An adaptive electrical control method and system for supercritical extraction

Abstract

The application discloses a kind of self-adapting electric control method and system for supercritical extraction.The method comprises the following steps: using multi-dimensional sensing assembly to collect the real-time operating parameters such as component concentration, pressure, temperature and fluid flow rate in supercritical extraction equipment, to obtain real-time detection sequence; constructing component-state correlation model of extraction process to generate reference control parameter library corresponding to different supercritical processes; based on real-time detection sequence and reference control parameter library, deviation analysis is carried out to generate parameter deviation time series data, and calibrated deviation parameters are obtained through dynamic threshold deviation calibration; according to the calibrated deviation parameters, an adaptive control strategy is constructed to adjust the frequency, amplitude and duration of the output signal of the electric control module, and to drive the actuator to act; feedback data after action is collected, and control parameters are iteratively optimized.The application realizes multi-parameter collaborative detection and dynamic threshold self-adaptive calibration, and improves the response speed and process stability of electric control.

Description

Technical Field

[0001] This invention relates to the field of electrical control technology, and in particular to an adaptive electrical control method and system for supercritical extraction. Background Technology

[0002] In recent years, with the widespread application of supercritical fluid extraction (SFE) technology in food, pharmaceutical, and chemical industries, the process precision, operational stability, and product extraction efficiency of the extraction process have become core industry requirements. This places higher demands on the adaptive adjustment capability, parameter control accuracy, and multi-parameter coordinated matching of the electrical control system for the extraction process. Adaptive electrical control methods for supercritical fluid extraction, with their unique advantages of multi-dimensional parameter coordinated detection, dynamic deviation calibration, adaptive strategy adjustment, and closed-loop iterative optimization, are gradually replacing traditional fixed parameter control and manual adjustment modes, becoming an important technical means to ensure the stability of the supercritical fluid extraction process and improve the quality and efficiency of product extraction. Currently, some methods related to electrical control for supercritical fluid extraction have been proposed. These methods mostly collect some operating parameters through a single sensor, adjust the electrical control output according to a preset fixed threshold, and then correct the control parameters by manually observing the process status. However, existing methods cannot comprehensively obtain real-time changes in key parameters such as component concentration, pressure, and temperature of the extraction system, and do not perform dynamic threshold deviation calibration, which easily leads to lag in control signal adjustment. Summary of the Invention

[0003] The present invention aims to provide an adaptive electrical control method and system for supercritical extraction, in order to solve the technical problem of lagging control signal adjustment caused by incomplete parameter acquisition and reliance on fixed thresholds for deviation judgment in the prior art, thereby realizing multi-parameter collaborative detection and dynamic threshold adaptive calibration in the supercritical extraction process, and improving the response speed and process stability of electrical control.

[0004] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: On one hand, the present invention provides an adaptive electrical control method for supercritical extraction, comprising the following steps: Step S1: Use multi-dimensional sensing components to collect real-time operating parameters in the supercritical extraction equipment to perform collaborative detection of component characteristics and operating status, and obtain a real-time detection sequence including the component concentration, system pressure, temperature and fluid flow rate of the extraction system; construct a component-state correlation model of the extraction process to generate a reference control parameter library corresponding to different supercritical processes; Step S2: Perform deviation analysis based on the real-time detection sequence and the reference control parameter library to generate parameter deviation time series data; perform dynamic threshold deviation calibration based on the parameter deviation time series data to obtain the calibrated deviation parameters; Step S3: Construct an adaptive control strategy based on the calibrated deviation parameters, and adjust the output signal of the electrical control module based on the adaptive control strategy to drive the actuator of the supercritical extraction equipment. Step S4: Collect equipment operating parameters and combination change data after the actuator moves in real time, and feed them back to the control strategy adjustment module to iteratively optimize the control parameters, thereby realizing adaptive electrical control of the supercritical extraction process.

[0005] On the other hand, the present invention provides an adaptive electrical control system for supercritical extraction that performs the above-described method, including a multi-dimensional sensing component, an electrical control module, an actuator, a data processing module, a reference parameter storage module, and a feedback adjustment module. The multi-dimensional sensing component is used to collect real-time operating parameters and component data within the supercritical extraction equipment, generate raw detection data, and transmit it to the data processing module. The data processing module is used to parse and fuse the raw detection data to generate a real-time detection sequence, and simultaneously retrieve the reference control parameter library in the reference parameter storage module for deviation analysis and calibration, generating calibrated deviation parameters and transmitting them to the electrical control module. The reference parameter storage module is used to store the reference control parameter library corresponding to different supercritical processes and the component-state correlation model of the extraction process. The electrical control module is used to construct an adaptive control strategy based on the calibrated deviation parameters, generate an electrical control output signal, and transmit it to the actuator. The actuator is used to receive the electrical control output signal, execute corresponding control actions, and adjust the operating state of the supercritical extraction equipment. The feedback adjustment module is used to collect feedback data after the actuator's actions, transmit it to the data processing module, drive iterative optimization of control parameters and control strategies, and realize adaptive electrical control of the supercritical extraction process.

[0006] Compared with the prior art, the present invention has the following significant advantages: 1. Comprehensive data acquisition and process-specific modeling: Real-time detection sequences fully present the dynamic fluctuation characteristics of each parameter during the extraction process, providing comprehensive and accurate basic data support for control and regulation, and avoiding control decision deviations caused by incomplete parameter acquisition; at the same time, the constructed component-state correlation model and benchmark control parameter library of the extraction process integrate the parameter matching rules corresponding to different supercritical processes, breaking the limitation of lack of process-specificity in parameter setting in traditional control, and providing a scientific reference for deviation analysis and control strategy formulation.

[0007] 2. Dynamic Threshold Deviation Calibration: Abandoning the traditional fixed threshold mode, it can adjust the deviation judgment standard in real time according to the actual dynamic fluctuation of parameters during the extraction process, effectively avoiding the problems of inaccurate deviation identification and control signal adjustment caused by fixed thresholds; the calibrated deviation parameters truly reflect the actual deviation of the extraction process, providing accurate and reliable deviation basis for the construction of adaptive control strategies, reducing process fluctuations caused by deviation judgment errors, and improving the adaptability of electrical control.

[0008] 3. Adaptive Control Strategy and Precise Drive: By adjusting the output signal of the electrical control module based on an adaptive control strategy, precise drive of the actuator is achieved, ensuring that the actuator's actions are highly matched with the actual needs of the extraction process. This avoids the problems of slow response and large adjustment errors that exist in manual adjustment. It also corrects parameter deviations in the extraction process in a timely manner, effectively suppresses process fluctuations, ensures that the extraction system is always in a reasonable operating range, and improves product extraction efficiency and quality.

[0009] 4. Closed-loop feedback and iterative optimization: Feedback data is transmitted to the control strategy adjustment module to promptly identify deficiencies in the control strategy execution process, enabling iterative optimization of control parameters. This ensures that the control strategy remains synchronized with the dynamic changes in the extraction process, further improving the adaptability and stability of the electrical control.

[0010] 5. Long-term operating results: The closed-loop iterative optimization mode can continuously correct control deviations, avoid problems such as decreased process stability and fluctuations in product extraction efficiency caused by the solidification of control parameters, effectively make up for the shortcomings of low efficiency and poor accuracy of manual correction of control parameters in existing methods, and can continuously improve the stability of the supercritical extraction process and the consistency of product extraction in the long term. Attached Figure Description

[0011] The present invention will now be described with reference to the accompanying drawings.

[0012] Figure 1 This is a schematic flowchart of the adaptive electrical control method for supercritical extraction according to the present invention. Figure 2 for Figure 1 A detailed flowchart of step S1; Figure 3 for Figure 1 A detailed flowchart of step S2; Figure 4 for Figure 3 A detailed flowchart of step S24; Figure 5 for Figure 1 A detailed flowchart of step S3; Figure 6 for Figure 1A detailed flowchart of step S4; Figure 7 for Figure 6 A detailed flowchart of step S44. Detailed Implementation

[0013] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] like Figure 1 The diagram shown is a flowchart illustrating the steps of the adaptive electrical control method for supercritical extraction according to the present invention. In this example, the adaptive electrical control method for supercritical extraction includes the following steps: Step S1: Use multi-dimensional sensing components to collect real-time operating parameters in the supercritical extraction equipment to perform collaborative detection of component characteristics and operating status, and obtain a real-time detection sequence including the component concentration, system pressure, temperature and fluid flow rate of the extraction system; construct a component-state correlation model of the extraction process to generate a reference control parameter library corresponding to different supercritical processes; In this embodiment of the invention, multi-dimensional sensing components are controlled to initiate collaborative detection at key points in the extraction chamber, separation chamber, and fluid transmission pipeline of the supercritical extraction equipment. This covers various processes including supercritical extraction, supercritical drying, and supercritical anhydrous dyeing, simultaneously collecting component characteristics and operational status data. The data collection process is conducted at fixed intervals: pressure data is collected every second in MPa, temperature data is collected in real-time in °C, and flow rate data is collected synchronously in m / s. Component concentration is calculated by combining the collected system spectral data with the concentration-intensity formula. For example, if the spectral intensity is 28 and the characteristic peak coefficient is 5.6, the calculated component concentration is 28 ÷ 5.6 = 5 g / L. All collected data are integrated in timestamp order to generate a real-time detection sequence containing the component concentration, system pressure, temperature, and fluid flow rate of the extraction system. Simultaneously, standard parameters were collected for each of the three processes to analyze the coupling relationship between component concentration and pressure, temperature, and flow rate, and a component-state correlation model was established. For example, the correlation formula for supercritical extraction is component concentration = (pressure × temperature) ÷ (flow rate × 0.8). Based on the model, the standard parameters of each process were classified and coded to generate a benchmark control parameter library. The library contains the component thresholds and benchmark values ​​of operating parameters for each process, ensuring that there is a clear basis for comparison in subsequent deviation analysis.

[0015] Step S2: Perform deviation analysis based on the real-time detection sequence and the reference control parameter library to generate parameter deviation time series data; perform dynamic threshold deviation calibration based on the parameter deviation time series data to obtain the calibrated deviation parameters; In this embodiment of the invention, by extracting real-time detection sequences and a reference control parameter library, spatiotemporal alignment is performed using timestamps as the core. Real-time parameters at the same timestamp and the same detection point are matched one-to-one with reference parameters, and data with non-overlapping timestamps are eliminated. The deviation value of each parameter is calculated one by one. The calculation logic is the real-time parameter value minus the reference parameter value. For range-type reference parameters, the difference between the real-time value and the midpoint of the range is taken. For example, if the real-time pressure is 10.3 MPa, the pressure reference range is 8-10 MPa, and the midpoint is 9 MPa, the pressure deviation = 10.3 - 9 = 1.3 MPa; if the real-time component concentration is 5.8 g / L, the reference value is 5 g / L, and the concentration deviation = 5.8 - 5 = 0.8 g / L. The parameter deviation time series data is then generated according to the timestamps. Fluctuation characteristics are extracted based on time-series data of parameter deviations. The rate of change, fluctuation period, and peak characteristics are calculated to construct a dynamic threshold calibration model. For example, if the concentration deviation peak is 0.8 g / L, the rate of change is 0.2 g / (L·s), and the stability coefficient is 0.5, the dynamic threshold is calculated as 0.8 × (1 + 0.2 ÷ 0.5) = 1.12 g / L. Each deviation value is compared with the corresponding dynamic threshold, and abnormal data exceeding the threshold are removed. Gaps are filled by averaging adjacent valid data to generate calibrated deviation parameters, ensuring the continuity and reliability of the deviation data.

