An intelligent control system for mixed oil leaching production based on artificial intelligence

Abstract

This invention relates to the field of intelligent control technology in oil and fat processing, specifically to an intelligent control system for mixed oil and fat leaching production based on artificial intelligence. The system includes a dielectric fingerprint acquisition module, a component intelligent analysis module, a coupled control and decision module, a rolling predictive control module, and a feedback correction module. Specifically: the dielectric fingerprint acquisition module generates a dielectric fingerprint spectrum reflecting the state of the mixed solution components; the component intelligent analysis module analyzes the corresponding oil concentration, solvent ratio, and colloid content; and the coupled control and decision module generates a temperature-solvent coupled control strategy. This invention improves the accuracy, responsiveness, and operational stability of the mixed oil and fat leaching process by analyzing the mixed solution components online using dielectric fingerprints and performing coupled prediction and feedback correction of temperature and solvent spraying based on component changes.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for oil processing, and in particular to an intelligent control system for the production of mixed oil leaching based on artificial intelligence. Background Technology

[0002] The leaching of mixed oils is a crucial step in oil processing. It typically involves contacting the oil-containing material with a solvent, transferring the oil from the solid phase to the oil-solvent mixture. Subsequent processes such as evaporation, condensation, and recovery then achieve oil extraction and solvent recycling. Solvent leaching processes are characterized by low oil residue and suitability for continuous production. Hexane solvents are widely used in the oil leaching industry, effectively reducing residual oil levels. However, the leaching process is influenced by factors such as feed oil content, bed thickness, solvent concentration, leaching temperature, and spray distribution, leading to continuous variations in oil concentration, solvent ratio, and colloid content across different leaching stages. Dielectric spectroscopy can reflect the dielectric response of materials under different frequency electric fields. Existing research indicates that broadband dielectric spectroscopy can be used for in-situ monitoring during the process. Therefore, using dielectric response data for identifying the state of the leaching mixture provides the technical foundation for online monitoring.

[0003] Current mixed oil leaching production control mainly relies on conventional process parameters such as temperature, flow rate, and pressure for adjustment. The states of key components, such as oil concentration, solvent ratio, and colloid content, are difficult to continuously, segmentally, and in real-time acquire within the leaching unit. This results in the control system adjusting the heating jacket and solvent spray nozzles based on delayed sampling and detection results or empirical settings, failing to accurately reflect states such as insufficient oil release, solvent supply deviations, and excessive colloid accumulation. This approach leads to independent temperature and solvent spray adjustments, lacking proactive assessment of feed fluctuations, solvent concentration fluctuations, and bed thickness fluctuations, resulting in problems such as localized insufficient leaching, increased solvent consumption, higher unit energy consumption, and delayed control response. Therefore, it is necessary to propose an artificial intelligence-based intelligent control system for mixed oil leaching production. This system generates dielectric fingerprints through real-time dielectric spectrum analysis, interprets oil concentration, solvent ratio, and colloid content online, and performs temperature-solvent coupled regulation, rolling prediction, and feedback correction based on this data. This addresses the technical challenges of difficulty in real-time perception of component states during the leaching process, lack of coupled decision-making basis for adjustment parameters, and difficulty in closed-loop correction of control results. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides an intelligent control system for the production of mixed oil leaching based on artificial intelligence.

[0005] An intelligent control system for mixed oil leaching production based on artificial intelligence includes a dielectric fingerprint acquisition module, a component intelligent analysis module, a coupled control and decision-making module, a rolling predictive control module, and a feedback correction module; wherein: Dielectric fingerprint acquisition module: used to acquire the real-time dielectric spectrum of the oil-solvent mixture in the leachator, and preprocess the real-time dielectric spectrum to generate a dielectric fingerprint spectrum reflecting the state of the mixture components. Component intelligent analysis module: used to receive the dielectric fingerprint spectrum and analyze the corresponding oil concentration, solvent ratio and colloidal content according to the dielectric response characteristics of different frequency bands in the dielectric fingerprint spectrum; Coupled regulation decision module: used to determine the current leaching state based on the oil concentration, solvent ratio and colloidal content, and combined with the target constraints of oil yield, solvent recovery rate and unit energy consumption, calculate the corresponding temperature adjustment amount and solvent spraying adjustment amount, and generate temperature-solvent coupled regulation strategy; Rolling Predictive Control Module: This module is used to match the temperature-solvent coupling control strategy with the current production conditions, including fluctuations in feed, solvent concentration, and material layer thickness. It predicts the changing trends of oil concentration, solvent ratio, and colloid content in the next control cycle and outputs control commands to the heating jackets and solvent spray nozzles of each section of the leachator based on the prediction results. Feedback correction module: Used to obtain the dielectric fingerprint spectrum and corresponding component analysis results at the next moment after the control command is executed, and to correct the temperature adjustment, solvent spraying adjustment and control command for the next control cycle based on the component change deviation and prediction deviation before and after execution.

[0006] Optionally, the dielectric fingerprint acquisition module includes a spectrum acquisition unit, a detection point association unit, and a sampling synchronization unit; wherein: Spectrum acquisition unit: used to acquire the complex dielectric constant and dielectric loss of the oil-solvent mixture in the frequency range of 1kHz-10MHz; Detection point association unit: used to designate the feed section, intermediate leaching section and discharge section of the leaching unit as dielectric detection points, and to add segment identifiers to the real-time dielectric spectrum collected by each dielectric detection point; Sampling synchronization unit: used to synchronously acquire the real-time dielectric spectrum of each dielectric detection point according to a sampling period of 5s-30s, forming real-time dielectric spectrum data with sampling time and segment identifier.

[0007] Optionally, the dielectric fingerprint acquisition module further includes a spectrum preprocessing unit and a spectrum generation unit; wherein: The spectrum preprocessing unit is used to perform temperature compensation, outlier removal, baseline correction, and normalization on real-time dielectric spectrum data to obtain standardized dielectric spectrum data. The spectrum generation unit is used to arrange the complex dielectric constant, dielectric loss, and rate of change of adjacent frequency points in the standardized dielectric spectrum data in ascending order of frequency, and generate a dielectric fingerprint spectrum associated with the corresponding dielectric detection point.

[0008] Optionally, the component intelligent analysis module includes a frequency band feature extraction unit and a component mapping analysis unit; wherein: Frequency band feature extraction unit: used to extract low-frequency polarization response features, mid-frequency dielectric loss features and high-frequency attenuation response features from dielectric fingerprint spectrum; Component mapping and analysis unit: used to determine the colloid content based on low-frequency polarization response characteristics, determine the solvent ratio based on mid-frequency dielectric loss characteristics, determine the oil concentration based on high-frequency attenuation response characteristics, and output the oil concentration, solvent ratio, and colloid content associated with the corresponding dielectric detection point.

[0009] Optionally, the coupling control decision module includes a leaching state determination unit; wherein: The leaching state determination unit is used to compare the oil concentration with a preset oil concentration range, the solvent ratio with a preset solvent ratio range, and the colloid content with a preset colloid content range, and determine the current leaching state based on the comparison results; the current leaching state includes insufficient oil release state, solvent supply deviation state, high colloid accumulation state, and stable leaching state.

[0010] Optionally, the coupled control decision module further includes a target constraint calculation unit and a control quantity generation unit; wherein: Target constraint calculation unit: used to call the target constraint weights corresponding to oil yield, solvent recovery rate and unit energy consumption according to the current leaching state, and form the target constraint combination for the current control cycle; Adjustment amount generation unit: used to determine the temperature adjustment direction and solvent spraying adjustment direction according to the target constraint combination, and calculate the corresponding temperature adjustment amount and solvent spraying adjustment amount.

