A method and system for intelligent regulation of glycine magnesium synthesis process

By decoupling the electrochemical response components through intelligent control methods and dynamically adjusting the stirring frequency and feeding valve, the problem of insufficient adaptability of stirring methods in the traditional magnesium glycin synthesis process is solved, thereby improving the accuracy of reaction endpoint determination and production efficiency.

CN123113784APending Publication Date: 2026-11-17HEBEI RIELS BIOTECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611585687.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-14
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the traditional synthesis of magnesium glycine, the fixed stirring method is difficult to adapt to changes in material viscosity, resulting in localized mass transfer limitations and inaccurate detection of mechanical fluctuations, which affects the determination of the reaction endpoint and reduces production efficiency.

Method used

By acquiring the impedance phase angle and instantaneous active power, filtering and dual-channel synchronous detection are performed to decouple the electrochemical response components. Combined with the macroscopic reaction dynamic index and delay difference component, the stirring frequency and feeding valve are dynamically adjusted to achieve intelligent control.

Benefits of technology

It improves the accuracy of reaction system state identification, avoids the accumulation of unreacted raw materials and side reactions, shortens the production cycle, and increases product yield and production efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113784A_ABST
    Figure CN123113784A_ABST
Patent Text Reader

Abstract

This application relates to the field of glycine magnesium production process control technology, and discloses an intelligent control method and system for the synthesis process of glycine magnesium. The method includes: acquiring the impedance phase angle and active power to extract baseline components; generating a disturbance frequency based on the active power baseline component, and performing dual-channel synchronous detection on the signal within an adaptive integration window; calculating the projection coefficient of the mechanically correlated component, and obtaining the electrochemical response component and intensity through residual decoupling; calculating the macroscopic reaction dynamic index and the response trajectory direction consistency coefficient, and simultaneously updating the basic stirring frequency and the feeding valve opening by combining the delay difference component of the decoupled response intensity; constructing a dynamic response activity index using the decoupling parameters, and determining the reaction endpoint by combining the macroscopic dynamic index and the delay difference component. This invention, through signal decoupling and multi-variable collaborative control, eliminates the fluctuation interference caused by mechanical stirring, improves local mass transfer conditions, and enhances the accuracy of endpoint determination.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of glycine magnesium production process control technology, specifically to an intelligent control method and system for the synthesis process of glycine magnesium. Background Technology

[0002] Magnesium glycinate, as an amino acid chelate, has applications in the food, pharmaceutical, and feed industries. Traditional magnesium glycinate synthesis processes typically employ a constant feed rate and fixed stirring speed control strategy. As the synthesis reaction progresses, especially in the later stages, the viscosity of the reaction system gradually increases. A fixed stirring speed is insufficient to adapt to these viscosity changes and cannot effectively eliminate localized mass transfer resistance within the reactor. This leads to the accumulation of unreacted raw materials in localized areas, resulting in supersaturation. This not only limits product yield but also increases the probability of side reactions.

[0003] Meanwhile, during state monitoring of the synthesis process, mechanical stirring fluctuations caused by the increase in fluid viscosity introduce significant interference to detection signals, such as impedance signals. This interference component related to mechanical action masks the characteristic evolution of the pure electrochemical response in the reaction system, making it difficult for the control system to accurately extract state parameters reflecting the reaction progress. Due to the inability to obtain accurate reaction state feedback, existing systems struggle to reliably determine the reaction endpoint, easily leading to premature reaction shutdown resulting in insufficient conversion or meaningless delays causing energy waste, thus reducing overall production efficiency. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent control method and system for the synthesis of magnesium glycinate, which solves the problems of limited mass transfer due to the difficulty of adapting to changes in material viscosity caused by fixed stirring in traditional synthesis processes, and inaccurate determination of the reaction endpoint caused by mechanical fluctuation interference in state detection.