[0016] Step S3: Construct an adaptive control strategy based on the calibrated deviation parameters, and adjust the output signal of the electrical control module based on the adaptive control strategy to drive the actuator of the supercritical extraction equipment. In this embodiment of the invention, deviation parameters after calibration are extracted, classified and analyzed according to parameter type, and the deviation values, trends, and durations of component concentration, pressure, temperature, and flow rate are clarified. The deviation level is determined using a deviation level assessment system. For example, a pressure deviation of 1.3 MPa and a duration of 4 seconds is classified as a level three deviation; a component concentration deviation of 0.8 g / L and a duration of 3 seconds is classified as a level two deviation. Based on the constructed component-state correlation model of the extraction process, the relationship between deviations and operating parameters is analyzed, and adjustment parameters are optimized. For example, a level three pressure deviation of 1.3 MPa indicates that the pressure regulating valve opening needs to be reduced by 16%; a level two component concentration deviation of 0.8 g / L indicates that the temperature needs to be increased by 1.6°C and the flow rate reduced by 0.12 m / s. An adaptive control strategy is constructed by integrating adjustment parameters, clarifying the adjustment sequence, amplitude, and duration, and converting the strategy into output signals that can be recognized by the electrical control module. For example, the pressure regulating valve control signal has a frequency of 50Hz, an amplitude of 6.8V, and a duration of 8 seconds, while the temperature controller signal has a frequency of 50Hz, an amplitude of 5.2V, and a duration of 10 seconds. The signals are then transmitted to the corresponding actuators to drive them to move synchronously.

[0017] Step S4: Collect equipment operating parameters and combination change data after the actuator moves in real time, and feed them back to the control strategy adjustment module to iteratively optimize the control parameters, thereby realizing adaptive electrical control of the supercritical extraction process.

[0018] In this embodiment of the invention, after the actuator completes the control action, it immediately controls the multi-dimensional sensing components to start real-time acquisition. The acquisition range is consistent, collecting equipment operating parameters and component change data after the actuator's action, such as pressure of 9.2 MPa, temperature of 41.6℃, flow rate of 0.98 m / s, and component concentration of 5.2 g / L after control. The data is organized according to timestamps, and abnormal data is removed by a data consistency detection algorithm to generate feedback data. The feedback data is transmitted to the control strategy adjustment module, and the feedback deviation is calculated by comparing it with the benchmark control parameter library. For example, the pressure feedback deviation is 9.2-9=0.2 MPa, and the component concentration feedback deviation is 5.2-5=0.2 g / L. Combining the calibrated deviation parameters and deviation level parameters, a control parameter optimization model is constructed to calculate the adjustment range. For example, if the pressure feedback deviation is 0.2 MPa, the amplitude of the pressure regulating valve control signal is adjusted to 5.3V and the duration is adjusted to 6 seconds; if the component concentration feedback deviation is 0.2 g / L, the amplitude of the temperature controller signal is adjusted to 4.9V and the duration is adjusted to 7 seconds. The adaptive control strategy is optimized, and the adjusted control parameters are fed back to the electrical control module. The control action and feedback detection process are repeated, and the optimization is continuously iterated until the deviation returns to the normal range, so as to realize the adaptive electrical control of the supercritical extraction process.

[0019] Furthermore, as an embodiment of the present invention, such as Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps: Step S11: Control the multi-dimensional sensing components to perform full-domain detection of the extraction chamber, separation chamber and fluid transmission pipeline of the supercritical extraction equipment, and collect raw detection data including system component spectral data, pressure fluctuation data, continuous temperature data and instantaneous fluid flow rate data; In this embodiment of the invention, a multi-dimensional sensing component is used to complete the full-domain detection and raw detection data acquisition of the supercritical extraction equipment, covering the core components of various supercritical technology equipment such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing. The multi-dimensional sensing component is controlled to start operation, performing full-domain detection of the extraction chamber, separation chamber, and fluid transmission pipeline of the supercritical extraction equipment. The detection range covers the inner cavity of the chamber, the inner wall of the pipeline, and all connection nodes, ensuring no blind spots in detection. The collected raw test data includes system component spectral data, pressure fluctuation data, continuous temperature data, and instantaneous fluid flow velocity data. System component spectral data is collected at fixed wavelength intervals, with the wavelength range set to a specific interval, recording the spectral intensity at different wavelengths. Pressure fluctuation data is recorded at fixed collection intervals in a specific pressure unit, capturing instantaneous pressure fluctuations. Continuous temperature data is collected in real time in degrees Celsius, recording the temperature at various points inside and outside the cavity and in the pipeline. Instantaneous fluid flow velocity data records the instantaneous flow velocity of the fluid in the pipeline in a specific velocity unit. The collection process is continuous for a fixed duration to ensure that the raw test data comprehensively reflects the equipment's operating status, providing a foundation for subsequent data processing and model building.

[0020] Step S12: Analyze the component characteristics of the raw detection data, extract the characteristic peaks of each target component through the spectral feature matching algorithm, and generate time-series data of component concentration by combining the correlation logic between component concentration and spectral intensity. In this embodiment of the invention, component feature analysis is performed on the original detection data to generate time-series component concentration data, supporting the subsequent construction of a correlation model. The spectral data of the system components in the original detection data are preprocessed to remove spectral noise signals and retain valid spectral data. A spectral feature matching algorithm is activated to compare the preprocessed spectral data with the characteristic spectra of known components, extracting the characteristic peaks of each target component. Each characteristic peak corresponds to a specific component's characteristic wavelength, and each target component has a unique characteristic peak. Combining the correlation logic between component concentration and spectral intensity, a concentration-intensity calculation formula is established: Component concentration = Spectral intensity ÷ Characteristic peak coefficient, where the characteristic peak coefficient is a fixed value inherent to each component. For example, if the characteristic peak coefficient of target component A is 5.2, and the spectral intensity is 26, the calculated concentration of component A is 26 ÷ 5.2 = 5, with the unit being a specific concentration unit. The component concentrations calculated at each acquisition time are sequentially organized to generate time-series component concentration data. This time-series data contains the target component concentrations at each time point, clearly presenting the variation of component concentration over time, adapting to the component monitoring needs of various supercritical processes.

[0021] Step S13: Perform time-series correlation processing on pressure fluctuation data, continuous temperature data, and instantaneous fluid velocity data, extract the inter-frame variation and steady-state characteristics of each parameter, and generate a sequence of operating state parameters; In this embodiment of the invention, a sequence of operating state parameters is generated by performing time-series correlation processing on pressure, temperature, and flow velocity data to reflect the operational stability of the equipment. Pressure fluctuation data, continuous temperature data, and instantaneous fluid flow velocity data are extracted from the original detection data and aligned according to the acquisition time sequence to ensure that the three correspond and match at the same time point. Time-series correlation calculations are performed on the pressure fluctuation data to calculate the pressure difference between two adjacent acquisition times, obtaining the pressure inter-frame change, and simultaneously calculating the average pressure over a fixed duration as the pressure steady-state characteristic. The same processing is applied to the continuous temperature data, calculating the temperature difference between adjacent times to obtain the temperature inter-frame change, and calculating the average temperature over a fixed duration as the temperature steady-state characteristic. For the instantaneous fluid flow velocity data, the velocity difference between adjacent times to obtain the velocity inter-frame change, and calculating the average velocity over a fixed duration as the velocity steady-state characteristic. For example, with a pressure acquisition interval of 1 second, the pressure data for 5 consecutive seconds are 10, 10.2, 9.9, 10.1, and 10.3. The calculated inter-frame changes are 0.2, -0.3, 0.2, and 0.2, respectively, and the average pressure over 5 seconds is 10.1, which serves as the steady-state pressure characteristic. The inter-frame changes and steady-state characteristics of each parameter are integrated in chronological order to generate a sequence of operational status parameters, fully presenting the changing characteristics of the equipment's operational parameters.

[0022] Step S14: Integrate the time series data of component concentration with the sequence of operating status parameters to construct a real-time detection sequence and clarify the spatiotemporal correspondence of each detection parameter; In this embodiment of the invention, a real-time detection sequence is constructed by fusing component concentration and time-series parameters, clarifying the spatiotemporal correspondence of the parameters. The component concentration time-series data generated in step S12 and the operating state parameter sequence generated in step S13 are extracted. Using the acquisition time as a benchmark, the component concentration data at the same time point are mapped one-to-one with the inter-frame changes and steady-state characteristics of pressure, temperature, and flow rate, ensuring data synchronization in the time dimension. The fused data is verified, and data with mismatched time points or abnormal values ​​are removed. For example, if component concentration data is missing at a certain moment, all operating state parameters corresponding to that moment are removed, ensuring the consistency of the fused data. The spatiotemporal correspondence of each detection parameter is clarified, i.e., the component concentration and operating state parameters at the same time point and the same detection location are correlated. All correlated data are integrated and arranged in chronological order to construct a real-time detection sequence. The real-time detection sequence includes component concentration, pressure fluctuation characteristics, temperature steady-state characteristics, and flow rate change characteristics at each moment, providing complete real-time data support for subsequent application of correlation models and adaptive control, adapting to the detection needs of various supercritical processes.

[0023] Step S15: Based on the operational requirements of different processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing, collect the standard values ​​of component concentration, standard range of pressure, standard range of temperature, and standard parameters of flow rate under the stable operating conditions of each process, and construct a component-state correlation model for the extraction process by combining the component-state coupling law. In this embodiment of the invention, standard parameters are collected and a component-state correlation model for the extraction process is constructed by combining the requirements of various supercritical processes. For the operational requirements of different processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing, the stable operation mode of the corresponding process is initiated. After the process operation status stabilizes, standard parameters for each process are collected. The collected standard parameters include standard values ​​for component concentration, standard pressure range, standard temperature range, and standard flow rate. Specifically, for the supercritical extraction process, the standard value for component concentration is set to a specific value, the standard pressure range to a specific range, the standard temperature range to a specific degree Celsius range, and the standard flow rate to a specific velocity value. For the supercritical drying process, the standard value for component concentration is another specific value, and the standard pressure, temperature, and flow rate parameters are adjusted accordingly to suit the drying process. The supercritical anhydrous dyeing process is similarly configured with corresponding standard parameters collected. Based on the collected standard parameters, the correlation between component concentration and pressure, temperature, and flow rate is analyzed, i.e., the component-state coupling law. A coupling calculation formula is established, for example, component concentration = (pressure × temperature) ÷ (flow rate × coupling coefficient). The coupling coefficient is set to a fixed value according to different processes. Based on this formula, a component-state correlation model of the extraction process is constructed. The model can predict the component concentration according to the operating parameters, or adjust the operating parameters according to the component concentration.

[0024] Step S16: Based on the component-state correlation model of the extraction process, classify and encode the standard parameters of each process to generate a benchmark control parameter library containing process type, component threshold, operating parameter baseline value and control response threshold.

[0025] In this embodiment of the invention, a baseline control parameter library is generated by classifying and encoding the standard parameters of each process, providing a baseline basis for adaptive control. Based on the component-state correlation model of the extraction process constructed in step S15, the standard parameters of the three processes—supercritical extraction, supercritical drying, and supercritical anhydrous dyeing—are classified and encoded respectively. The encoding rules are set with unique codes according to the process type. For example, the supercritical extraction process is coded as 01, supercritical drying as 02, and supercritical anhydrous dyeing as 03. The standard parameters of each process are further refined and encoded. The component threshold codes include upper and lower limits. For example, the component concentration threshold codes for the supercritical extraction process are 01001 (upper limit) and 01002 (lower limit). The operating parameter baseline values ​​are encoded to correspond to the standard values ​​of pressure, temperature, and flow rate. For example, the pressure baseline value is coded as 01003, and the temperature baseline value is coded as 01004. The control response threshold codes correspond to the trigger thresholds when the parameters deviate from the baseline values. For example, when the pressure deviates from the baseline value by 0.5, the trigger response is coded as 01005. All codes and corresponding parameters are integrated to generate a baseline control parameter library. The library contains process type, component threshold, operating parameter baseline value and control response threshold. Each code corresponds to a unique parameter, ensuring that the baseline parameters of the corresponding process can be quickly called during subsequent adaptive control, so as to achieve precise control of various supercritical technical equipment.