[0011] Optionally, the coupled control decision module further includes a coupled verification unit; wherein: Coupled verification unit: used to perform linkage verification of temperature regulation and solvent spray regulation. When the temperature regulation is directed to increase the temperature and the solvent ratio is lower than the preset lower limit of the solvent ratio, the solvent spray regulation is increased; when the solvent spray regulation is directed to increase the temperature and the temperature of the oil-solvent mixture is lower than the preset lower limit of the leaching temperature, the temperature regulation is increased; when the colloid content is higher than the preset upper limit of the colloid content, the temperature regulation is decreased and the increase of the solvent spray regulation is limited, thereby generating a temperature-solvent coupled control strategy.

[0012] Optionally, the rolling prediction control module includes a disturbance amount construction unit and a component trend prediction unit; wherein: Disturbance quantity construction unit: used to obtain the feed flow rate, feed oil content, solvent concentration and material layer thickness in the current production condition, and calculate the difference with the corresponding data of the previous control cycle to obtain the feed fluctuation, solvent concentration fluctuation and material layer thickness fluctuation. Component trend prediction unit: It is used to take the temperature-solvent coupling control strategy, feed fluctuation, solvent concentration fluctuation and material layer thickness fluctuation as prediction inputs, and output the predicted change trend of oil concentration, solvent ratio and colloid content in the next control cycle.

[0013] Optionally, the rolling prediction control module further includes a control command allocation unit; wherein: Control command allocation unit: Based on the predicted trends of oil concentration, solvent ratio and colloid content, it allocates the temperature adjustment to the heating jacket of each section of the leachator, allocates the solvent spray adjustment to the corresponding solvent spray port, and generates control commands containing target heating power, target spray flow rate and execution sequence.

[0014] Optionally, the feedback correction module includes a deviation calculation unit and a control correction unit; wherein: Deviation calculation unit: used to obtain the dielectric fingerprint spectrum and corresponding component analysis results at the next moment after the control command is executed, and to calculate the deviation of oil concentration change, solvent ratio change and colloidal content change before and after execution. At the same time, it calculates the prediction deviation between the predicted component state and the actual component analysis results in the next control cycle.

[0015] Control correction unit: used to correct the temperature adjustment amount, solvent spraying adjustment amount, and control commands for the next control cycle based on the deviations in oil concentration, solvent ratio, colloid content, and prediction.

[0016] The beneficial effects of this invention are: This invention acquires the real-time dielectric spectrum of the oil-solvent mixture in the leaching unit and generates a dielectric fingerprint spectrum, enabling online acquisition of oil concentration, solvent ratio, and colloid content during the leaching process, eliminating reliance on delayed manual sampling and detection. By analyzing the correspondence between dielectric response characteristics of different frequency bands and component states, leaching states such as insufficient oil release, solvent supply deviation, and excessive colloid accumulation can be identified in a timely manner, thus providing continuous, segmented, and real-time status data for subsequent regulation.

[0017] This invention generates a temperature-solvent coupled control strategy based on oil concentration, solvent ratio, and colloid content. It also predicts component change trends in the next control cycle by incorporating feed fluctuations, solvent concentration fluctuations, and material layer thickness fluctuations. This allows control commands for each section of the heating jacket and solvent spray nozzle to be allocated according to the actual leaching state. Simultaneously, feedback corrections are made to the temperature adjustment, solvent spray adjustment, and control commands for the next control cycle based on the executed dielectric fingerprint and component analysis results. This forms a closed-loop control process, improving the responsiveness and stability of leaching control and reducing the risks of insufficient local leaching, increased solvent consumption, and increased unit energy consumption. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the intelligent control system for mixed oil leaching production according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the coupling control decision module in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0022] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0023] like Figures 1-2 As shown, an intelligent control system for the extraction of mixed oils based on artificial intelligence includes a dielectric fingerprint acquisition module, a component intelligent analysis module, a coupled control and decision-making module, a rolling predictive control module, and a feedback correction module; wherein: Dielectric fingerprint acquisition module: used to acquire the real-time dielectric spectrum of the oil-solvent mixture in the leachator, and preprocess the real-time dielectric spectrum to generate a dielectric fingerprint spectrum reflecting the state of the mixture components. The dielectric fingerprint acquisition module includes a spectrum acquisition unit, a detection point association unit, and a sampling synchronization unit; wherein: Spectrum acquisition unit: used to acquire the complex dielectric constant and dielectric loss of the oil-solvent mixture in the frequency range of 1kHz-10MHz; Detection point association unit: used to designate the feed section, intermediate leaching section and discharge section of the leaching unit as dielectric detection points, and to add segment identifiers to the real-time dielectric spectrum collected by each dielectric detection point; Sampling synchronization unit: used to synchronously acquire the real-time dielectric spectrum of each dielectric detection point according to a sampling period of 5s-30s, forming real-time dielectric spectrum data with sampling time and segment identifier.

[0024] The dielectric fingerprint acquisition module also includes a spectrum preprocessing unit and a spectrum generation unit; wherein: The spectrum preprocessing unit is used to perform temperature compensation, outlier removal, baseline correction, and normalization on real-time dielectric spectrum data to obtain standardized dielectric spectrum data. The spectrum generation unit is used to arrange the complex dielectric constant, dielectric loss, and rate of change of adjacent frequency points in the standardized dielectric spectrum data in ascending order of frequency, and generate a dielectric fingerprint spectrum associated with the corresponding dielectric detection point.

[0025] Specifically, the spectrum preprocessing unit first performs temperature compensation processing. Since the dielectric response of the oil-solvent mixture in the leaching tank shifts with temperature, 45℃ is used as the reference temperature to uniformly correct the complex dielectric constant and dielectric loss collected at each frequency point. The temperature compensation formula is as follows: ; ,in, Indicates the first Complex permittivity after temperature compensation at each frequency point; Indicates the first The complex permittivity obtained from sampling at each frequency point; Indicates the first Dielectric loss after temperature compensation at each frequency point; Indicates the first Dielectric loss obtained from sampling at each frequency point; This indicates the temperature of the grease-solvent mixture at the current dielectric detection point; Represents the reference temperature, with a value of ; This represents the temperature compensation coefficient for the complex permittivity, with a value of [value missing]. ; This represents the dielectric loss temperature compensation coefficient, with a value of [value missing]. ; This indicates the frequency point number. This processing method corrects the dielectric spectrum collected at different temperatures to the same temperature reference, reducing the impact of temperature fluctuations on the stability of the dielectric fingerprint. After temperature compensation, the spectrum preprocessing unit performs abnormal frequency point removal. Local deviation values ​​are calculated for the compensated complex dielectric constant and dielectric loss of adjacent frequencies within the same sampling period. When the deviation value of any frequency point exceeds a preset abnormal threshold, that frequency point is marked as an abnormal frequency point and removed. The abnormality determination formula is:

