, predicting the polysaccharide extraction result based on the extracted temperature, pH, material-liquid ratio, ultrasonic intensity, and ultrasonic time, and comparing the prediction result with the prior model data. When there is a deviation between the prediction result and the prior model data, the intelligent calculation module will recalculate the extraction process parameters and inform the feedback adjustment module. The feedback adjustment module adjusts the heating system, cooling system, ultrasonic generator, acid-base adjustment module, etc. according to the obtained instructions to ensure that all indicators during the polysaccharide extraction process maintain the commanded state.

[0066] In another exemplary embodiment, an alarm module is further provided in the various variable control devices in the above-mentioned component extraction process. The warning module is used to give a warning when the parameter values of any one or more target process parameters at each time point within the current time period meet the warning conditions; the warning conditions are that the parameter values of the target process parameters at one or more time points within the current time period are not within the preset range corresponding to the target process parameters. The specific principle is as follows: When the warning module detects an abnormal situation, it can automatically trigger an alarm to prompt the operator to handle it in time. Temperature has a significant impact on the extraction effect. Too high or too low temperature may cause the degradation of polysaccharides or low extraction efficiency; the pH value affects the solubility and stability of polysaccharides, so the pH value should be ensured to be within a suitable range; the ultrasonic intensity directly affects the extraction effect. Too strong ultrasonic waves may cause overheating of the solvent and even damage the polysaccharide structure; while too weak ultrasonic intensity may lead to poor extraction effect; too low solid-liquid ratio may cause too high solvent concentration and reduce the extraction efficiency; while too high solid-liquid ratio may lead to a decrease in the ultrasonic efficiency; too long or too short ultrasonic duration will affect the extraction effect. Too long ultrasonic duration may cause overheating and degradation of the extract, and too short duration may lead to insufficient extraction. Therefore, the warning principle of the warning module is as follows: Set a minimum threshold and a maximum threshold , and the warning formula is: ; If the current state exceeds the set range , then set the warning signal to 1 and trigger a warning; otherwise, no warning is triggered.

[0067] In the above formula, the parameter contains the normalized parameter values of process parameters such as extraction temperature, pH value, ultrasonic intensity, solid-liquid ratio, ultrasonic duration, etc.

[0068] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0069] In this article, specific examples are used to elaborate on the principles and implementation methods of this application. The descriptions of the above embodiments are only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A method for predicting the results of polysaccharide extraction, characterized in that, Comprising: Obtaining parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process; The target process parameters are obtained by means of correlation degree calculation; Inputting the parameter values of different target process parameters at each time point within the current time period into a trained multi-layer perceptron network model for polysaccharide extraction prediction to obtain the polysaccharide extraction rate.

2. The polysaccharide extraction result prediction method according to claim 1, wherein The specific obtaining method of the target process parameters includes: Obtaining parameter values of different process parameters at each historical time point during the polysaccharide extraction process and the polysaccharide extraction rate at each historical time point; Performing normalization processing on the parameter values of different process parameters at each historical time point to obtain the normalized parameter values of different process parameters at each historical time point; Calculating the correlation coefficient between each process parameter and the polysaccharide extraction rate according to the normalized parameter values of different process parameters at each historical time point and the polysaccharide extraction rate at each historical time point; Selecting the process parameters with a correlation coefficient greater than a preset coefficient threshold as the target process parameters.

3. The polysaccharide extraction result prediction method according to claim 2, characterized in that The formula for calculating the correlation coefficient is: ； ； Among them, is the correlation coefficient between the process parameter i and the polysaccharide extraction rate, is the discrimination coefficient between the process parameter i and the polysaccharide extraction rate at the k-th historical time point; is the polysaccharide extraction rate at the -th historical time point, is the normalized parameter value of the process parameter i at the -th historical time point, K is the number of historical time points, is the discrimination constant, is the polysaccharide extraction rate at the -th historical time point, is the normalized parameter value of the process parameter i at the -th historical time point.

4. The polysaccharide extraction result prediction method according to claim 1, wherein Obtaining parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process specifically includes: Obtaining the sensor signal of the target process parameter within the current time period; Denosing the sensor signal to obtain the denoised sensor signal; Sampling and performing analog-to-digital conversion on the denoised sensor signal to obtain the parameter values of the target process parameter at each time point within the current time period.

5. A method for controlling multiple variables in the process of ingredient extraction, characterized in that, Comprising: Using the polysaccharide extraction result prediction method described in any one of claims 1-4 to predict the polysaccharide extraction rate during the component extraction process to obtain a polysaccharide extraction rate prediction result; When the absolute value of the difference between the polysaccharide extraction rate prediction result and the target value of the polysaccharide extraction rate is greater than the absolute value threshold, determining the control amount of each target process parameter according to the parameter values of different target process parameters at the current moment; Controlling the component extraction process according to the control amount of each target process parameter.

6. The method for controlling multiple variables in the component extraction process according to claim 5, characterized in that, Determining the control amount of each target process parameter according to the parameter values of different target process parameters at the current moment specifically includes: Determining the control amount of each target process parameter according to the parameter values of different target process parameters at the current moment by using a fuzzy control algorithm.

7. The method for controlling multiple variables in the component extraction process according to claim 5, characterized in that, Determining the control amount of each target process parameter according to the parameter values of different target process parameters at the current moment specifically includes: Determining the control amount of each target process parameter according to the parameter values of different target process parameters at the current moment based on the single-input single-output model of each target process parameter.

8. The method for controlling multiple variables in the component extraction process according to claim 7, characterized in that The single-input single-output model is constructed in the following manner: Constructing a multivariable coupling control model for the component extraction process; Converting the multivariable coupling control model into a multi-input single-output model according to the coupling relationship between the input and the output; Respectively taking each input in the multi-input single-output model as the target input and taking the inputs other than the target input as disturbances, and decomposing the multi-input single-output model to obtain a plurality of single-input single-output models.

9. A device for controlling multiple variables during a component extraction process, characterized in that, Comprising: A sensing signal acquisition system and a polysaccharide content prediction and control module; The said sensing signal acquisition system is used to acquire the parameter values of different target process parameters at each time point within the current time period during the polysaccharide extraction process; The said polysaccharide content prediction and control module is used to adopt the multiple variable control methods in the component extraction process described in any one of claims 5-8 to obtain the control quantity of each target process parameter and control the polysaccharide extraction process.

10. The multi-variable control device in the component extraction process according to claim 9, characterized in that, It further includes: An early warning module; The said early warning module is used to give an early warning when the parameter values of any one or more target process parameters at each time point within the current time period meet the early warning conditions; The early warning condition is that the parameter values of the target process parameter at one or more time points within the current time period are not within the preset range corresponding to the target process parameter.

Images