[0026] Furthermore, as an embodiment of the present invention, such as Figure 3 As shown, Figure 1 A detailed flowchart of step S2 is shown below. In this embodiment, step S2 includes the following steps: Step S21: Retrieve the reference parameters corresponding to the current supercritical process from the reference control parameter library, including the reference time series of component concentration, the reference fluctuation range of pressure, the reference time series of temperature, and the reference range of flow rate, and generate a reference parameter sequence; In this embodiment of the invention, a benchmark parameter sequence is generated by obtaining the benchmark parameters corresponding to the current process from the benchmark control parameter library, providing a comparison benchmark for subsequent deviation calculation. First, the type of supercritical process currently in operation is identified, specifying whether it is supercritical extraction, supercritical drying, or supercritical anhydrous dyeing. For example, the current supercritical extraction process corresponds to code 01. Based on the process type and corresponding code, matching benchmark parameters are retrieved from the benchmark control parameter library. The retrieval process is precise according to the code correspondence, ensuring that the parameters are fully compatible with the current process. The retrieved benchmark parameters include the component concentration benchmark time series, pressure benchmark fluctuation range, temperature benchmark time series, and flow rate benchmark range. The component concentration benchmark time series is set at fixed time intervals, for example, one benchmark value every 1 second, covering the entire process; the pressure benchmark fluctuation range is set to 8-10 MPa; the temperature benchmark time series is set according to the process stage, with the benchmark temperature for the extraction stage being 40℃ and for the separation stage being 30℃; the flow rate benchmark range is 0.8-1.2 m / s. All retrieved benchmark parameters are organized in timestamp order to clarify the benchmark values ​​and ranges corresponding to each time node, and a benchmark parameter sequence is generated. The sequence structure is consistent with the real-time detection sequence mentioned above to ensure that subsequent spatiotemporal alignment and deviation calculation can be carried out smoothly, and to meet the benchmark comparison requirements of various supercritical technical equipment.

[0027] Step S22: Spatiotemporally align the real-time detection sequence with the reference parameter sequence, calculate the difference between each detection parameter and the corresponding reference parameter according to the timestamp correspondence, and generate parameter deviation time series data; In this embodiment of the invention, parameter deviation time-series data is generated by completing the spatiotemporal alignment and parameter deviation calculation of the real-time detection sequence and the reference parameter sequence. The previously generated real-time detection sequence and reference parameter sequence are extracted, and spatiotemporal alignment is performed using timestamps as the core. Parameters at the same timestamp and the same detection point in the two sequences are matched one-to-one to ensure complete spatiotemporal matching. Data with non-overlapping timestamps are removed to avoid misalignment in deviation calculation. Based on the timestamp correspondence, the difference between each detection parameter and its corresponding reference parameter is calculated one by one. The calculation logic is the detection parameter value minus the reference parameter value (for range-type parameters, the difference between the detection value and the midpoint of the range is taken). For example, at a certain timestamp, the real-time component concentration is 6 g / L, corresponding to a baseline concentration of 5 g / L, resulting in a calculated component concentration deviation of 6 - 5 = 1 g / L; the real-time pressure is 9.5 MPa, with the midpoint of the pressure baseline fluctuation range being 9 MPa, resulting in a calculated pressure deviation of 9.5 - 9 = 0.5 MPa; the real-time temperature is 41℃, corresponding to a baseline temperature of 40℃, resulting in a deviation of 41 - 40 = 1℃; the real-time flow velocity is 1.1 m / s, with the midpoint of the flow velocity baseline range being 1.0 m / s, resulting in a deviation of 1.1 - 1.0 = 0.1 m / s. By sequentially organizing the deviation values ​​of all parameters according to their timestamps, time-series data of parameter deviations is generated. This data includes the deviation values ​​of component concentration, pressure, temperature, and flow velocity at each time point, clearly presenting the degree and time-series changes of parameter deviations from the baseline, providing data support for subsequent deviation fluctuation analysis.

[0028] Step S23: Perform time series fluctuation analysis on the parameter deviation time series data to extract the rate of change of deviation, fluctuation period and peak characteristics, and generate deviation fluctuation parameters; In this embodiment of the invention, by analyzing the time-series data of parameter deviations, fluctuation characteristics are extracted and deviation fluctuation parameters are generated to reflect the variation law of deviations. The time-series data of parameter deviations are extracted and classified according to parameter type, and time-series fluctuation analysis is performed on component concentration deviation, pressure deviation, temperature deviation, and flow rate deviation respectively. The rate of change of each parameter deviation is calculated by dividing the difference in deviation between two adjacent time stamps by the time interval. For example, if the component concentration deviation is 1 g / L, 1.2 g / L, and 1.5 g / L for three consecutive time stamps, with a time interval of 1 second, the rate of change for the first two time stamps is calculated to be (1.2-1)÷1=0.2 g / (L·s), and the rate of change for the last two time stamps is (1.5-1.2)÷1=0.3 g / (L·s). The time interval between repeated deviation values ​​is statistically analyzed to determine the fluctuation period. For example, if the pressure deviation peaks every 5 seconds, the fluctuation period is 5 seconds. The maximum deviation value in the time-series data of each parameter deviation is extracted as the peak feature. For example, the maximum value of the temperature deviation is 2℃, and the maximum value of the flow rate deviation is 0.3 m / s. The rate of change, fluctuation period, and peak characteristics of each parameter are organized according to parameter type, the fluctuation details of each parameter are clarified, deviation fluctuation parameters are generated, and the dynamic change law of deviation is fully presented. This provides core feature data for the subsequent construction of dynamic threshold calibration model and adapts to the deviation analysis needs of various supercritical processes.

[0029] Step S24: Construct a dynamic threshold calibration model based on deviation fluctuation parameters, and calculate the dynamic threshold for deviation judgment by combining the composition change characteristics and operating stability requirements of the supercritical process, and optimize the deviation judgment criteria. In this embodiment of the invention, a dynamic threshold calibration model is constructed to calculate the dynamic threshold and optimize the deviation judgment criteria, thereby improving the adaptability of deviation judgment. Based on the deviation fluctuation parameters and combined with the component change characteristics and operational stability requirements of the supercritical process, a dynamic threshold calibration model is constructed. The model takes the deviation fluctuation parameters as input and the dynamic threshold as output, integrating the correlation between the component concentration change rate, the fluctuation period of the operating parameters, and the stability requirements. For example, in the supercritical extraction process, the greater the component concentration change rate, the more relaxed the dynamic threshold needs to be; the more stable the operational state, the tighter the dynamic threshold needs to be. The model calculation formula is set as: Dynamic Threshold = Peak Feature × (1 + Change Rate ÷ Stability Coefficient), where the stability coefficient is set according to the process type: 0.5 for supercritical extraction, 0.4 for supercritical drying, and 0.6 for supercritical anhydrous dyeing. Taking component concentration deviation as an example, the peak value is 1.5 g / L, and the rate of change is 0.3 g / (L·s). The calculated dynamic threshold is 1.5 × (1 + 0.3 ÷ 0.5) = 1.5 × 1.6 = 2.4 g / L. The pressure deviation peak value is 0.8 MPa, and the rate of change is 0.1 MPa / s. The calculated dynamic threshold is 0.8 × (1 + 0.1 ÷ 0.5) = 0.8 × 1.2 = 0.96 MPa. The dynamic thresholds for temperature and flow rate deviations are calculated using the same logic. All dynamic thresholds are integrated, the original deviation judgment criteria are optimized, and fixed thresholds are replaced with dynamic thresholds. This adapts the deviation judgment criteria to the real-time operating status of the process, improving the rationality of deviation judgment.

[0030] Step S25: Based on the optimized deviation judgment criteria, filter and calibrate the parameter deviation time series data, remove abnormal deviation points, and generate calibrated deviation parameters.

[0031] In this embodiment of the invention, the parameter deviation time series data is filtered and calibrated according to the optimized deviation judgment criteria to generate calibrated deviation parameters, ensuring the reliability of the deviation data. The parameter deviation time series data and the optimized deviation judgment criteria (dynamic threshold) are extracted, and the deviation values ​​of each parameter at each time point are compared with the corresponding dynamic threshold. Abnormal deviation points whose deviation values ​​exceed the dynamic threshold are screened out, and these abnormal deviation points and all parameter deviation data at their corresponding time points are removed to prevent abnormal data from affecting subsequent control decisions. For example, if the component concentration dynamic threshold is 2.4 g / L, and the component concentration deviation at a certain time point is 2.6 g / L, exceeding the dynamic threshold, it is determined to be an abnormal deviation point, and the component concentration, pressure, temperature, and flow rate deviation data at that time point are removed; if the pressure dynamic threshold is 0.96 MPa, and the pressure deviation at a certain time point is 0.9 MPa, not exceeding the threshold, the deviation data is retained. The deviation data after outlier removal is calibrated by calculating the average of adjacent valid deviation values ​​to fill the data gaps left by outlier removal. For example, if deviation data for a certain timestamp is removed, the average deviation value of the two timestamps before and after it is taken as the calibration deviation value for that timestamp, ensuring the continuity of the deviation data. After calibration, all valid deviation data are organized in chronological order to generate calibrated deviation parameters. The data is complete and without anomalies, providing accurate deviation data for the subsequent adaptive electrical control of various supercritical technical equipment.

[0032] Furthermore, as an embodiment of the present invention, such as Figure 4 As shown, Figure 3 A detailed flowchart of step S24 is shown. In this embodiment, step S24 includes the following steps: S241: Extract the deviation change rate, fluctuation period and peak characteristics from the deviation fluctuation parameters, and combine them with the component mass transfer law and fluid flow characteristics in the supercritical extraction process to obtain the correlation coefficient between deviation fluctuation and process stability; construct a deviation impact assessment model based on the correlation coefficient, analyze the degree of influence of different deviation fluctuation parameters on supercritical extraction efficiency and component separation effect, and generate deviation impact weight parameters. In this embodiment of the invention, an evaluation model is constructed and deviation influence weight parameters are generated by extracting deviation fluctuation characteristics and calculating correlation coefficients. From the deviation fluctuation parameters, the deviation change rate, fluctuation period, and peak value characteristics are extracted one by one. Combined with the component mass transfer laws and fluid flow characteristics of each process—supercritical extraction, supercritical drying, and supercritical anhydrous dyeing—the correlation between deviation fluctuation and process stability is analyzed, and the correlation coefficient is calculated. The correlation coefficient is calculated by dividing the covariance of the deviation fluctuation parameter and the process stability parameter by the product of their standard deviations. For example, if the deviation change rate is 0.2 g / (L·s), the process stability parameter is 0.8, the covariance is 0.128, and the standard deviations are 0.1 and 0.2 respectively, the correlation coefficient is calculated as 0.128 ÷ (0.1 × 0.2) = 6.4. A deviation impact assessment model is constructed based on correlation coefficients. The model integrates the correlation logic between correlation coefficients and process efficiency and separation effect, analyzes the influence degree of different deviation fluctuation parameters, and calculates the deviation influence weight parameters. The total weight parameters are 1. For example, the deviation change rate weight is 0.4, the fluctuation period weight is 0.3, and the peak feature weight is 0.3. In the supercritical drying process, the weights are adjusted to change rate 0.5, fluctuation period 0.2, and peak feature 0.3. The model clarifies the proportion of influence of each deviation fluctuation parameter on the extraction efficiency and component separation effect of various supercritical processes, providing a basis for subsequent initial threshold calculation.