[0026] When satisfied At that time, the judgment of the first Several frequency points are abnormal frequency points; among them... Indicates the first Local deviation values ​​at each frequency point; Indicates the first The average complex permittivity of each frequency point, two frequencies preceding and following each other. Indicates the first The average dielectric loss of each frequency point, two frequency points before and after it. Indicates the first The standard deviation of the complex permittivity of each frequency point is the difference between the two adjacent frequencies. Indicates the first The standard deviation of dielectric loss for each frequency point is calculated from the two frequency points immediately preceding and following it; 0.001 is used as a stabilizing term to prevent the denominator from being zero. This processing removes abnormal frequency points caused by sensor jitter, electromagnetic interference, and contact state fluctuations, ensuring the continuity and reliability of subsequent spectrum generation. After removing abnormal frequency points, the spectrum preprocessing unit performs baseline correction on the remaining frequency points. Using the reference dielectric spectrum pre-acquired under unloaded solvent conditions as baseline data, the baseline values ​​of the corresponding frequency points are subtracted from the temperature-compensated complex dielectric constant and dielectric loss. The baseline correction formula is: ; ;in, Indicates the first Complex permittivity after baseline correction at each frequency point; Indicates the first The reference complex permittivity at each frequency point; Indicates the first Dielectric loss after baseline correction at each frequency point; Indicates the first The baseline dielectric loss at each frequency point is used as a reference. This processing removes the baseline response offset caused by the detection probe, solvent background, and fixed pipeline environment, making the dielectric spectrum more focused on reflecting the compositional state of the oil-solvent mixture. After baseline correction, the spectrum preprocessing unit performs normalization, mapping the complex dielectric constant and dielectric loss at each frequency point to a unified numerical range. The normalization formula is: ; ,in, Indicates the first The complex permittivity normalized to each frequency point; Indicates the first Dielectric loss after frequency point normalization; It represents the minimum value of the complex permittivity after baseline correction within the same sampling period; This represents the maximum value of the complex permittivity after baseline correction within the same sampling period; This represents the minimum dielectric loss after baseline correction within the same sampling period; This represents the maximum value of dielectric loss after baseline correction within the same sampling period; 0.001 is the normalized stability term; through this processing, the numerical scale differences between different detection segments and different sampling batches can be eliminated, so that dielectric data has a unified spectral representation basis.

[0027] Finally, the spectrum generation unit combines the normalized complex permittivity, normalized dielectric loss, rate of change of complex permittivity, and rate of change of dielectric loss at the same dielectric detection point and the same sampling time in ascending order of frequency to form a dielectric fingerprint spectrum.

[0028] Intelligent component analysis module: used to receive dielectric fingerprint spectrum and analyze the corresponding oil concentration, solvent ratio and colloidal content based on the dielectric response characteristics of different frequency bands in the dielectric fingerprint spectrum; The component intelligent analysis module includes a frequency band feature extraction unit and a component mapping analysis unit; wherein: The frequency band feature extraction unit receives the dielectric fingerprint spectrum output by the dielectric fingerprint acquisition module and divides it into low-frequency, mid-frequency, and high-frequency bands according to frequency range; the low-frequency band is 1kHz-100kHz, the mid-frequency band is 100kHz-1MHz, and the high-frequency band is 1MHz-10MHz. The unit extracts low-frequency polarization response features, mid-frequency dielectric loss features, and high-frequency attenuation response features from these three bands. The low-frequency polarization response features characterize the interfacial polarization response of colloidal substances in an oil-solvent mixture under a low-frequency electric field. The unit reads the normalized complex dielectric constant from the low-frequency band and calculates the mean of the normalized complex dielectric constant to obtain the low-frequency polarization response features. The calculation formula is as follows: ,in, Indicates the characteristics of low-frequency polarization response; Represents the set of frequency points in the low-frequency band; Indicates the number of frequency points in the low-frequency band; Indicates the frequency point number; Indicates the first The normalized complex dielectric constant at each frequency point. The mid-frequency dielectric loss characteristic is used to characterize the dielectric energy dissipation state of the oil-solvent mixture under a mid-frequency electric field; the frequency band feature extraction unit reads the normalized dielectric loss within the mid-frequency band and calculates the mean normalized dielectric loss of the mid-frequency band to obtain the mid-frequency dielectric loss characteristic. The calculation formula is: ,in, Indicates the characteristics of dielectric loss at intermediate frequencies; Represents the set of mid-frequency band frequencies; Indicates the number of frequency points in the mid-band; Indicates the first The dielectric loss is normalized at each frequency point. The high-frequency attenuation response characteristics are used to characterize the attenuation trend of the dielectric response of the oil-solvent mixture as the frequency increases under a high-frequency electric field. The frequency band feature extraction unit reads the normalized complex dielectric constant from the high-frequency band and calculates the high-frequency attenuation response characteristics based on the variation amplitude of adjacent frequency points in the logarithmic frequency coordinate. The calculation formula is: ,in, Indicates the characteristics of high-frequency attenuation response; Represents the set of high-frequency bands; Indicates the number of frequency points in the high-frequency band; Indicates the first The frequency of each frequency point; Indicates the first The frequency of each frequency point; Indicates the first The natural logarithm of the frequency at each frequency point; Indicates the first The complex permittivity after normalization at each frequency point.

[0029] The component mapping analysis unit is used to analyze the colloidal content, solvent ratio, and oil concentration based on the low-frequency polarization response characteristics, mid-frequency dielectric loss characteristics, and high-frequency decay response characteristics. Specifically, the component mapping analysis unit calls pre-calibrated component mapping coefficients to map the low-frequency polarization response characteristics to colloidal content, the mid-frequency dielectric loss characteristics to solvent ratio, and the high-frequency decay response characteristics to oil concentration. The calculation formula is as follows: ; ; ,in, Indicates colloid content; Indicates the solvent ratio; Indicates the concentration of oil and fat; This represents the colloidal content mapping coefficient corresponding to the low-frequency polarization response characteristics; Indicates the intercept of the colloid content mapping; This represents the solvent ratio mapping coefficient corresponding to the mid-frequency dielectric loss characteristics; Indicates the solvent ratio mapping intercept; This represents the oil concentration mapping coefficient corresponding to the high-frequency attenuation response characteristics. The mapping intercept represents the oil concentration. The mapping coefficient and the mapping intercept are determined through offline calibration samples, which include oil-solvent mixtures with known oil concentrations, solvent ratios, and colloid contents. By dividing the dielectric fingerprint spectrum into low-frequency, mid-frequency, and high-frequency bands, and extracting low-frequency polarization response characteristics, mid-frequency dielectric loss characteristics, and high-frequency attenuation response characteristics respectively, the colloid content, solvent ratio, and oil concentration can be identified with distinct dielectric response sources.

[0030] Coupled regulation decision module: It is used to determine the current leaching state based on oil concentration, solvent ratio and colloidal content, and calculate the corresponding temperature regulation amount and solvent spray regulation amount in combination with the target constraints of oil yield, solvent recovery rate and unit energy consumption, and generate temperature-solvent coupled regulation strategy. The coupled control decision module includes a leaching state determination unit; wherein: Leaching state determination unit: It is used to compare the oil concentration with the preset oil concentration range, the solvent ratio with the preset solvent ratio range, and the colloid content with the preset colloid content range, and determine the current leaching state based on the comparison results; the current leaching state includes insufficient oil release state, solvent supply deviation state, high colloid accumulation state, and stable leaching state.