[0005] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of this invention provides a method for intelligent regulation of the glycine magnesium synthesis process, comprising the following steps: The impedance phase angle and instantaneous active power are acquired and filtered to obtain the baseline components of the impedance phase angle and active power; the disturbance frequency is determined based on the active power baseline components to generate a stirring frequency command, and the transient fluctuation components of active power are extracted. Within the adaptive integration window determined based on the disturbance frequency, the impedance phase angle and the transient fluctuation component of the active power are subjected to dual-channel synchronous detection. The projection coefficient of the mechanically related component is calculated to subtract the response component with the same frequency as the power, and the decoupled electrochemical response component and the decoupled response intensity are obtained. The macroscopic reaction dynamic index is calculated using the baseline components of the impedance phase angle and the active power; a decoupled response vector is constructed based on the decoupled electrochemical response components to form a response trajectory, and the response trajectory direction consistency coefficient is calculated. The decoupled response intensity is processed by delay difference to obtain delay difference component. The delay difference component is used to update the basic stirring frequency. The macroscopic reaction dynamic index, the response trajectory direction consistency coefficient and the delay difference component are combined to generate the opening adjustment command. Based on the decoupling response intensity and the phase after phase expansion, the dynamic response activity index is calculated; the dynamic response activity index, the macroscopic reaction dynamic index, and the delay difference component are jointly judged, and a valve closing command is output when the reaction endpoint determination condition is met.

[0006] By injecting a perturbation frequency that adaptively adjusts with the load into the stirring system and employing dual-channel detection and projection calculations, individual electrochemical response characteristics are separated from the impedance signal containing mechanical fluctuation interference. Based on this, by extracting the consistency characteristics between the macroscopic baseline and the microscopic trajectory direction, the stirring speed and the opening of the feeding valve are dynamically coordinated to improve local mass transfer conditions in the later stages of the reaction. Furthermore, the timing of feeding stoppage is controlled based on multi-parameter joint judgment conditions, thereby increasing product yield and shortening the production cycle.

[0007] Furthermore, the step of determining the disturbance frequency based on the active power baseline component to generate a stirring frequency command, and extracting the active power transient fluctuation component, specifically includes: After normalizing the active power baseline component to rated power, it is multiplied by a set amplitude proportional gain coefficient and a frequency proportional coefficient to obtain the stirring frequency disturbance amplitude and the disturbance frequency within the effective operating range. The angular frequencies corresponding to the disturbance frequencies are discretely accumulated within a discrete sampling period to generate a continuous disturbance phase. A sinusoidal reference signal is generated using the disturbance phase. The sinusoidal reference signal is multiplied by the stirring frequency disturbance amplitude and then superimposed onto the current base stirring frequency to obtain the stirring frequency command. The difference between the instantaneous active power and the active power baseline component is used to obtain the active power transient fluctuation component.

[0008] By using the normalized active power baseline component as the basis for adjustment, the system can adjust the amplitude and frequency of the detected perturbation according to the slow changing trend of the operating load and maintain the continuity of phase evolution, providing a computational benchmark for subsequent extraction of co-frequency features.

[0009] Furthermore, the dual-channel synchronous detection specifically includes: The time length between the current moment and the position in historical data where the disturbance phase difference reaches the number of complete cycles is used as the adaptive integration window; a sine reference signal and a cosine reference signal are generated based on the disturbance phase; the impedance phase angle is multiplied by the sine reference signal and the cosine reference signal respectively and integrated within the adaptive integration window to obtain the impedance phase in-phase response component and the impedance phase quadrature response component; the active power transient fluctuation component is multiplied by the sine reference signal and integrated within the adaptive integration window to obtain the power frequency response component.

[0010] Because the adaptive integral window is aligned with the complete disturbance period, it reduces the spectral leakage error caused by frequency changes, and facilitates orthogonal decomposition of impedance and power signals on the same time scale.

[0011] Furthermore, the calculation of the mechanically correlated component projection coefficients to subtract the power-frequency response component, thereby obtaining the decoupled electrochemical response component and the decoupled response intensity, specifically includes: The window linear projection relationships of the impedance phase in-phase response component and the impedance phase quadrature response component with respect to the power frequency response component are calculated respectively to obtain the corresponding mechanically correlated component projection coefficients. Using the mechanically correlated component projection coefficients, the parts linearly correlated with the power frequency response component are subtracted from the impedance phase in-phase response component and the impedance phase quadrature response component respectively to obtain the decoupled electrochemical response components as real and imaginary parts. The root of the sum of squares of the real and imaginary parts is taken to obtain the decoupled response intensity.

[0012] By utilizing the linear projection relationship within the calculation window, the system determines the interference weight of mechanical stirring on impedance measurement. By subtracting the part that is linearly related to the power frequency response, the influence of mechanical disturbance on impedance measurement is eliminated, and decoupling parameters that reflect the electrochemical reaction state independently are obtained.