[0033] S242: Based on the deviation influence weight parameter and the target extraction requirements of the current process, calculate the initial threshold for deviation judgment; collect the component concentration change rate in the supercritical extraction equipment in real time, calculate the instantaneous concentration change rate through the first derivative of the component concentration time series data, and integrate the instantaneous concentration change rates of each time stamp to generate a concentration change rate sequence. In this embodiment of the invention, by setting the target extraction efficiency and component separation standards according to the deviation influence weight parameter and the target extraction requirements of the current process, the initial threshold for deviation judgment is calculated. The initial threshold is calculated by multiplying the peak characteristics of each deviation fluctuation parameter by the corresponding weight parameter and then summing them. For example, in the current supercritical extraction process, the peak deviation change rate is 0.3 g / (L·s) with a weight of 0.4, the peak fluctuation period is 5s with a weight of 0.3, and the peak characteristic is 1.5 g / L with a weight of 0.3. The initial threshold is calculated as (0.3×0.4) + (5×0.3) + (1.5×0.3) = 0.12 + 1.5 + 0.45 = 2.07 g / L. Real-time component concentration data within the supercritical fluid extraction equipment is collected and combined with time-series component concentration data. The instantaneous rate of change of concentration is calculated using the first derivative. The first derivative is calculated by dividing the difference in component concentration between two adjacent time stamps by the time interval. For example, if the concentrations at adjacent time stamps are 5 g / L and 5.3 g / L, with a time interval of 1 second, the instantaneous rate of change of concentration is calculated as (5.3-5) ÷ 1 = 0.3 g / (L·s). All calculated instantaneous rates of change of concentration are then organized in time stamp order to generate a concentration rate of change sequence, clearly presenting the instantaneous state of component concentration changes and providing data support for subsequent threshold adjustments.

[0034] S243: Obtain the steady-state values ​​of operating parameters, detect the steady-state values ​​of operating parameters, extract the steady-state ranges of pressure, temperature, and flow rate parameters, calculate the fluctuation amplitude of each parameter within the steady-state range, and generate steady-state fluctuation values ​​of operating parameters; based on the concentration change rate sequence and the steady-state fluctuation values ​​of operating parameters, construct a threshold adjustment model, and analyze the influence of concentration change rate and steady-state fluctuation values ​​on the deviation judgment threshold. In this embodiment of the invention, pressure, temperature, and flow rate parameters during the operation of supercritical equipment are extracted, and the average value of each parameter over a fixed period of time is calculated as the steady-state value of the operating parameters, for example, pressure 10 MPa, temperature 40℃, and flow rate 1.0 m / s. The steady-state values ​​of the operating parameters are detected, and the steady-state ranges of each parameter are defined. The steady-state range for pressure is 9.8-10.2 MPa, for temperature it is 39-41℃, and for flow rate it is 0.9-1.1 m / s. The fluctuation amplitude of each parameter within the steady-state range is calculated by subtracting the minimum value from the maximum value within the steady-state range, resulting in a pressure fluctuation amplitude of 0.4 MPa, a temperature fluctuation amplitude of 2℃, and a flow rate fluctuation amplitude of 0.2 m / s. These are then integrated to generate the steady-state fluctuation values ​​of the operating parameters. Based on the concentration change rate sequence and the currently calculated steady-state fluctuation value of the operating parameters, a threshold adjustment model is constructed. The model takes the concentration change rate and steady-state fluctuation value as inputs and the threshold adjustment amount as outputs. The influence of the two on the deviation judgment threshold is analyzed. That is, the larger the concentration change rate and the larger the steady-state fluctuation value, the larger the threshold adjustment range, and vice versa. This model is adapted to the operating characteristics of various supercritical processes.

[0035] S244: Based on the influence law, the initial threshold is dynamically corrected. When the concentration change rate increases or the steady-state fluctuation value of the operating parameter exceeds the preset range, the dynamic threshold is reduced to improve the sensitivity of deviation detection. When the concentration change rate decreases or the operating parameter is in a stable state, the dynamic threshold is expanded to avoid misjudgment, thereby outputting the corrected dynamic threshold. The corrected dynamic threshold is integrated into the deviation judgment logic to optimize the deviation judgment standard and clarify the judgment rules corresponding to different degrees of deviation.

[0036] In this embodiment of the invention, the calculated initial threshold is dynamically corrected according to the analyzed influence law, and a correction coefficient is set. The correction coefficient = (concentration change rate + steady-state fluctuation value) ÷ baseline correction value. The baseline correction value is set according to the process type: 0.5 for supercritical extraction, 0.4 for supercritical drying, and 0.6 for supercritical anhydrous dyeing. When the concentration change rate increases to 0.5 g / (L·s) and the steady-state fluctuation value of the operating parameters exceeds the preset range (such as pressure fluctuation amplitude of 0.5 MPa), the correction coefficient = (0.5 + 0.5) ÷ 0.5 = 2. The initial threshold of 2.07 g / L is corrected to 2.07 ÷ 2 = 1.035 g / L, thus reducing the dynamic threshold to improve the sensitivity of deviation detection. When the concentration change rate decreases to 0.1 g / (L·s) and the operating parameters are in a stable state (fluctuation amplitude within the preset range), the correction coefficient = (0.1 + 0.2) ÷ 0.5 = 0.6. The initial threshold is corrected to 2.07 ÷ 0.6 = 3.45 g / L, thus expanding the dynamic threshold to avoid misjudgment. The corrected dynamic threshold is then output. The revised dynamic threshold is incorporated into the deviation judgment logic, and the deviation judgment criteria are optimized. It is clearly defined that a deviation value exceeding the dynamic threshold is a serious deviation, reaching 80% of the dynamic threshold is a minor deviation, and being below 80% is normal. The judgment rules corresponding to different degrees of deviation are distinguished to ensure that the deviation judgment is consistent with the real-time operating status of various supercritical processes, and to provide an accurate basis for subsequent deviation calibration.

[0037] Furthermore, as an embodiment of the present invention, such as Figure 5 As shown, Figure 1 A detailed flowchart of step S3 is shown below. In this embodiment, step S3 includes the following steps: Step S31: Classify and analyze the deviation parameters after calibration, classify them according to component concentration deviation, pressure deviation, temperature deviation and flow rate deviation, extract the magnitude, trend and duration of each type of deviation, and generate deviation classification feature data; In this embodiment of the invention, the calibrated deviation parameters are classified, features are extracted, and deviation classification feature data is generated to support deviation level assessment. The calibrated deviation parameters are extracted and classified according to parameter type, clearly divided into four categories: component concentration deviation, pressure deviation, temperature deviation, and flow rate deviation. The classification process strictly adheres to parameter attributes to ensure no confusion or omission, adapting to the parameter monitoring needs of various processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing. The core characteristics of each type of deviation are extracted, including numerical magnitude, trend, and duration. The numerical magnitude is directly obtained from the specific values ​​of the calibrated deviation parameters, such as a component concentration deviation of 1.2 g / L, a pressure deviation of 0.3 MPa, a temperature deviation of 1.5℃, and a flow rate deviation of 0.15 m / s. The trend is determined by calculating the difference in deviation between three adjacent time stamps. For example, if the component concentration deviations are consecutively 1.0 g / L, 1.2 g / L, and 1.4 g / L, with a difference of 0.2 g / L for each, the trend is determined to be continuously increasing. The duration is calculated by multiplying the number of consecutive time stamps where the deviation value exceeds the normal range by the acquisition interval. For example, if the acquisition interval is 1 second, and the deviation exceeds the normal range for 6 consecutive time stamps, the duration is 6 seconds. These three characteristics of each type of deviation are then categorized to generate deviation classification feature data, clearly presenting the specific status of each type of deviation and laying the foundation for subsequent deviation level assessment.

[0038] Step S32: Based on the deviation classification feature data and combined with the control requirements of the supercritical process, construct a deviation level evaluation system, classify various deviations into different levels, and generate deviation level parameters; In this embodiment of the invention, a deviation level assessment system is constructed based on deviation classification feature data and the control requirements of various supercritical processes. The assessment system uses the magnitude, trend, and duration of deviation as the core assessment indicators, and clarifies the classification standards for various deviation levels. For example, in supercritical extraction processes, a component concentration deviation ≤ 0.5 g / L, duration ≤ 2 seconds, and a stable trend are classified as Level 1 deviation; 0.5 g / L < deviation ≤ 1.5 g / L, 2 seconds < duration ≤ 5 seconds, and a slowly increasing trend are classified as Level 2 deviation; deviation > 1.5 g / L, duration > 5 seconds, and a rapidly increasing trend are classified as Level 3 deviation; pressure deviation ≤ 0.2 MPa is Level 1, 0.2 MPa < deviation ≤ 0.8 MPa is Level 2, and deviation > 0.8 MPa is Level 3; temperature and flow rate deviations are classified in the same way. By comparing with the classification criteria, the level of each type of deviation is determined one by one. For example, a component concentration deviation of 1.2 g / L, a duration of 4 seconds, and a slow increasing trend are judged as a level 2 deviation; a pressure deviation of 0.3 MPa is also judged as a level 2 deviation. All deviation levels are integrated to generate deviation level parameters, clarifying the level classification of each type of deviation, and providing a basis for subsequent control strategy retrieval.

[0039] Step S33: Based on the deviation level parameter, retrieve the corresponding control strategy template from the benchmark control parameter library, combine it with the component-state correlation model of the extraction process, optimize the adjustment parameters of the control strategy, and generate an adaptive control strategy. In this embodiment of the invention, by defining the level and impact of various deviations based on deviation level parameters, and considering the current supercritical process type, a corresponding control strategy template is retrieved from the benchmark control parameter library. For example, in the current supercritical drying process, template 2 is retrieved for a second-level deviation. This template includes the basic adjustment direction and sequence of pressure, temperature, and flow rate. Combining the constructed component-state correlation model of the extraction process, the relationship between deviation levels and process parameters is analyzed, and the adjustment parameters of the control strategy are optimized. The calculation of adjustment parameters considers the magnitude and trend of deviation values. For example, for a second-level deviation of 1.2 g / L in component concentration, the model calculates that the temperature needs to be increased by 2°C and the flow rate reduced by 0.1 m / s to bring the deviation back to normal; for a second-level deviation of 0.3 MPa in pressure, the calculation shows that the pressure regulating valve opening needs to be reduced by 10%, with the adjustment range proportional to the deviation value. The optimized adjustment parameters are integrated into the control strategy template, defining the adjustment sequence, adjustment range, and duration, generating an adaptive control strategy. This strategy adapts to the deviation state of the current process, specifically addressing various deviation problems and meeting the control requirements of various supercritical technical equipment.

[0040] Step S34: Convert the adaptive control strategy into a control signal that the electrical control module can recognize, determine the output frequency, amplitude and duration of the control signal, and generate the corresponding output signal of the electrical control module; In this embodiment of the invention, an adaptive control strategy is extracted, the specific requirements of each adjustment action are clarified, and these are converted into control signals recognizable by the electrical control module. The conversion process follows the signal reception standard of the electrical control module, specifying the output frequency, amplitude, and duration of the control signals. For example, to control the temperature controller to rise by 2°C, the corresponding control signal output frequency is set to 50Hz, the amplitude to 5V, and the duration to 10 seconds, ensuring that the temperature controller can rise steadily. To control the flow rate regulating pump to reduce the flow rate by 0.1m / s, the control signal output frequency is 60Hz, the amplitude to 4V, and the duration to 8 seconds. To control the pressure regulating valve to reduce the opening by 10%, the output frequency is 50Hz, the amplitude to 6V, and the duration to 5 seconds. All control signals are integrated, sorted according to the adjustment order, and the actuator corresponding to each control signal is identified. Output signals corresponding to the electrical control module are generated, with a unified signal format and clearly defined parameters, ensuring that the electrical control module can accurately receive and execute them, providing instruction support for subsequent actuator actions.

[0041] Step S35: Transmit the output signal corresponding to the electrical control module to the actuator of the supercritical extraction equipment, drive the actuator to perform corresponding actions, including starting and stopping the pressure regulating valve, temperature controller, and flow rate regulating pump, and adjusting parameters to achieve real-time control of the extraction process.