[0031] The aforementioned preset oil concentration range, preset solvent ratio range, and preset colloid content range are not fixed empirical values, but are established based on historical production samples formed during the stable production phase of the target leaching process. Specifically, data from no fewer than 100 consecutive production batches under stable production conditions are continuously selected, and the corresponding oil concentration, solvent ratio, and colloid content for each batch are statistically analyzed. Their mean and standard deviation are calculated separately. The mean is used as the center, and twice the standard deviation is used as the normal fluctuation range to establish each preset range. The calculation formulas are as follows: Oil concentration range center value: ; Standard deviation of oil concentration: ; Preset range for oil concentration: ;in, Indicates the first The oil concentration corresponding to a stable production batch; This represents the average concentration of oils and fats. Indicates the standard deviation of oil concentration; Indicates the preset oil concentration range; This indicates the number of stable production samples, with a value of no less than 100; the solvent ratio range and colloidal content range are established using the same method; the solvent ratio preset range is: ; Preset range for colloid content: ;in, This represents the average solvent ratio; Indicates the standard deviation of the solvent ratio; Indicates a preset solvent ratio range; This represents the average colloid content; Indicates the standard deviation of colloid content; It indicates the preset colloidal content range; by using historical statistical data from the stable production stage as the basis for judgment, each preset range can directly correspond to the actual process capability of the current production line, rather than using a fixed empirical threshold. Therefore, different production devices can form corresponding judgment standards after being calibrated with historical data.

[0032] The leaching state determination unit first determines whether the oil concentration, solvent ratio, and colloid content fall within the corresponding preset ranges, and generates three determination flags. The expressions for the corresponding determination flags are as follows: Indicators for determining oil concentration: ; Solvent ratio determination criteria: ; Criteria for determining colloid content: ;in, Indicates the current oil concentration; Indicates the current solvent ratio; Indicates the current colloid content; These represent the lower and upper limits of the preset range for oil concentration, respectively. These represent the lower and upper limits of the preset solvent ratio range, respectively. This indicates the upper limit of the preset range for colloid content; Indicates the indicator for determining oil concentration; Indicates the solvent ratio determination mark.

[0033] After obtaining the three determination flags, the leaching state determination unit determines the current leaching state according to a fixed determination rule, the specific rule being: when When this occurs, the current state is determined to be one of insufficient oil release. This state indicates that the current oil concentration is lower than the range corresponding to the stable leaching stage, meaning that the rate at which oil enters the mixture per unit time is lower than normal. ,and or At that time, it is determined that the current state is a solvent supply deviation. Among them, This indicates that the solvent ratio is too low. This indicates that the solvent ratio is too high; both situations indicate that the solvent supply deviates from normal operating conditions. When ,and At this point, the current state is determined to be one of excessive colloid accumulation. While the oil concentration and solvent ratio are within the normal range, the colloid content exceeds the normal range, indicating that colloids are gradually accumulating in the mixture, requiring further adjustment. When the following conditions are met... When the current state is determined to be a stable leaching state, it means that the oil concentration, solvent ratio, and colloid content are all within the normal fluctuation range corresponding to the stable production stage.

[0034] The coupled control decision module also includes a target constraint calculation unit and a regulation quantity generation unit; wherein: Target constraint calculation unit: used to call the target constraint weights corresponding to oil yield, solvent recovery rate and unit energy consumption according to the current leaching state, and form the target constraint combination for the current control cycle; Adjustment amount generation unit: used to determine the temperature adjustment direction and solvent spraying adjustment direction according to the target constraint combination, and calculate the corresponding temperature adjustment amount and solvent spraying adjustment amount.

[0035] Specifically, the target constraint calculation unit pre-sets weight combinations corresponding to four types of leaching states. When the current leaching state is insufficient oil release, the oil yield weight is 0.55, the solvent recovery rate weight is 0.25, and the unit energy consumption weight is 0.20, so that the current control cycle prioritizes meeting the demand for increased oil release. When the current leaching state is a solvent supply deviation state, the oil yield weight is 0.30, the solvent recovery rate weight is 0.50, and the unit energy consumption weight is 0.20, so that the current control cycle prioritizes correcting the solvent supply deviation. When the current leaching state is a high colloid accumulation state, the oil yield weight is 0.20, the solvent recovery rate weight is 0.35, and the unit energy consumption weight is 0.45, so that the current control cycle prioritizes limiting the energy consumption increase caused by excessive heating and excessive spraying. When the current leaching state is a stable leaching state, the oil yield weight is 0.40, the solvent recovery rate weight is 0.30, and the unit energy consumption weight is 0.30, so that the current control cycle maintains a balance between yield, recovery rate, and energy consumption. In each of the above weight combinations, the sum of the three weights is 1. The combination of objective constraints is expressed as follows: ,in, This represents the combination of target constraints for the current control cycle; Indicates the weight of oil yield; Indicates the weight of solvent recovery rate; This represents the weight of unit energy consumption. Further calculations are made for oil yield deviation, solvent recovery rate deviation, and unit energy consumption deviation. Oil yield deviation characterizes the difference between the current oil yield and the target oil yield; solvent recovery rate deviation characterizes the difference between the current solvent recovery rate and the target solvent recovery rate; and unit energy consumption deviation characterizes the degree to which the current unit energy consumption exceeds the target unit energy consumption. The corresponding expressions are: ; ; ,in, This indicates deviation in oil yield; Indicates the target oil yield; Indicates the current oil yield; This indicates a deviation in solvent recovery rate; Indicates the target solvent recovery rate; This indicates the current solvent recovery rate; Indicates the deviation in unit energy consumption; Indicates the target unit energy consumption; This represents the current unit energy consumption; the above deviation value serves as the input for the adjustment quantity generation unit to determine the adjustment direction and adjustment range.

[0036] The adjustment generation unit is used to determine the temperature adjustment direction and solvent spraying adjustment direction based on the target constraint combination and three types of target deviations. Specifically, when the oil yield deviation corresponding to the insufficient oil release state is positive, the temperature adjustment direction is to increase the temperature, and the solvent spraying adjustment direction is to increase the spraying direction. When the solvent recovery rate deviation corresponding to the solvent supply deviation state is positive and the current solvent ratio is lower than the lower limit of the preset solvent ratio range, the solvent spraying adjustment direction is to increase the spraying direction. When the current solvent ratio is higher than the upper limit of the preset solvent ratio range, the solvent spraying adjustment direction is to decrease the spraying direction. When the unit energy consumption deviation corresponding to the high colloid accumulation state is positive, the temperature adjustment direction is to decrease the temperature, and the solvent spraying adjustment direction is to limit the increase. When the current leaching state is a stable leaching state, the temperature adjustment direction and the solvent spraying adjustment direction are fine-tuned according to the weighted result of the target deviation.

[0037] The expressions for calculating the temperature regulation and solvent spray regulation by the regulation generation unit are as follows: ; ;in, This indicates the temperature adjustment amount for the current control cycle, in units of... ; This indicates the solvent spray adjustment amount for the current control cycle, in units of... ; This represents the temperature regulation coefficient, with a value of [value missing]. ; This represents the solvent spray adjustment coefficient, with a value of [value missing]. ; Indicates the weight of oil yield; Indicates the weight of solvent recovery rate; Indicates the weight of unit energy consumption; This indicates deviation in oil yield; This indicates a deviation in solvent recovery rate; Indicates the deviation in unit energy consumption; Represents the amplitude limiting function, when Time output ,when Time output ,when Time output Through limiting processing, the temperature adjustment within a single control cycle is restricted to a certain value. to Within the range, the solvent spray adjustment amount is limited to to Within the range. By applying the target constraint weights corresponding to oil yield, solvent recovery rate, and unit energy consumption based on the current leaching state, the adjustment targets under different leaching states can be given clear priorities. Then, by calculating the temperature adjustment amount and solvent spraying adjustment amount based on the target deviation, the leaching state determination result can be converted into quantifiable basic control parameters, providing a stable decision basis for subsequent temperature-solvent coupled control.