[0013] Furthermore, the calculation of the macroscopic response dynamic index using the impedance phase angle baseline component and the active power baseline component specifically includes: Calculate the time first derivatives of the impedance phase angle baseline component and the active power baseline component at the current sampling time; divide the time first derivatives of the impedance phase angle baseline component and the active power baseline component by their respective sliding standard deviations within the adaptive integration window for scaling; add the two scale-standardized values ​​to obtain the macroscopic response dynamic index characterizing the combination result of the relative changes in the baselines.

[0014] By using the sliding standard deviation to scale the baseline derivatives of different dimensions, the numerical dimension difference between impedance and power signals is eliminated, enabling the synthesized macroscopic reaction dynamic index to uniformly reflect the state evolution rate of the reaction system at the macroscopic level.

[0015] Furthermore, the construction of the decoupled response vector to form the response trajectory and the calculation of the response trajectory direction consistency coefficient specifically include: Extract the decoupled response vector sequence corresponding to consecutive sampling times within the current adaptive integration window; calculate the magnitude difference between the first and last decoupled response vectors within the adaptive integration window as the overall displacement of the response trajectory, and calculate the cumulative magnitude difference between the decoupled response vectors at adjacent sampling times as the cumulative path length; remove the overall displacement from the cumulative path length to obtain the response trajectory direction consistency coefficient, which characterizes the degree of change of the pure electrochemical response along a consistent direction.

[0016] By calculating the ratio of the total displacement of the response trajectory to the cumulative path length, the orientation of the decoupled response vector's change in the complex plane can be quantified, making it easier to distinguish between disordered fluctuations and effective response processes in the system.

[0017] Furthermore, the step of performing delay differential processing on the decoupling response intensity to obtain a delay difference component, and using the delay difference component to update the basic stirring frequency, specifically includes: The historical decoupling response intensity, which is separated from the current time by one adaptive integral window, is read through a circular data buffer queue. The difference between the current decoupling response intensity and the historical decoupling response intensity is used to obtain the delay difference component, which characterizes the degree of net change in the response. When the delay difference component is negative or positive, the discrete integral accumulation update in the opposite direction is performed on the basic stirring frequency in combination with the integral speed regulation gain, and the integral accumulation in the corresponding direction is stopped when the operating frequency limit boundary is reached.

[0018] By updating the basic stirring frequency using the delay difference component, the system adjusts the stirring force in real time according to the evolution direction of the decoupled response, thereby improving the local mass transfer deterioration caused by viscosity increase in the later stage of the reaction. At the same time, the frequency limiting rule is combined to prevent the integral accumulation from exceeding the safe operating range.

[0019] Furthermore, the step of generating the opening adjustment command by combining the macroscopic response dynamic index, the response trajectory direction consistency coefficient, and the delay difference component specifically includes: Extract the positive change portion and the absolute negative change portion of the delay difference component; using a preset base maintenance opening degree as the base term, multiply the macroscopic response dynamic index by a proportional coefficient as the positive driving term; multiply the positive change portion by the response trajectory direction consistency coefficient and the adjustment gain corresponding to the positive delay difference component as the first negative suppression term, and multiply the absolute negative change portion by the adjustment gain corresponding to the negative delay difference component as the second negative suppression term; add the base term and the positive driving term, and subtract the first negative suppression term and the second negative suppression term to generate the opening degree adjustment command.

[0020] In the generation logic of the opening adjustment command, when the delay difference component changes positively or negatively, a negative suppression effect is introduced through the consistency coefficient of the response trajectory direction and the adjustment gain, so as to achieve a balance between controlling the local concentration from overload and maintaining the feeding efficiency.

[0021] Furthermore, the step of calculating the dynamic response activity index based on the decoupling response intensity and the phase after phase expansion processing; and jointly judging the dynamic response activity index, the macroscopic response dynamic index, and the delay difference component, specifically includes: Within the adaptive integration window, the product of the square of the decoupled response intensity and the time change rate of the phase is integrated to obtain the dynamic response activity index; the time change rate of the dynamic response activity index and the time change rate of the macroscopic response dynamic index are obtained; when the absolute values ​​of the dynamic response activity index and its time change rate, the macroscopic response dynamic index and its time change rate, and the delay difference component are all synchronously less than or equal to the threshold corresponding to the endpoint constraint, and the duration meets the set number of disturbance cycles, the endpoint determination condition is confirmed to be met.