[0042] In this embodiment of the invention, the output signal corresponding to the electrical control module is transmitted to various actuators of the supercritical extraction equipment via electrical transmission lines. During transmission, signal strength is monitored to ensure no signal attenuation or interference, adapting to the control requirements of actuators in various supercritical extraction, supercritical drying, and supercritical anhydrous dyeing equipment. After the signal is transmitted to the pressure regulating valve, it drives the valve to operate, reducing the opening by 10% as required by the control signal, adjusting the pressure inside the equipment, and gradually returning the pressure deviation to normal. After being transmitted to the temperature controller, it drives the temperature controller to start the heating program, increasing the temperature by 2°C according to the set amplitude and duration, stabilizing the temperature inside the equipment. After being transmitted to the flow rate regulating pump, it drives the pump body to adjust its speed, reducing the flow rate by 0.1 m / s, controlling the fluid transmission speed. All actuators operate synchronously, and equipment operating parameters are collected in real time during the operation, providing feedback on the adjustment effect. If the deviation does not return to normal, the control signal continues to be transmitted until the deviation reaches the target, achieving real-time control of the extraction process, ensuring stable operation of various supercritical processes, and improving extraction efficiency and component separation effect.

[0043] Furthermore, as an embodiment of the present invention, such as Figure 6 As shown, Figure 1 A detailed flowchart of step S4 is shown below. In this embodiment, step S4 includes the following steps: Step S41: After the actuator moves, control the multi-dimensional sensing components to collect the operating parameters and component change data of the supercritical extraction equipment in real time, including the adjusted pressure, temperature, flow rate data and component concentration time series data, and generate a feedback data sequence. In this embodiment of the invention, after the actuator completes the first control action, it immediately controls the multi-dimensional sensing components to start real-time data acquisition. The acquisition range covers the extraction chamber, separation chamber, and fluid transmission pipeline of the supercritical extraction equipment, adapting to the feedback monitoring needs of various processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing. The acquired operating parameters and component change data include adjusted pressure, temperature, flow rate data, and component concentration time-series data. Pressure data is acquired at 1-second intervals in MPa, temperature data is acquired in real time in °C, flow rate data is acquired synchronously in m / s, and component concentration time-series data is acquired at fixed wavelength intervals and the concentration value is calculated. For example, after the actuator is adjusted, the collected pressure data is 9.7 MPa, temperature is 42℃, flow rate is 0.9 m / s, and component concentrations are 5.8 g / L, 5.9 g / L, and 6.0 g / L respectively. All data are organized according to the collection timestamp, and data with signal interruption during the collection process are removed to generate a feedback data sequence. The sequence fully presents the operating status and component changes of the equipment after adjustment, providing a basis for subsequent effectiveness verification and deviation calculation.

[0044] Step S42: Verify the validity of the feedback data sequence, remove abnormal feedback data through a data consistency detection algorithm to ensure the authenticity and reliability of the feedback data, and generate verified feedback data; In this embodiment of the invention, by extracting the feedback data sequence, a data consistency detection algorithm is activated to verify the data one by one. The verification logic involves comparing the same parameters at different collection points under the same time stamp, calculating the parameter difference, and if the difference exceeds a set range, it is determined to be abnormal feedback data and discarded. For example, at the same time stamp, the pressure collection value of the extraction chamber is 9.7 MPa, and the pressure collection value of the separation chamber is 10.5 MPa. The difference between the two is 0.8 MPa, which exceeds the set difference range of 0.5 MPa. Therefore, the pressure data at this time stamp is determined to be abnormal data and discarded. In the component concentration data, the collection value at a certain time stamp is 8.0 g / L, which has a large difference from the adjacent time stamps of 5.9 g / L and 6.0 g / L. Therefore, it is determined to be abnormal data and discarded. After verification, the remaining valid data is organized and the missing timestamps after removing abnormal data are filled in. The filling method is to take the average of adjacent valid data. For example, if a certain timestamp data is removed, the average of the pressure of the two timestamps before and after it is taken as the filling. The post-verification feedback data is generated to ensure that the data is continuous, true and reliable, and to provide accurate data for subsequent comparison and analysis.

[0045] Step S43: Compare the verified feedback data with the corresponding benchmark parameters in the benchmark control parameter library, calculate the feedback deviation data, and analyze the control effect of the actuator's actions on the extraction process; In this embodiment of the invention, by extracting the feedback data after verification, the reference parameters corresponding to the current supercritical process are retrieved from the reference control parameter library. For example, the current process is a supercritical anhydrous dyeing process, with a pressure reference range of 9.5-10.5 MPa, a temperature reference range of 40-43℃, a flow rate reference range of 0.8-1.1 m / s, and a component concentration reference value of 6.0 g / L. The feedback data after verification is compared with the corresponding reference parameters one by one, and the feedback deviation data is calculated. The calculation method is to subtract the reference parameter from the verification data (for range-type parameters, the difference between the verification data and the midpoint of the range is taken). For example, if the verification pressure is 9.7 MPa and the midpoint of the reference range is 10.0 MPa, the feedback pressure deviation is 9.7-10.0=-0.3 MPa; if the verification temperature is 42℃ and the midpoint of the reference range is 41.5℃, the feedback temperature deviation is 42-41.5=0.5℃; if the verification component concentration is 6.0 g / L, the feedback concentration deviation is 6.0-6.0=0 g / L. By analyzing the feedback deviation data, the control effect of the actuator's actions is improved. The smaller the absolute value of the deviation, the better the control effect. For example, if the component concentration deviation is 0, it means that the control action has fully achieved the expected result. The pressure and temperature deviations are small, indicating a good control effect, which provides a basis for subsequent control strategy optimization.

[0046] Step S44: Based on the feedback deviation data, combined with the calibrated deviation parameters and deviation level parameters, construct a control parameter optimization model, adjust the parameters of the electrical control output signal, and optimize the adaptive control strategy; In this embodiment of the invention, a control parameter optimization model is constructed based on feedback deviation data, combined with calibrated deviation parameters and deviation level parameters. The model integrates the correlation logic of feedback deviation, calibrated deviation, and deviation level. Its core is to adjust the parameters of the electrical control output signal according to the magnitude and direction of the feedback deviation. For example, if the feedback pressure deviation is -0.3 MPa (below the reference midpoint), the calibrated pressure deviation is 0.3 MPa, and the deviation level is level two, the model calculates that the pressure regulating valve opening needs to be increased by 5%, corresponding to adjusting the amplitude of the electrical control output signal from 6V to 5V and the duration from 5 seconds to 6 seconds. If the feedback temperature deviation is 0.5℃, the calibrated temperature deviation is 1.5℃, and the deviation level is level two, the calculation calculates that the temperature controller's temperature rise needs to be reduced by 0.5℃, the control signal output frequency remains unchanged at 50Hz, and the amplitude is adjusted from 5V to 4.5V. The adjusted control signal parameters are integrated into the original adaptive control strategy, optimizing the adjustment sequence and duration to generate an optimized adaptive control strategy. This ensures that the strategy can specifically correct feedback deviations, improve control accuracy, and adapt to the control requirements of various supercritical processes.

[0047] Step S45: Feed the optimized control strategy and control parameters back to the electrical control module, and iteratively execute the control actions and feedback detection process to achieve adaptive closed-loop control of the supercritical extraction process.

[0048] In this embodiment of the invention, the optimized control strategy and control parameters are fed back to the electrical control module via an electrical transmission line. Upon receiving the feedback, the electrical control module updates the original control logic and generates a new electrical control output signal based on the optimized control parameters. Subsequently, the actuator is driven to act according to the new control signal, for example, increasing the opening of the pressure regulating valve by 5% and decreasing the temperature controller's heating amplitude by 0.5°C. After the action is completed, the multi-dimensional sensing components are controlled to collect feedback data again, repeating steps S41-S44 for data verification, deviation calculation, and strategy optimization. For example, after the first iteration, the feedback pressure deviation is reduced to -0.1MPa and the temperature deviation to 0.2°C. The control parameters are then optimized again, increasing the pressure regulating valve opening by 2% and adjusting the temperature controller amplitude to 4.8V. The control actions and feedback detection process are continuously iterated until the feedback deviation returns to the normal range, achieving adaptive closed-loop control of the supercritical extraction process. This ensures the continuous and stable operation of various processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing, improving process efficiency and product quality.

[0049] Furthermore, as an embodiment of the present invention, such as Figure 7 As shown, Figure 6 A detailed flowchart of step S44 is shown in this embodiment. Step S44 includes the following steps: S441: Extract the feedback deviation value, deviation change trend and steady-state deviation characteristics of each parameter from the feedback deviation data, combine the deviation level of the calibrated deviation parameter, obtain the correlation between the feedback deviation and the initial deviation, and generate the deviation correlation parameter. In this embodiment of the invention, the feedback deviation value, deviation trend, and steady-state deviation characteristics of each parameter are extracted one by one from the feedback deviation data. The feedback deviation value is directly read from the calculation results of the verified feedback data and the benchmark parameters, such as pressure feedback deviation -0.3MPa, temperature feedback deviation 0.5℃, component concentration feedback deviation 0g / L, and flow rate feedback deviation -0.05m / s. The deviation trend is determined by calculating the difference in feedback deviation between three adjacent time stamps. For example, if the pressure feedback deviation is continuously -0.3MPa, -0.25MPa, and -0.2MPa, with a difference of 0.05MPa, the trend is determined to be continuously decreasing. The steady-state deviation characteristic is calculated by the average value of the feedback deviation over a fixed period of time. For example, the average pressure feedback deviation over 5 seconds is -0.25MPa, which is used as the pressure steady-state deviation characteristic. By combining the deviation levels of the calibrated deviation parameters (such as pressure level 2 deviation, temperature level 2 deviation), the difference between the feedback deviation and the initial calibrated deviation is calculated. For example, if the initial calibrated deviation of pressure is 0.3 MPa, the feedback deviation is -0.3 MPa, and the difference is -0.6 MPa; if the initial calibrated deviation of temperature is 1.5℃, the feedback deviation is 0.5℃, and the difference is -1.0℃. These differences and deviation characteristics are integrated to generate deviation correlation parameters, clarifying the degree of correlation and the changing pattern between the feedback deviation and the initial deviation.

[0050] S442: Based on the deviation correlation parameters, analyze the control efficiency of the actuator's actions and calculate the control effect evaluation parameters, which reflect the actuator's ability to correct deviations. In this embodiment of the invention, by determining the difference and duration between the feedback deviation and the initial deviation based on the deviation correlation parameter, the control efficiency of the actuator's action is analyzed. The control efficiency is reflected in the amount of deviation correction per unit time. The parameters for evaluating the control effect are calculated by dividing the absolute value of the difference between the feedback deviation and the initial calibration deviation by the duration of the actuator action. For example, if the initial calibration deviation of pressure is 0.3 MPa, the feedback deviation is -0.3 MPa, the absolute value of the difference is 0.6 MPa, and the actuator action duration is 5 seconds, the calculated pressure control effect evaluation parameter is 0.6 ÷ 5 = 0.12 MPa / s; if the initial calibration deviation of temperature is 1.5℃, the feedback deviation is 0.5℃, the absolute value of the difference is 1.0℃, and the action duration is 10 seconds, the calculated temperature control effect evaluation parameter is 1.0 ÷ 10 = 0.1℃ / s; if the initial calibration deviation of component concentration is 1.2 g / L, the feedback deviation is 0 g / L, the absolute value of the difference is 1.2 g / L, and the action duration is 8 seconds, the calculated concentration control effect evaluation parameter is 1.2 ÷ 8 = 0.15 g / (L·s). The larger the value of the control effect evaluation parameter, the stronger the ability of the actuator to correct the deviation. By integrating the evaluation results of all parameters, a complete control effect evaluation parameter is generated, which provides a basis for subsequent adjustment of control parameters.