[0038] The coupled control decision module also includes a coupled verification unit; wherein: Coupled verification unit: used to perform linkage verification of temperature regulation and solvent spray regulation. When the temperature regulation is directed to increase the temperature and the solvent ratio is lower than the preset lower limit of the solvent ratio, the solvent spray regulation is increased; when the solvent spray regulation is directed to increase the temperature and the temperature of the oil-solvent mixture is lower than the preset lower limit of the leaching temperature, the temperature regulation is increased; when the colloid content is higher than the preset upper limit of the colloid content, the temperature regulation is decreased and the increase of the solvent spray regulation is limited, thereby generating a temperature-solvent coupled control strategy.

[0039] Specifically, let the initial temperature adjustment be denoted as... The initial solvent spray adjustment amount is denoted as ;in, Indicates a rise in temperature. Indicates a drop in temperature; This indicates an increase in the solvent spray volume. This indicates a reduction in solvent spray volume; the temperature adjustment after verification output from the coupled verification unit is denoted as... The adjusted solvent spray volume after verification is recorded as follows: The preset lower limit of the leaching temperature is taken as follows: The preset upper limit of the leaching temperature is set to The target leaching temperature is taken as This temperature range is determined based on continuous operating data of the oil-solvent mixture during a stable production phase. Specifically, in production batches where oil yield, solvent recovery rate, and unit energy consumption are all within stable ranges, the main temperature distribution range of the mixture is selected as the preset leaching temperature range, and its median value is used as the target leaching temperature. (Lower limit of solvent ratio) upper limit of solvent ratio and the upper limit of colloid content We will continue to use the preset interval boundaries previously determined based on stable production samples, namely: ; ; ,in, Indicates the lower limit of the preset solvent ratio; This indicates the upper limit of the preset solvent ratio; This indicates the preset upper limit of colloid content; This represents the average solvent ratio in a stable production sample. This represents the standard deviation of the solvent ratio in a stable production sample. This represents the average colloid content in a stable production sample. This represents the standard deviation of colloid content in a stable production sample.

[0040] When the initial temperature adjustment is set to increase the temperature, and the current solvent ratio is below the preset lower limit, it indicates that the temperature increase will enhance the solvent consumption trend in the oil-solvent mixture. Since the current solvent ratio is already below the stable range, the coupling verification unit adds a solvent compensation amount to the initial solvent spray adjustment amount. The solvent compensation amount is calculated based on the degree of deviation of the current solvent ratio from the lower limit. ; ;in, This indicates the solvent spray compensation amount, in units of... ; Indicates the current solvent ratio; Indicates the preset lower limit of solvent ratio; 6 indicates the corresponding spray compensation coefficient for solvent ratio deviation; 4 indicates the maximum allowable spray compensation amount for a single coupling verification, in units of... ; This indicates the solvent spray adjustment amount after verification, in units of... ; This indicates the initial solvent spray adjustment amount, in units of... ; Represents the amplitude limiting function, when Time output ,when Time output ,when Time output The rule limits the solvent spray compensation amount to... Within the range, and limit the adjusted solvent spray volume after verification to within to Within the range, the spray adjustment amplitude remains stable within a single control cycle.

[0041] When the initial solvent spray adjustment is set to increase, and the current temperature of the oil-solvent mixture is below the preset lower limit of the leaching temperature, it indicates that increasing the spray will further remove heat. Since the current temperature is already below the stable leaching temperature range, the coupling verification unit adds a temperature compensation amount to the initial temperature adjustment. The temperature compensation amount is calculated based on the degree of deviation of the current temperature from the lower limit. ; ;in, This indicates the amount of temperature compensation, in units of... ; This indicates the current temperature of the oil-solvent mixture, in units of... ; This indicates the preset lower limit of the leaching temperature, with a value of [value missing]. ; This represents the temperature rise compensation coefficient corresponding to the temperature deviation; 2 represents the maximum allowable temperature rise compensation amount for a single coupling check, in units of... ; This indicates the temperature adjustment amount after verification, in units of... ; This indicates the initial temperature adjustment amount, in units of... The rule limits the amount of temperature compensation to... Within the range, and limit the verified temperature adjustment to within to Within this range, the temperature compensation when the spray increases has a clear boundary.

[0042] When the current colloid content exceeds the preset upper limit, it indicates that the colloid accumulation in the oil-solvent mixture has exceeded the stable production range. At this point, further heating or significantly increasing solvent spraying will exacerbate the fluctuations in the mixture's state. The coupled verification unit performs suppression correction on the temperature adjustment amount and limits the positive increase in the solvent spraying adjustment amount. The colloid over-limit suppression amount is calculated based on the degree to which the colloid content exceeds the upper limit: ; ; ;in, This indicates the amount of colloid exceeding the limit of inhibition, in units of... ; Indicates the current colloid content; 1 indicates the preset upper limit of colloidal content; 2 indicates the temperature suppression coefficient corresponding to the deviation of colloidal content; 1.5 indicates the maximum temperature adjustment that can be deducted in a single coupling check, in units of 1. ; The value represents the solvent spray adjustment amount after verification, in percentage (%); 3 represents the maximum allowable positive spray increase in a single control cycle under colloid exceeding the limit, in percentage (%). This rule weakens the temperature rise when the colloid content exceeds the limit, and limits the positive increase in solvent spray amount to a certain level. Within.

[0043] When multiple verification conditions are simultaneously met, the coupled verification unit executes the verification in the following order: first, verification of excessive colloid content; second, verification of lower temperature limit; and third, verification of lower solvent ratio limit. This priority order is based on the fact that excessive colloid content directly affects the stability of the mixture, thus limiting the increase in temperature and spray volume; temperatures below the lower limit weaken the continuity of the leaching process, requiring temperature compensation before increasing spray volume; and solvent ratios below the lower limit are used to compensate for insufficient solvent during the heating process. By fixing the verification order, conflicts in adjustment direction caused by multiple verification rules acting simultaneously can be avoided. After completing the above-mentioned linked verification, the coupled verification unit adjusts the verified temperature amount... And the calibrated solvent spray adjustment amount The combination is a temperature-solvent coupled control strategy.

[0044] Rolling Predictive Control Module: This module is used to match the temperature-solvent coupling control strategy with the current production conditions, including fluctuations in feed, solvent concentration, and material layer thickness. It predicts the changing trends of oil concentration, solvent ratio, and colloid content in the next control cycle and outputs control commands to the heating jackets and solvent spray nozzles of each section of the leachator based on the prediction results. The rolling predictive control module includes a disturbance quantity construction unit and a component trend prediction unit. The disturbance quantity construction unit is used to obtain the feed flow rate, feed oil content, solvent concentration and material layer thickness in the current control cycle, and calculate the difference with the corresponding data in the previous control cycle to convert the changes in production conditions into a unified disturbance quantity. The control cycle is 30s-120s, which is an integer multiple of the sampling cycle of the dielectric fingerprint acquisition module.

[0045] Specifically, the disturbance construction unit first calculates the feed fluctuation, which is determined by the changes in feed flow rate and feed oil content. The calculation formula is as follows: ,in, Indicates the fluctuation in feed volume; This indicates the feed flow rate for the current control cycle; This indicates the feed flow rate in the previous control cycle; This represents the baseline feed flow rate during the stable production phase. This indicates the oil content of the feed during the current control cycle; This indicates the oil content of the feed in the previous control cycle; The baseline feed oil content represents the stable production stage; 0.6 represents the weighting of feed flow rate variation; and 0.4 represents the weighting of feed oil content variation. Then, the solvent concentration fluctuation and the bed thickness fluctuation are calculated separately using the following formulas: ; ,in, This indicates the amount of solvent concentration fluctuation; This indicates the solvent concentration in the current control cycle; This indicates the solvent concentration in the previous control cycle; Indicates the baseline solvent concentration during the stable production phase; This indicates the amount of fluctuation in the material layer thickness; This indicates the material layer thickness during the current control cycle; Indicates the thickness of the material layer in the previous control cycle; This represents the baseline material layer thickness during the stable production phase. Through the above normalization process, feed fluctuations, solvent concentration fluctuations, and material layer thickness fluctuations are converted into dimensionless disturbance inputs.

[0046] The component trend prediction unit receives the temperature-solvent coupling control strategy and the aforementioned perturbation, and combines them with the current component analysis results to form the prediction input. The prediction input includes the current oil concentration, current solvent ratio, current colloid content, verified temperature adjustment, verified solvent spray adjustment, feed fluctuation, solvent concentration fluctuation, and bed thickness fluctuation. The change in oil concentration in the next control cycle is calculated using the following formula: The change in solvent ratio during the next control cycle is calculated using the following formula: The change in colloid content during the next control period is calculated using the following formula: ;in, This indicates the predicted change in oil concentration during the next control period. This indicates the predicted change in solvent ratio within the next control period; This indicates the predicted change in colloid content within the next control period; This indicates the temperature adjustment amount after verification; This indicates the solvent spray adjustment amount after verification; Indicates the fluctuation in feed volume; This indicates the amount of solvent concentration fluctuation; This indicates the fluctuation in material layer thickness. The above coefficients were calibrated from historical samples during the stable production and controlled adjustment phases, and positive and negative relationships were set according to the response directions of oil concentration, solvent ratio, and colloid content to temperature adjustment, spray adjustment, and production disturbances.

[0047] The component trend prediction unit obtains the predicted component state for the next control period based on the current component analysis results and predicted changes. The calculation formula is as follows: ; ; ,in, This indicates the predicted oil concentration for the next control cycle; This indicates the predicted solvent ratio for the next control cycle; This indicates the predicted colloidal content for the next control cycle; Indicates the current oil concentration; Indicates the current solvent ratio; This indicates the current colloid content. Therefore, the predicted trend output by the component trend prediction unit includes the predicted change in oil concentration. Predicted changes in solvent ratio Predicted changes in colloid content And predicting the oil concentration for the next control cycle. Predicting solvent ratio and predicting colloidal content The prediction result serves as the basis for subsequent control command allocation.

[0048] The rolling predictive control module also includes a control command allocation unit; wherein: Control command allocation unit: Based on the predicted trends of oil concentration, solvent ratio and colloid content, it allocates the temperature adjustment to the heating jacket of each section of the leachator, allocates the solvent spray adjustment to the corresponding solvent spray port, and generates control commands containing target heating power, target spray flow rate and execution sequence.

[0049] Specifically, the leachator is divided into a feeding section, an intermediate leaching section, and a discharge section along the material flow direction. Each of the three leaching sections is equipped with a corresponding heating jacket and solvent spray nozzle. The control command distribution unit determines the adjustment priority based on the predicted component change trends of each section: when the predicted oil concentration in a certain leaching section is lower than the corresponding target range, that section is designated as the key section for temperature adjustment; when the predicted solvent ratio in a certain leaching section is lower than the corresponding target range, that section is designated as the key section for solvent spray adjustment; when the predicted colloid content in a certain leaching section is higher than the corresponding target range, the temperature adjustment priority of that section is reduced and the spray increase in that section is limited.

[0050] During the temperature regulation allocation process, the control command allocation unit distributes the temperature regulation amount according to the predicted oil concentration deviation of the feeding section, intermediate leaching section, and discharge section. If the predicted oil concentration deviation is the largest in the feeding section, then 50% of the temperature regulation amount is allocated to the heating jacket of the feeding section, 30% to the intermediate leaching section, and 20% to the discharge section. If the predicted oil concentration deviation is the largest in the intermediate leaching section, then 50% is allocated to the intermediate leaching section, and 25% is allocated to both the feeding and discharge sections. If the predicted oil concentration deviation is the largest in the discharge section, then 50% is allocated to the discharge section, 30% to the intermediate leaching section, and 20% to the feeding section. In this way, the temperature regulation is preferentially applied to the leaching section where the oil release is most insufficient.

[0051] During the solvent spray adjustment process, the control command allocation unit allocates the solvent spray adjustment amount according to the predicted solvent ratio deviation of each segment. If the predicted solvent ratio of a certain leaching segment is lower than the lower limit of the corresponding target range, the spray flow rate of that segment's spray nozzle is increased; if the predicted solvent ratio of a certain leaching segment is higher than the upper limit of the corresponding target range, the spray flow rate of that segment's spray nozzle is decreased; if the predicted solvent ratio is within the target range, the original spray flow rate of that segment's spray nozzle is maintained. When multiple leaching segments have solvent ratio deviations, the control command allocation unit determines the spray adjustment sequence according to the absolute value of the deviation from largest to smallest.

[0052] Control commands are generated based on the above allocation results. These commands include the target heating power of each heating jacket segment, the target spray flow rate of each solvent spray nozzle, and the execution sequence. The target heating power is calculated by adding the temperature adjustment amount of the corresponding segment to the current heating power, and the target spray flow rate is calculated by adding the solvent spray adjustment amount of the corresponding segment to the current spray flow rate. The execution sequence is set to first adjust the target spray flow rate of the solvent spray nozzle, then adjust the target heating power of the corresponding heating jacket, ensuring the solvent supply reaches a stable range before temperature compensation. During execution, the control command allocation unit writes the target heating power and target spray flow rate of each segment into the corresponding control command table, and adds a control cycle number, leaching segment identifier, and execution time identifier, forming control commands that can be sent to the leaching unit's field control terminal. These control commands are only used in the current control cycle and are regenerated in the next control cycle based on the new dielectric fingerprint spectrum and predicted component trends.

[0053] Feedback correction module: Used to obtain the dielectric fingerprint spectrum and corresponding component analysis results at the next moment after the control command is executed, and to correct the temperature adjustment, solvent spraying adjustment and control command for the next control cycle based on the component change deviation and prediction deviation before and after execution.

[0054] The feedback correction module includes a deviation calculation unit and a control correction unit. The deviation calculation unit receives the dielectric fingerprint spectrum at the next moment after the control command is executed and calls the component intelligent analysis module to obtain the oil concentration, solvent ratio, and colloidal content at the next moment. Simultaneously, it reads the oil concentration, solvent ratio, and colloidal content before the control command is executed, as well as the predicted component state for the next control cycle output by the rolling prediction control module. The deviation calculation unit first calculates the component change deviation before and after execution, using the following formula: ; ; ,in, This indicates the deviation in the change of oil concentration; This indicates the deviation in solvent ratio variation; This indicates the deviation in colloid content variation; This indicates the oil concentration before the control command is executed; Indicates the solvent ratio before the control command is executed; Indicates the colloid content before the control command is executed; This indicates the oil concentration after the control command is executed; This indicates the solvent ratio after the control command is executed; This indicates the colloid content after the control command is executed.

[0055] The deviation calculation unit then calculates the prediction deviation. The prediction deviation is used to characterize the difference between the predicted component state output by the rolling prediction control module and the actual component analysis result after execution. The calculation formula is as follows: ; ; ;in, This indicates a deviation in the predicted oil concentration. This indicates the deviation in the solvent ratio prediction; This indicates a deviation in the predicted colloid content. This indicates the predicted oil concentration for the next control cycle; This indicates the predicted solvent ratio for the next control cycle; This indicates the predicted colloid content for the next control cycle.