[0051] S443: Combining the evaluation parameters of the control effect, retrieve the optimization rules in the control parameter optimization model, and determine the adjustment direction and adjustment range of the control parameters based on the target extraction effect of the supercritical process; In this embodiment of the invention, by combining the evaluation parameters of the control effect, the corresponding optimization rules are retrieved from the control parameter optimization model. The optimization rules are directly related to the evaluation parameters of the control effect; the smaller the evaluation parameter, the larger the adjustment range. The adjustment direction is determined based on the sign of the feedback deviation. The current target extraction effect of the supercritical process is to stabilize the component concentration at the baseline value and maintain the pressure, temperature, and flow rate within the baseline range. By referring to the optimization rules and combining the evaluation parameters of the control effect of each parameter with the direction of the feedback deviation, the adjustment direction and range are determined. For example, if the pressure control effect evaluation parameter is 0.12 MPa / s and the feedback deviation is negative (below the reference midpoint), the adjustment direction is determined to be increasing the pressure. The adjustment range is calculated as (reference midpoint - feedback deviation) × control coefficient. The control coefficient is set according to the process type; for supercritical extraction, it is 0.8. The calculated pressure adjustment range is (10.0 - 9.7) × 0.8 = 0.24 MPa, corresponding to a 5% increase in the opening of the pressure regulating valve. If the temperature control effect evaluation parameter is 0.1℃ / s and the feedback deviation is positive (above the reference midpoint), the adjustment direction is to decrease the temperature rise. The adjustment range is (42 - 41.5) × 0.8 = 0.4℃, corresponding to a 0.5℃ decrease in the temperature rise of the temperature controller. This clarifies the adjustment direction and specific range of all control parameters.

[0052] S444: Based on the adjustment direction and adjustment magnitude, optimize the frequency, amplitude and duration of the electrical control output signal, and at the same time adjust the deviation judgment logic and adjustment priority in the adaptive control strategy, thereby outputting optimized control parameters and control strategy.

[0053] In this embodiment of the invention, the frequency, amplitude, and duration of the electrical control output signal are optimized based on a determined adjustment direction and amplitude. For example, if the pressure regulating valve needs to increase its opening by 5%, the original control signal amplitude is adjusted from 6V to 5V, the duration from 5 seconds to 6 seconds, and the output frequency remains unchanged at 50Hz. If the temperature controller needs to reduce the temperature rise by 0.5℃, the original control signal amplitude is adjusted from 5V to 4.5V, the duration from 10 seconds to 8 seconds, and the output frequency remains unchanged at 50Hz. If the flow rate regulating pump feedback deviation is -0.05m / s, the adjustment direction is to increase the flow rate, the control signal amplitude is adjusted from 4V to 4.2V, the duration from 8 seconds to 7 seconds, and the output frequency is 60Hz. Simultaneously, the deviation judgment logic in the adaptive control strategy is adjusted. Based on the feedback deviation change trend, the dynamic threshold for pressure deviation judgment is fine-tuned by 0.1MPa, and the dynamic threshold for temperature deviation is fine-tuned by 0.2℃. The adjustment priority is adjusted, with the component concentration control priority increased to ensure that the component concentration is stable at the reference value, followed by the adjustment of pressure, temperature, and flow rate. The system integrates and optimizes control signal parameters, deviation judgment logic, and adjustment priorities, and outputs optimized control parameters and adaptive control strategies to ensure that the strategies can specifically correct feedback deviations, improve control accuracy, and adapt to the closed-loop control requirements of various supercritical processes.

[0054] Furthermore, the step of analyzing the control efficiency of the actuator's actions based on deviation correlation parameters and calculating the control effect evaluation parameters includes the following steps: Based on the feedback deviation data and the calibrated deviation parameters, the deviation correction amount for each parameter is calculated, which is the difference between the calibrated deviation parameters and the feedback deviation data, and a deviation correction amount sequence is generated. In this embodiment of the invention, by extracting feedback deviation data and calibrated deviation parameters, two sets of data with the same timestamp and the same parameter are identified, and the deviation correction amount of each parameter is calculated one by one. The calculation logic is the value of the calibrated deviation parameter minus the value of the feedback deviation data, ensuring that the correction amount can accurately reflect the degree of correction of the deviation by the actuator action. For example, in the current supercritical fluid extraction process, at a certain time stamp, the pressure calibration deviation is 0.3 MPa, and the feedback pressure deviation is -0.3 MPa. The calculated pressure deviation correction is 0.3 - (-0.3) = 0.6 MPa. The temperature calibration deviation is 1.5℃, and the feedback temperature deviation is 0.5℃, so the deviation correction is 1.5 - 0.5 = 1.0℃. The component concentration calibration deviation is 1.2 g / L, and the feedback concentration deviation is 0 g / L, so the deviation correction is 1.2 - 0 = 1.2 g / L. The flow rate calibration deviation is 0.15 m / s, and the feedback flow rate deviation is -0.05 m / s, so the deviation correction is 0.15 - (-0.05) = 0.2 m / s. Following the time stamp order, the deviation corrections for each parameter at each time point are sequentially organized, and abnormal differences occurring during the calculation process are eliminated to ensure data continuity. This generates a deviation correction sequence, fully presenting the specific values ​​and time-series changes of each parameter deviation correction, providing a foundation for subsequent time-series analysis.

[0055] Furthermore, time-series analysis is performed on the deviation correction sequence to extract the rate of change, correction period, and steady-state correction value of the deviation correction, generating correction characteristic parameters; combined with the actuator's action parameters, including action amplitude, action frequency, and response time, the actuator's control characteristic parameters are obtained. In this embodiment of the invention, by performing time-series analysis on the deviation correction amount sequence, classifying and processing according to parameter type, the rate of change, correction period, and steady-state correction value of each parameter deviation correction amount are extracted one by one. The rate of change is calculated by dividing the difference in deviation correction amount between two adjacent timestamps by the acquisition interval. For example, if the pressure deviation correction amount is continuously 0.6MPa, 0.55MPa, and 0.5MPa, and the acquisition interval is 1 second, the rate of change = (0.55-0.6)÷1 = -0.05MPa / s. The correction period is the time interval at which the deviation correction amount tends to stabilize. For example, if the component concentration deviation correction amount tends to stabilize every 4 seconds, the correction period is 4 seconds. The steady-state correction value is calculated as the average value of the deviation correction amount within a fixed time period. For example, if the average value of the temperature deviation correction amount within 5 seconds is 0.9℃, it is used as the steady-state temperature correction value, and these are integrated to generate correction characteristic parameters. Simultaneously, the actuator's motion parameters are extracted, including motion amplitude, motion frequency, and response time. Motion amplitude is the actual adjustment amount of the actuator, such as adjusting the opening of the pressure regulating valve by 10% or raising the temperature of the temperature controller by 2°C. Motion frequency is the number of times the actuator moves per second, set to 2 times / second. Response time is the time from when the actuator receives the signal to when it starts moving, set to 0.3 seconds. These parameters are integrated to generate the actuator's control characteristic parameters, clarifying the actuator's motion capability.

[0056] Furthermore, based on the modified characteristic parameters and the control characteristic parameters, a control effect evaluation model is constructed to analyze the correlation between the deviation correction amount and the actuator action parameters, and to quantify the control effect. Based on the quantification results, control effect evaluation parameters are generated, which cover three dimensions: deviation correction efficiency, steady-state control accuracy, and response speed.

[0057] In this embodiment of the invention, a control effect evaluation model is constructed based on the correction characteristic parameters and the control characteristic parameters. The model integrates the correlation between the two and quantifies the control effect by analyzing the matching degree between the deviation correction amount and the actuator action parameters. The quantification process is carried out in three dimensions: deviation correction efficiency is calculated as the steady-state correction value divided by the actuator response time. For example, if the steady-state pressure correction value is 0.5 MPa and the response time is 0.3 seconds, the correction efficiency = 0.5 ÷ 0.3 ≈ 1.67 MPa / s; steady-state control accuracy is calculated as the ratio of the steady-state correction value to the calibrated deviation parameter. For example, if the steady-state temperature correction value is 0.9℃ and the calibrated deviation is 1.5℃, the steady-state control accuracy = 0.9 ÷ 1.5 = 0.6; the response speed is directly taken as the actuator response time, and the rate of change in the correction characteristic parameters is combined to supplement the response speed evaluation. For example, the larger the absolute value of the rate of change, the faster the response speed. The response speed evaluation value is calculated as 1 ÷ response time, i.e., 1 ÷ 0.3 ≈ 3.33. By integrating the quantitative results from three dimensions, parameters for evaluating the control effect are generated, including deviation correction efficiency of 1.67 MPa / s, steady-state control accuracy of 0.6, and response speed evaluation value of 3.33. These three dimensions comprehensively reflect the actuator's ability to correct deviations, providing accurate quantitative basis for subsequent control parameter adjustments and strategy optimization, and adapting to the control needs of various supercritical processes.

[0058] Furthermore, obtaining the control characteristic parameters of the actuator by combining the actuator's action parameters, including action amplitude, action frequency, and response time, includes the following steps: Collect the motion parameters of the actuator, including motion amplitude, motion frequency and response time, and combine them with the deviation correction trend in the feedback deviation data to obtain the preliminary correlation parameters between the motion parameters and the deviation correction. In this embodiment of the invention, the action parameters of the supercritical extraction equipment actuator are collected in real time, including the action amplitude, action frequency, and response time. The action amplitude is determined according to the actual adjustment amount of the actuator, such as adjusting the pressure regulating valve opening by 10%, raising the temperature controller temperature by 2°C, and adjusting the flow rate regulating pump speed by 100 r / min. The action frequency is calculated by counting the number of actions of the actuator per unit time, set to 2 times / second, counted at 1-second intervals, with the number of actions in 5 consecutive seconds being 2, 2, 2, 2, 2, to ensure stable action frequency. The response time is the time from when the actuator receives the electrical control signal to when it starts to act, obtained by timing. For example, if the signal is emitted at 0 seconds and the actuator starts to act at 0.3 seconds, the response time is 0.3 seconds. The feedback deviation data generated in step S43 is extracted, and the deviation correction trend of each parameter is analyzed. For example, the pressure feedback deviation gradually changes from -0.3MPa to -0.25MPa and -0.2MPa, with a correction trend of continuous decrease; the temperature feedback deviation changes from 0.5°C to 0.4°C and 0.3°C, with a correction trend of continuous decrease. Calculate the correlation value between the motion parameters and the deviation correction trend. Correlation value = motion amplitude × deviation correction difference ÷ response time. For example, if the pressure motion amplitude is 10%, the deviation correction difference is 0.05 MPa, and the response time is 0.3 seconds, the correlation value = 10% × 0.05 ÷ 0.3 ≈ 0.017. Integrate the correlation values ​​of all parameters to generate preliminary correlation parameters and clarify the basic correlation between motion parameters and deviation correction.

[0059] Furthermore, based on the preliminary correlation parameters, the correspondence between the motion amplitude and the deviation correction amplitude is analyzed to generate the amplitude adaptation factor; combined with the amplitude adaptation factor, motion frequency, and response time, a motion timing response model is constructed, the synchronicity characteristics of the motion response are extracted, and timing response parameters are generated. In this embodiment of the invention, based on preliminary correlation parameters, the correspondence between the action amplitude and the deviation correction amplitude of each actuator is analyzed one by one. The deviation correction amplitude is the difference in deviation correction between two adjacent timestamps. For example, a pressure action amplitude of 10% corresponds to a deviation correction amplitude of 0.05 MPa, and a temperature action amplitude of 2℃ corresponds to a deviation correction amplitude of 0.1℃. The amplitude adaptation factor is calculated by dividing the deviation correction amplitude by the action amplitude. For example, the pressure amplitude adaptation factor = 0.05 ÷ 10% = 0.5 MPa / %; the temperature amplitude adaptation factor = 0.1 ÷ 2 = 0.05℃ / ℃. Combining the amplitude adaptation factor with the action frequency and response time of the actuator, an action timing response model is constructed. The model takes action parameters as input and deviation correction timing changes as output, analyzing the time synchronization between action issuance and deviation correction. Extract the synchronicity features of the action response and calculate the synchronicity difference. The synchronicity difference is the difference between the action issuance time and the deviation correction start time. For example, if the action issuance time is 0.3 seconds and the deviation correction start time is 0.4 seconds, the synchronicity difference is 0.1 seconds. Statistically calculate the synchronicity difference of multiple timestamps and use the average value as a timing response parameter. For example, if the synchronicity differences of 5 timestamps are 0.1, 0.12, 0.08, 0.11, and 0.09, the average value is (0.1 + 0.12 + 0.08 + 0.11 + 0.09) ÷ 5 = 0.1 seconds. Generate timing response parameters to clarify the synchronicity level of the action response.

[0060] Furthermore, based on the timing response parameters and deviation correlation parameters, the control lag characteristics and correction accuracy of the actuator's actions are analyzed to generate control response characteristics; by integrating the control response characteristics and amplitude adaptation factors, the dynamic control capability of the actuator is extracted to generate the control characteristic parameters of the actuator.

[0061] In this embodiment of the invention, the control lag characteristics and correction accuracy of the actuator action are analyzed based on the timing response parameters and deviation correlation parameters. The control lag characteristics are reflected by the synchronization difference in the timing response parameters. The larger the synchronization difference, the more obvious the lag. The lag coefficient is calculated as: lag coefficient = synchronization difference ÷ response time. For example, if the synchronization difference is 0.1 seconds and the response time is 0.3 seconds, the lag coefficient = 0.1 ÷ 0.3 ≈ 0.33. The correction accuracy is calculated as the ratio of the deviation correction amount to the calibrated deviation parameter. For example, if the pressure deviation correction amount is 0.6 MPa and the calibrated deviation is 0.3 MPa, the correction accuracy = 0.6 ÷ 0.3 = 2. The lag coefficient and correction accuracy are integrated to generate the control response characteristics. By integrating control response characteristics and amplitude adaptation factors, the dynamic control capability of the actuator is extracted. This dynamic control capability is quantified from three dimensions: lag degree, correction accuracy, and amplitude adaptability. A comprehensive dynamic control value is calculated: Comprehensive value = Amplitude adaptation factor × (1 - Lag coefficient) × Correction accuracy. For example, for pressure, with a lag factor of 0.5 MPa / %, a lag coefficient of 0.33, and a correction accuracy of 2, the comprehensive value = 0.5 × (1 - 0.33) × 2 = 0.67 MPa / %. For temperature, with a lag factor of 0.05℃ / ℃, a lag coefficient of 0.33, and a correction accuracy of 1.2, the comprehensive value = 0.05 × (1 - 0.33) × 1.2 ≈ 0.04℃ / ℃. By integrating the comprehensive dynamic control value, lag coefficient, correction accuracy, and amplitude adaptation factor of each parameter, control characteristic parameters of the actuator are generated, comprehensively reflecting the actuator's dynamic control capability and providing core parameter support for the subsequent construction of a control effect evaluation model.

[0062] Furthermore, the present invention also provides an adaptive electrical control system for supercritical extraction, used to execute the adaptive electrical control method for supercritical extraction as described above. This adaptive electrical control system for supercritical extraction includes a multi-dimensional sensing component, an electrical control module, an actuator, a data processing module, a reference parameter storage module, and a feedback adjustment module. The multi-dimensional sensing component is used to collect real-time operating parameters and component data within the supercritical extraction equipment, generate raw detection data, and transmit it to the data processing module. The data processing module is used to parse and fuse the raw detection data to generate a real-time detection sequence, and simultaneously retrieve the reference control parameter library from the reference parameter storage module. Deviation analysis and calibration are performed to generate calibrated deviation parameters, which are then transmitted to the electrical control module. The benchmark parameter storage module stores a library of benchmark control parameters for different supercritical processes and a component-state correlation model for the extraction process. The electrical control module constructs an adaptive control strategy based on the calibrated deviation parameters, generates electrical control output signals, and transmits them to the actuators. The actuators receive the electrical control output signals, execute corresponding control actions, and adjust the operating state of the supercritical extraction equipment. The feedback adjustment module collects feedback data after the actuators' actions, transmits it to the data processing module, drives iterative optimization of control parameters and control strategies, and achieves adaptive electrical control of the supercritical extraction process.

[0063] In this embodiment of the invention, multi-dimensional sensing components are deployed at key points in the extraction chamber, separation chamber, and fluid transmission pipeline of the supercritical extraction equipment, covering the core monitoring areas of various processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing, ensuring no blind spots in data acquisition. The acquisition process is initiated at fixed intervals: pressure data is acquired at 1-second intervals, temperature data is acquired in real time, component spectral data is acquired at fixed wavelength intervals, and flow rate data is acquired synchronously. The acquired real-time operating parameters and component data include pressure, temperature, flow rate, and system component spectral data. For example, during supercritical extraction, the acquired pressure data is 10.2 MPa, the temperature is 41°C, and the flow rate is 1.1 m / s. The component spectral data records spectral intensity at specific wavelength intervals. All acquired data is compiled, invalid data generated by signal interruptions during acquisition is removed, and raw detection data is generated. The raw detection data is transmitted to the data processing module via an electrical transmission line. During transmission, signal strength is monitored to ensure no data attenuation or loss, providing complete and accurate basic data for subsequent data processing.

[0064] The data processing module receives raw detection data transmitted from multi-dimensional sensing components, classifies and analyzes the data, removes noise signals from component spectral data, extracts the characteristic peaks of target components through spectral feature matching, and calculates the component concentration using the concentration-intensity calculation formula. For example, if the spectral intensity is 26 and the characteristic peak coefficient is 5.2, the calculated component concentration is 26 ÷ 5.2 = 5 g / L. It calculates inter-frame changes and steady-state characteristics for pressure, temperature, and flow rate data, integrates all analyzed data according to timestamps, and generates a real-time detection sequence. Simultaneously, it establishes a connection with the reference parameter storage module, retrieves the corresponding process reference control parameter library, and performs spatiotemporal alignment between the real-time detection sequence and the reference parameter sequence, calculating the deviation value of each parameter one by one. For example, if the real-time pressure is 10.2 MPa and the midpoint of the reference range is 9 MPa, the calculated pressure deviation is 10.2 - 9 = 1.2 MPa. The deviation data is screened and calibrated by combining dynamic thresholds, and abnormal data exceeding the dynamic thresholds are removed. The gaps are filled by the average of adjacent valid data to generate calibrated deviation parameters. The calibrated deviation parameters are then transmitted to the electrical control module to ensure that the data processing process conforms to the process requirements and the deviation calibration logic is consistent.

[0065] The baseline parameter storage module uses a dedicated storage unit to store data categorized by process type. It stores baseline control parameter libraries and component-state correlation models for three processes: supercritical extraction, supercritical drying, and supercritical anhydrous dyeing. The baseline control parameter library includes component thresholds, baseline operating parameter values, and control response thresholds for each process. For example, the upper limit for component concentration thresholds in supercritical extraction is 6 g / L, and the lower limit is 4 g / L, with a pressure baseline range of 8-10 MPa; the pressure baseline range for supercritical drying is 12-15 MPa, and the temperature baseline range is 50-55℃. The component-state correlation models for the extraction processes have different coupling coefficients set according to the process. For example, the coupling coefficient for supercritical extraction is 0.8, and the correlation model calculation formula is component concentration = (pressure × temperature) ÷ (flow rate × 0.8); the coupling coefficient for supercritical anhydrous dyeing is 1.2, and the calculation formula is adjusted accordingly. The storage unit maintains stable operation in real time, ensuring no data loss or corruption, and can quickly respond to the retrieval needs of the data processing module and electrical control module.

[0066] The electrical control module receives the calibrated deviation parameters transmitted from the data processing module, classifies and analyzes these parameters, and clarifies the deviation values, trends, and durations of component concentration, pressure, temperature, and flow rate. It then determines the deviation level using a deviation level assessment system; for example, a pressure deviation of 1.2 MPa is classified as a level three deviation. Combining this with the component-state correlation model of the extraction process retrieved from the reference parameter storage module, the module analyzes the relationship between deviations and operating parameters, optimizing the control strategy's adjustment parameters. For instance, a level three pressure deviation of 1.2 MPa indicates that the pressure regulating valve opening needs to be reduced by 15%. The optimized control strategy is then converted into control signals recognizable by the electrical control module, specifying the signal output frequency, amplitude, and duration. For example, the pressure regulating valve control signal output frequency is 50 Hz, amplitude is 6.5 V, and duration is 7 seconds; the temperature controller control signal output frequency is 50 Hz, amplitude is 4.8 V, and duration is 9 seconds. All control output signals are transmitted to the actuators in the adjustment sequence, ensuring accurate and timely transmission of control signal parameters.

[0067] The actuator comprises three core components: a pressure regulating valve, a temperature controller, and a flow rate regulating pump. These components respectively regulate the pressure, temperature, and fluid flow rate within the equipment, adapting to the control requirements of various supercritical processes. It receives control output signals from the electrical control module and executes corresponding actions according to the signal parameters. For example, upon receiving a control signal from the pressure regulating valve, it drives the valve opening to decrease by 15%, adjusting the valve's opening degree through mechanical transmission, and monitoring the valve opening in real time to ensure the adjustment range meets the signal requirements. Upon receiving a control signal from the temperature controller, it initiates a heating or cooling program, adjusting the temperature according to the amplitude and duration; for example, an amplitude of 4.8V corresponds to a temperature increase of 1.8℃, lasting 9 seconds to complete the heating action. Upon receiving a control signal from the flow rate regulating pump, it adjusts the pump speed to change the fluid transmission rate. All actuators operate synchronously, with real-time feedback on the action status during the process, ensuring precise implementation of control actions and gradually bringing the equipment's operating parameters back to baseline values.

[0068] After the actuator completes its control action, the feedback adjustment module immediately initiates data acquisition. The acquisition range is consistent with that of the multi-dimensional sensing components, collecting data on pressure, temperature, flow rate, and component concentration after the actuator's action. For example, after control, the acquired data includes a pressure of 9.3 MPa, a temperature of 40.2℃, a flow rate of 1.0 m / s, and a component concentration of 5.1 g / L. This data is then processed according to timestamps to generate feedback data. After removing abnormal data, data verification is completed. The verified feedback data is transmitted to the data processing module via a transmission line. The data processing module calculates the feedback deviation, for example, a pressure feedback deviation of 9.3 - 9 = 0.3 MPa, and a temperature feedback deviation of 40.2 - 41 = -0.8℃. Combining the feedback deviation, the calibrated deviation parameters, and the deviation level parameters, the data processing module and the electrical control module optimize the control parameters. For example, with a pressure feedback deviation of 0.3 MPa, the amplitude of the pressure regulating valve control signal is adjusted to 5.2V, and the duration is adjusted to 5 seconds. This iterative optimization of control parameters and control strategies promotes the implementation of adaptive closed-loop control.

[0069] Other aspects of this invention that are not detailed herein are all conventional techniques known to those skilled in the art.

[0070] It should be noted that the terms “comprising,” “including,” or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0071] The scope of protection of this invention is not limited to the technical solutions disclosed in the specific embodiments. Any modifications, equivalent substitutions, improvements, etc., made to the above embodiments based on the technical essence of this invention shall fall within the scope of protection of this invention.