[0056] The control correction unit corrects the temperature adjustment and solvent spray adjustment for the next control cycle based on the above deviations. Temperature correction is mainly determined based on deviations in oil concentration, colloid content, and oil concentration prediction. Solvent spray correction is mainly determined based on solvent ratio deviations and solvent ratio prediction. The correction formula is as follows: ; ;in, This indicates the temperature adjustment amount after correction in the next control cycle; This indicates the solvent spray adjustment amount after correction in the next control cycle; This indicates the temperature adjustment amount after the current control cycle verification; This indicates the solvent spray adjustment amount after the current control cycle verification; 0.3 represents the temperature correction coefficient corresponding to the prediction deviation of oil concentration; 0.3 represents the temperature suppression coefficient corresponding to the prediction deviation of colloid content; 0.2 represents the temperature following coefficient corresponding to the prediction deviation of oil concentration; 0.5 represents the spray correction coefficient corresponding to the prediction deviation of solvent ratio; 0.2 represents the spray following coefficient corresponding to the prediction deviation of solvent ratio. Represents the amplitude limiting function, when Time output ,when Time output ,when Time output When the predicted oil concentration deviation is positive, it indicates that the actual oil concentration after execution is higher than the predicted value. The control correction unit increases the correction magnitude of the temperature adjustment in the next control cycle. When the predicted colloid content deviation is positive, it indicates that the actual colloid content after execution is higher than the predicted value. The control correction unit decreases the temperature adjustment in the next control cycle to prevent further accumulation of colloids. When the predicted solvent ratio deviation is positive, it indicates that the actual solvent ratio after execution is higher than the predicted value. The control correction unit increases the reverse adjustment sensitivity of the spray correction in the next control cycle to bring the solvent spray volume back to the stable range. The control correction unit updates the control command for the next control cycle based on the corrected temperature adjustment and solvent spray adjustment. Specifically, the control correction unit updates the control command for the next control cycle. As the input quantity for calculating the target heating power of each segment of the heating jacket in the next control cycle, This serves as the input for calculating the target spray flow rate of each solvent spray nozzle in the next control cycle.

[0057] This embodiment is applied to a continuous mixed oil leaching production line. The leaching unit is divided into a feeding section, an intermediate leaching section, and a discharge section along the material flow direction. Each section is equipped with a heating jacket and a solvent spray nozzle. The production object is a pre-treated mixed oily material. During the leaching process, an oil-solvent mixture is formed. The system uses oil concentration, solvent ratio, and colloid content as the criteria for judging the leaching state and performs linkage control on the heating jacket and solvent spray nozzle.

[0058] Before the system was put into operation, 120 production batches that had been running continuously and stably were selected as calibration samples. The stable leaching conditions were found to have an oil concentration range of 24.0%-28.5%, a solvent ratio range of 60.0%-66.0%, and a colloid content range of 0.65%-1.35%. Simultaneously, the stable leaching temperature range was determined to be 50℃-60℃, with a target leaching temperature of 55℃. These ranges served as the benchmark for subsequent leaching condition determination and adjustment.

[0059] During system operation, the dielectric fingerprint acquisition module collects real-time dielectric spectra of the oil-solvent mixture at the feeding section, intermediate leaching section, and discharge section, with a sampling frequency range of 1kHz-10MHz and a sampling period of 15s. Each set of real-time dielectric spectra includes the complex dielectric constant and dielectric loss, along with the sampling time and section identifier. After performing temperature compensation, outlier removal, baseline correction, and normalization on the real-time dielectric spectra, the system generates dielectric fingerprint maps corresponding to each leaching section.

[0060] Within a certain control cycle, the intelligent component analysis module, based on the dielectric fingerprint spectrum, obtains the following results: Oil concentration in the feed section is 22.8%, solvent ratio is 58.6%, and colloid content is 0.92%; oil concentration in the intermediate leaching section is 24.9%, solvent ratio is 60.8%, and colloid content is 1.08%; oil concentration in the discharge section is 27.6%, solvent ratio is 63.5%, and colloid content is 1.18%. The system compares these component data with preset ranges and determines that the oil concentration in the feed section is below 24.0% and the solvent ratio is below 60.0%, indicating that the current leaching state is one of insufficient oil release accompanied by low solvent supply.

[0061] In this state, the coupled control decision module invokes the target constraint weights corresponding to the insufficient oil release state, where the oil yield weight is 0.55, the solvent recovery rate weight is 0.25, and the unit energy consumption weight is 0.20. Based on the judgment that the current oil concentration and solvent ratio are low, the system generates an initial temperature adjustment of +2.0℃ and an initial solvent spray adjustment of +5.0%. Since the current solvent ratio is lower than the preset lower limit and the temperature adjustment points to a temperature increase, after coupling verification, the solvent spray adjustment is corrected to +6.5%, and the temperature adjustment remains at +2.0℃, forming the temperature-solvent coupled control strategy for the current control cycle.

[0062] The rolling predictive control module further collects current production condition disturbance data. During this control cycle, the feed flow rate increases by 3.2% compared to the previous control cycle, the feed oil content decreases by 1.1%, the solvent concentration decreases by 1.8%, and the bed thickness increases by 2.5%. The system converts this data into feed fluctuations, solvent concentration fluctuations, and bed thickness fluctuations, and combines this with a temperature-solvent coupling control strategy to predict the component change trends for the next control cycle. The prediction results are as follows: in the feed section, the oil concentration increases to 24.3%, the solvent ratio increases to 60.5%, and the colloid content increases to 1.02%; in the intermediate leaching section, the oil concentration increases to 25.8%, the solvent ratio increases to 61.7%, and the colloid content increases to 1.15%; in the discharge section, the oil concentration increases to 28.1%, the solvent ratio increases to 64.0%, and the colloid content increases to 1.23%.

[0063] Based on the prediction results, the control command allocation unit determines the feeding section as the main adjustment section, allocating 50% of the temperature adjustment to the feeding section heating jacket, 30% to the intermediate leaching section heating jacket, and 20% to the discharging section heating jacket. Simultaneously, it allocates 55% of the solvent spray adjustment to the feeding section spray nozzle, 30% to the intermediate leaching section spray nozzle, and 15% to the discharging section spray nozzle. This generates the control commands for the current control cycle: increase the target temperature of the feeding section heating jacket by 1.0℃, the intermediate leaching section by 0.6℃, and the discharging section by 0.4℃; increase the spray flow rate of the feeding section by 3.6%, the intermediate leaching section by 2.0%, and the discharging section by 0.9%. The execution sequence is set to first adjust the solvent spray flow rate of each section, then adjust the heating jacket temperature of each section after a 20-second delay.

[0064] After the control command is executed, the system re-acquires the dielectric fingerprint spectrum at the next sampling moment and analyzes it to obtain the component data for the next moment: the oil concentration in the feed section is 24.1%, the solvent ratio is 60.2%, and the colloid content is 1.01%; the oil concentration in the intermediate leaching section is 25.6%, the solvent ratio is 61.5%, and the colloid content is 1.13%; and the oil concentration in the discharge section is 28.0%, the solvent ratio is 63.8%, and the colloid content is 1.22%. The feedback correction module compares the actual component analysis results with the prediction results of the previous cycle and finds that the prediction deviations are all within the set allowable range. In the next control cycle, only the temperature adjustment and solvent spray adjustment are slightly corrected, with the temperature adjustment being corrected from +2.0℃ to +1.2℃ and the solvent spray adjustment being corrected from +6.5% to +3.5%.

[0065] During an 8-hour continuous operation verification process, after adopting this system, the duration for which the oil concentration was below the lower limit of the stability range decreased from 42 minutes under the original fixed control method to 13 minutes, the duration for which the solvent ratio was below the lower limit of the stability range decreased from 36 minutes to 11 minutes, and the duration for which the colloid content exceeded the upper limit of the stability range decreased from 28 minutes to 9 minutes. The unit energy consumption decreased from 78.5 kWh / t to 73.2 kWh / t, and the solvent recovery rate increased from 94.1% to 96.0%. These results demonstrate that this system can identify the state of the mixed solution components in real time through dielectric fingerprinting and can adjust the temperature and solvent spraying in conjunction with component changes, thus maintaining stability in the leaching process under disturbed conditions.

[0066] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0067] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent control system for mixed oil extraction production based on artificial intelligence, characterized in that, It includes a dielectric fingerprint acquisition module, a component intelligent analysis module, a coupled control decision module, a rolling predictive control module, and a feedback correction module; wherein: Dielectric fingerprint acquisition module: used to acquire the real-time dielectric spectrum of the oil-solvent mixture in the leachator, and preprocess the real-time dielectric spectrum to generate a dielectric fingerprint spectrum reflecting the state of the mixture components. Component intelligent analysis module: used to receive the dielectric fingerprint spectrum and analyze the corresponding oil concentration, solvent ratio and colloidal content according to the dielectric response characteristics of different frequency bands in the dielectric fingerprint spectrum; Coupled regulation decision module: used to determine the current leaching state based on the oil concentration, solvent ratio and colloidal content, and combined with the target constraints of oil yield, solvent recovery rate and unit energy consumption, calculate the corresponding temperature adjustment amount and solvent spraying adjustment amount, and generate temperature-solvent coupled regulation strategy; Rolling Predictive Control Module: This module is used to match the temperature-solvent coupling control strategy with the current production conditions, including fluctuations in feed, solvent concentration, and material layer thickness. It predicts the changing trends of oil concentration, solvent ratio, and colloid content in the next control cycle and outputs control commands to the heating jackets and solvent spray nozzles of each section of the leachator based on the prediction results. Feedback correction module: Used to obtain the dielectric fingerprint spectrum and corresponding component analysis results at the next moment after the control command is executed, and to correct the temperature adjustment, solvent spraying adjustment and control command for the next control cycle based on the component change deviation and prediction deviation before and after execution.

2. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 1, characterized in that, The dielectric fingerprint acquisition module includes a spectrum acquisition unit, a detection point association unit, and a sampling synchronization unit; wherein: Spectrum acquisition unit: used to acquire the complex dielectric constant and dielectric loss of the oil-solvent mixture in the frequency range of 1kHz-10MHz; Detection point association unit: used to designate the feed section, intermediate leaching section and discharge section of the leaching unit as dielectric detection points, and to add segment identifiers to the real-time dielectric spectrum collected by each dielectric detection point; Sampling synchronization unit: used to synchronously acquire the real-time dielectric spectrum of each dielectric detection point according to a sampling period of 5s-30s, forming real-time dielectric spectrum data with sampling time and segment identifier.

3. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 2, characterized in that, The dielectric fingerprint acquisition module further includes a spectrum preprocessing unit and a spectrum generation unit; wherein: The spectrum preprocessing unit is used to perform temperature compensation, outlier removal, baseline correction, and normalization on real-time dielectric spectrum data to obtain standardized dielectric spectrum data. The spectrum generation unit is used to arrange the complex dielectric constant, dielectric loss, and rate of change of adjacent frequency points in the standardized dielectric spectrum data in ascending order of frequency, and generate a dielectric fingerprint spectrum associated with the corresponding dielectric detection point.

4. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 1, characterized in that, The component intelligent analysis module includes a frequency band feature extraction unit and a component mapping analysis unit; wherein: Frequency band feature extraction unit: used to extract low-frequency polarization response features, mid-frequency dielectric loss features and high-frequency attenuation response features from dielectric fingerprint spectrum; Component mapping and analysis unit: used to determine the colloid content based on low-frequency polarization response characteristics, determine the solvent ratio based on mid-frequency dielectric loss characteristics, determine the oil concentration based on high-frequency attenuation response characteristics, and output the oil concentration, solvent ratio, and colloid content associated with the corresponding dielectric detection point.

5. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 1, characterized in that, The coupling control decision module includes a leaching state determination unit; wherein: The leaching state determination unit is used to compare the oil concentration with a preset oil concentration range, the solvent ratio with a preset solvent ratio range, and the colloid content with a preset colloid content range, and determine the current leaching state based on the comparison results; the current leaching state includes insufficient oil release state, solvent supply deviation state, high colloid accumulation state, and stable leaching state.

6. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 5, characterized in that, The coupled control decision module further includes a target constraint calculation unit and an adjustment quantity generation unit; wherein: Target constraint calculation unit: used to call the target constraint weights corresponding to oil yield, solvent recovery rate and unit energy consumption according to the current leaching state, and form the target constraint combination for the current control cycle; Adjustment amount generation unit: used to determine the temperature adjustment direction and solvent spraying adjustment direction according to the target constraint combination, and calculate the corresponding temperature adjustment amount and solvent spraying adjustment amount.

7. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 6, characterized in that, The coupled control decision module further includes a coupled verification unit; wherein: Coupled verification unit: used to perform linkage verification of temperature regulation and solvent spray regulation. When the temperature regulation is directed to increase the temperature and the solvent ratio is lower than the preset lower limit of the solvent ratio, the solvent spray regulation is increased; when the solvent spray regulation is directed to increase the temperature and the temperature of the oil-solvent mixture is lower than the preset lower limit of the leaching temperature, the temperature regulation is increased; when the colloid content is higher than the preset upper limit of the colloid content, the temperature regulation is decreased and the increase of the solvent spray regulation is limited, thereby generating a temperature-solvent coupled control strategy.

8. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 1, characterized in that, The rolling prediction control module includes a disturbance quantity construction unit and a component trend prediction unit; wherein: Disturbance quantity construction unit: used to obtain the feed flow rate, feed oil content, solvent concentration and material layer thickness in the current production condition, and calculate the difference with the corresponding data of the previous control cycle to obtain the feed fluctuation, solvent concentration fluctuation and material layer thickness fluctuation. Component trend prediction unit: It is used to take the temperature-solvent coupling control strategy, feed fluctuation, solvent concentration fluctuation and material layer thickness fluctuation as prediction inputs, and output the predicted change trend of oil concentration, solvent ratio and colloid content in the next control cycle.

9. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 8, characterized in that, The rolling prediction control module further includes a control command allocation unit; wherein: Control command allocation unit: Based on the predicted trends of oil concentration, solvent ratio and colloid content, it allocates the temperature adjustment to the heating jacket of each section of the leachator, allocates the solvent spray adjustment to the corresponding solvent spray port, and generates control commands containing target heating power, target spray flow rate and execution sequence.

10. The intelligent control system for mixed oil extraction production based on artificial intelligence according to claim 1, characterized in that, The feedback correction module includes a deviation calculation unit and a control correction unit; wherein: Deviation calculation unit: used to obtain the dielectric fingerprint spectrum and corresponding component analysis results at the next moment after the control command is executed, and to calculate the deviation of oil concentration change, solvent ratio change and colloidal content change before and after execution. At the same time, it calculates the prediction deviation between the predicted component state and the actual component analysis results in the next control cycle. Control correction unit: used to correct the temperature adjustment amount, solvent spraying adjustment amount, and control commands for the next control cycle based on the deviations in oil concentration, solvent ratio, colloid content, and prediction.

Images