Claims

1. An adaptive electrical control method for supercritical extraction, characterized in that, Includes the following steps: Step S1: Use multi-dimensional sensing components to collect real-time operating parameters in the supercritical extraction equipment to perform collaborative detection of component characteristics and operating status, and obtain a real-time detection sequence including the component concentration, system pressure, temperature and fluid flow rate of the extraction system; construct a component-state correlation model of the extraction process to generate a reference control parameter library corresponding to different supercritical processes; Step S2: Perform deviation analysis based on the real-time detection sequence and the reference control parameter library to generate parameter deviation time series data; perform dynamic threshold deviation calibration based on the parameter deviation time series data to obtain the calibrated deviation parameters; Step S3: Construct an adaptive control strategy based on the calibrated deviation parameters, and adjust the output signal of the electrical control module based on the adaptive control strategy to drive the actuator of the supercritical extraction equipment. Step S4: Collect equipment operating parameters and combination change data after the actuator moves in real time, and feed them back to the control strategy adjustment module to iteratively optimize the control parameters, thereby realizing adaptive electrical control of the supercritical extraction process.

2. The adaptive electrical control method for supercritical extraction according to claim 1, characterized in that, Step S1 includes: Step S11: Control the multi-dimensional sensing components to perform full-domain detection of the extraction chamber, separation chamber and fluid transmission pipeline of the supercritical extraction equipment, and collect raw detection data including system component spectral data, pressure fluctuation data, continuous temperature data and instantaneous fluid flow rate data; Step S12: Analyze the component characteristics of the raw detection data, extract the characteristic peaks of each target component through the spectral feature matching algorithm, and generate time-series data of component concentration by combining the correlation logic between component concentration and spectral intensity. Step S13: Perform time-series correlation processing on pressure fluctuation data, continuous temperature data, and instantaneous fluid velocity data, extract the inter-frame variation and steady-state characteristics of each parameter, and generate a sequence of operating state parameters; Step S14: Integrate the time series data of component concentration with the sequence of operating status parameters to construct a real-time detection sequence and clarify the spatiotemporal correspondence of each detection parameter; Step S15: Based on the operational requirements of different processes such as supercritical extraction, supercritical drying, and supercritical anhydrous dyeing, collect the standard values ​​of component concentration, standard range of pressure, standard range of temperature, and standard parameters of flow rate under the stable operating conditions of each process, and construct a component-state correlation model for the extraction process by combining the component-state coupling law. Step S16: Based on the component-state correlation model of the extraction process, classify and encode the standard parameters of each process to generate a benchmark control parameter library containing process type, component threshold, operating parameter baseline value and control response threshold.

3. The adaptive electrical control method for supercritical extraction according to claim 1, characterized in that, Step S2 includes: Step S21: Retrieve the reference parameters corresponding to the current supercritical process from the reference control parameter library, including the reference time series of component concentration, the reference fluctuation range of pressure, the reference time series of temperature, and the reference range of flow rate, and generate a reference parameter sequence; Step S22: Spatiotemporally align the real-time detection sequence with the reference parameter sequence, calculate the difference between each detection parameter and the corresponding reference parameter according to the timestamp correspondence, and generate parameter deviation time series data; Step S23: Perform time series fluctuation analysis on the parameter deviation time series data to extract the rate of change of deviation, fluctuation period and peak characteristics, and generate deviation fluctuation parameters; Step S24: Construct a dynamic threshold calibration model based on deviation fluctuation parameters, and calculate the dynamic threshold for deviation judgment by combining the composition change characteristics and operating stability requirements of the supercritical process, and optimize the deviation judgment criteria. Step S25: Based on the optimized deviation judgment criteria, filter and calibrate the parameter deviation time series data, remove abnormal deviation points, and generate calibrated deviation parameters.

4. The adaptive electrical control method for supercritical extraction according to claim 3, characterized in that, Step S24 includes: Step S241: Extract the deviation change rate, fluctuation period and peak characteristics from the deviation fluctuation parameters, and obtain the correlation coefficient between deviation fluctuation and process stability by combining the component mass transfer law and fluid flow characteristics in the supercritical extraction process; construct a deviation impact assessment model based on the correlation coefficient, analyze the degree of influence of different deviation fluctuation parameters on supercritical extraction efficiency and component separation effect, and generate deviation impact weight parameters. Step S242: Calculate the initial threshold for deviation judgment based on the deviation influence weight parameter and the target extraction requirements of the current process; collect the component concentration change rate in the supercritical extraction equipment in real time, calculate the instantaneous concentration change rate by the first derivative of the component concentration time series data, and integrate the instantaneous concentration change rates of each time stamp to generate a concentration change rate sequence. Step S243: Obtain the steady-state values ​​of the operating parameters, and detect the steady-state values ​​of the operating parameters. Extract the steady-state ranges of pressure, temperature, and flow rate parameters, calculate the fluctuation amplitude of each parameter within the steady-state range, and generate the steady-state fluctuation values ​​of the operating parameters. Based on the concentration change rate sequence and the steady-state fluctuation values ​​of the operating parameters, construct a threshold adjustment model and analyze the influence of the concentration change rate and steady-state fluctuation values ​​on the deviation judgment threshold. Step S244: Based on the influence law, the initial threshold is dynamically corrected. When the concentration change rate increases and the steady-state fluctuation value of the operating parameter exceeds the preset range, the dynamic threshold is reduced to improve the sensitivity of deviation detection. When the concentration change rate decreases and the operating parameter is in a stable state, the dynamic threshold is expanded to avoid misjudgment, thereby outputting the corrected dynamic threshold. The corrected dynamic threshold is integrated into the deviation judgment logic to optimize the deviation judgment standard and clarify the judgment rules corresponding to different degrees of deviation.

5. The adaptive electrical control method for supercritical extraction according to claim 1, characterized in that, Step S3 includes: Step S31: Classify and analyze the deviation parameters after calibration, classify them according to component concentration deviation, pressure deviation, temperature deviation and flow rate deviation, extract the magnitude, trend and duration of each type of deviation, and generate deviation classification feature data; Step S32: Based on the deviation classification feature data and combined with the control requirements of the supercritical process, construct a deviation level evaluation system, classify various deviations into different levels, and generate deviation level parameters; Step S33: Based on the deviation level parameter, retrieve the corresponding control strategy template from the benchmark control parameter library, combine it with the component-state correlation model of the extraction process, optimize the adjustment parameters of the control strategy, and generate an adaptive control strategy. Step S34: Convert the adaptive control strategy into a control signal that the electrical control module can recognize, determine the output frequency, amplitude and duration of the control signal, and generate the corresponding output signal of the electrical control module; Step S35: Transmit the output signal corresponding to the electrical control module to the actuator of the supercritical extraction equipment, drive the actuator to perform corresponding actions, including starting and stopping the pressure regulating valve, temperature controller, and flow rate regulating pump, and adjusting parameters to achieve real-time control of the extraction process.

6. The adaptive electrical control method for supercritical extraction according to claim 1, characterized in that, Step S4 includes: Step S41: After the actuator moves, control the multi-dimensional sensing components to collect the operating parameters and component change data of the supercritical extraction equipment in real time, including the adjusted pressure, temperature, flow rate data and component concentration time series data, and generate a feedback data sequence. Step S42: Verify the validity of the feedback data sequence, remove abnormal feedback data through a data consistency detection algorithm to ensure the authenticity and reliability of the feedback data, and generate verified feedback data; Step S43: Compare the verified feedback data with the corresponding benchmark parameters in the benchmark control parameter library, calculate the feedback deviation data, and analyze the control effect of the actuator's actions on the extraction process; Step S44: Based on the feedback deviation data, combined with the calibrated deviation parameters and deviation level parameters, construct a control parameter optimization model, adjust the parameters of the electrical control output signal, and optimize the adaptive control strategy; Step S45: Feed the optimized control strategy and control parameters back to the electrical control module, and iteratively execute the control actions and feedback detection process to achieve adaptive closed-loop control of the supercritical extraction process.

7. The adaptive electrical control method for supercritical extraction according to claim 6, characterized in that, Step S44 includes: Step S441: Extract the feedback deviation value, deviation change trend and steady-state deviation characteristics of each parameter from the feedback deviation data, combine the deviation level of the calibrated deviation parameter, obtain the correlation between the feedback deviation and the initial deviation, and generate the deviation correlation parameter. Step S442: Based on the deviation correlation parameters, analyze the control efficiency of the actuator's actions and calculate the control effect evaluation parameters, which reflect the actuator's ability to correct deviations. Step S443: Combining the evaluation parameters of the control effect, retrieve the optimization rules in the control parameter optimization model, and determine the adjustment direction and adjustment range of the control parameters according to the target extraction effect of the supercritical process; Step S444: Based on the adjustment direction and adjustment magnitude, optimize the frequency, amplitude and duration of the electrical control output signal, and at the same time adjust the deviation judgment logic and adjustment priority in the adaptive control strategy, thereby outputting the optimized control parameters and control strategy.

8. The adaptive electrical control method for supercritical extraction according to claim 7, characterized in that, The method for analyzing the control efficiency of the actuator's actions based on deviation correlation parameters and calculating the control effect evaluation parameters includes: Based on the feedback deviation data and the calibrated deviation parameters, the deviation correction amount for each parameter is calculated, which is the difference between the calibrated deviation parameters and the feedback deviation data, and a deviation correction amount sequence is generated. Time series analysis is performed on the deviation correction amount sequence to extract the rate of change, correction period and steady-state correction value of the deviation correction amount, and generate correction characteristic parameters; combined with the actuator's action parameters, including action amplitude, action frequency and response time, the control characteristic parameters of the actuator are obtained. Based on the correction characteristic parameters and control characteristic parameters, a control effect evaluation model is constructed to analyze the correlation between the deviation correction amount and the actuator action parameters and quantify the control effect. Based on the quantification results, control effect evaluation parameters are generated, which cover three dimensions: deviation correction efficiency, steady-state control accuracy, and response speed.

9. The adaptive electrical control method for supercritical extraction according to claim 8, characterized in that, The process of obtaining the control characteristic parameters of the actuator by combining the actuator's motion parameters, including motion amplitude, motion frequency, and response time, includes: Collect the motion parameters of the actuator, including motion amplitude, motion frequency and response time, and combine them with the deviation correction trend in the feedback deviation data to obtain the preliminary correlation parameters between the motion parameters and the deviation correction. Based on the preliminary correlation parameters, the correspondence between the motion amplitude and the deviation correction amplitude is analyzed to generate the amplitude adaptation factor; combined with the amplitude adaptation factor, motion frequency, and response time, a motion timing response model is constructed, the synchronicity characteristics of the motion response are extracted, and timing response parameters are generated. Based on the timing response parameters and deviation correlation parameters, the control lag characteristics and correction accuracy of the actuator's actions are analyzed to generate control response characteristics. By integrating the control response characteristics and amplitude adaptation factors, the dynamic control capability of the actuator is extracted, and the control characteristic parameters of the actuator are generated.

10. An adaptive electrical control system for supercritical extraction, characterized in that, The adaptive electrical control system for supercritical extraction, as described in claim 1, comprises a multi-dimensional sensing component, an electrical control module, an actuator, a data processing module, a reference parameter storage module, and a feedback adjustment module. The multi-dimensional sensing component is used to collect real-time operating parameters and component data within the supercritical extraction equipment, generate raw detection data, and transmit it to the data processing module. The data processing module parses and fuses the raw detection data to generate a real-time detection sequence. Simultaneously, it retrieves the benchmark control parameter library from the benchmark parameter storage module for deviation analysis and calibration, generating calibrated deviation parameters and transmitting them to the electrical control module. The benchmark parameter storage module stores benchmark control parameter libraries corresponding to different supercritical processes and the component-state correlation model of the extraction process. The electrical control module constructs an adaptive control strategy based on the calibrated deviation parameters, generates electrical control output signals, and transmits them to the actuators. The actuators receive the electrical control output signals, execute corresponding control actions, and adjust the operating state of the supercritical extraction equipment. The feedback adjustment module collects feedback data after the actuators' actions and transmits it to the data processing module, driving iterative optimization of control parameters and control strategies to achieve adaptive electrical control of the supercritical extraction process.

Images