[0022] By combining the decoupled response intensity with the phase change rate to form a dynamic response activity index, and establishing joint judgment conditions together with the macroscopic response dynamic index, the delay difference component and its change rate, the interference of local fluctuations on the judgment results is reduced, and premature termination of the response or excessive waste is avoided.

[0023] A second aspect of this invention provides an intelligent control system for the synthesis of magnesium glycine, applied to the intelligent control method for the synthesis of magnesium glycine described in the first aspect of this invention, comprising: The reaction module includes a reaction vessel equipped with a stirring mechanism, a jacket, and a feed pipeline. The reaction vessel is used to contain the reactants and provide space for the chemical reaction and mixing of the reactants. The jacket is used to introduce an external heat exchange medium to exchange heat with the reactants inside the reaction vessel. The sensing module includes an AC impedance detection unit installed inside the reactor and an active power calculation unit for acquiring the active power of the stirring motor. The active power calculation unit is used to acquire the impedance phase angle of the reactants and the instantaneous active power during the operation of the stirring motor. The active power calculation unit can be deployed in the motor drive unit inside the execution module and transmit the calculated instantaneous active power to the control module. The execution module, including a frequency converter, a stirring motor, and a feeding regulating valve, is used to adjust the operating frequency of the stirring motor and the feeding rate of the reaction raw materials according to the control commands output by the control module. The control module, including a programmable logic controller, is used to perform signal processing and related component decoupling processing on impedance phase angle and active power, and generate stirring frequency command and feeding regulating valve opening adjustment command based on the processing results. The communication module, including the industrial fieldbus, is used to connect the sensing module, the execution module, and the control module, enabling the transmission of detection data to the control module and the issuance of control commands to the execution module.

[0024] This invention provides a method and system for intelligent regulation of the glycine magnesium synthesis process. It has the following beneficial effects: 1. This invention adaptively generates stirring frequency disturbances based on the active power baseline component, and performs dual-channel synchronous detection and linear projection calculations on impedance phase angle and power transient fluctuations within the corresponding integration window. This allows the interference components linearly related to mechanical stirring to be subtracted from the mixed signal, thereby separating out the decoupled response parameters that reflect the electrochemical state of the reactants. This eliminates the influence of stirring fluctuations caused by fluid viscosity changes on signal detection and improves the accuracy of reaction system state identification.

[0025] 2. This invention constructs a multivariate joint regulation mechanism for the basic stirring frequency and the opening of the feed regulating valve by combining the macroscopic reaction dynamic index, the response trajectory direction consistency coefficient, and the delay difference component of the decoupled response intensity. Furthermore, while maintaining the basic feed drive, a negative suppression term is dynamically introduced based on the direction and consistency of signal evolution, coordinating the stirring kinetic energy and feed rate in real time. This avoids local accumulation of unreacted raw materials and side reactions caused by increased material viscosity in the later stages of the reaction, thus ensuring the yield of the synthesized product.

[0026] 3. This invention constructs a dynamic response activity index by utilizing decoupled response intensity and phase change rate, and sets it together with the macroscopic response dynamic index, delay difference component, and their corresponding time change rate as joint judgment conditions. This allows for a multi-dimensional judgment logic that integrates signal dissipation and the synchronous convergence characteristics of macroscopic and microscopic states, filtering out interference from local random fluctuations in the later stages of the response, avoiding premature termination or ineffective waste in the production process, and shortening the single-batch production cycle while ensuring product quality. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram of the baseline feature separation and perturbation instruction injection logic of the present invention; Figure 4 This is a schematic diagram illustrating the adaptive integral window mapping and dual-channel residual cancellation principle of the present invention. Figure 5 This is a schematic diagram illustrating the construction logic of the state evolution of the dual-scale system in this invention. Figure 6 This is a schematic diagram of the active identification control principle based on historical response self-difference of the present invention; Figure 7 This is a schematic diagram of the reaction endpoint determination principle based on dynamic energy dissipation of the present invention; Figure 8 This is a comparison diagram of the evolution of the decoupled response vector complex plane trajectory of the present invention; Figure 9 This is a comparative chart showing the evolution of yield and by-product content across multiple batches of the product according to the present invention. Detailed Implementation

[0028] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only 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.

[0029] Reference Figure 1 This invention provides an intelligent control system for the synthesis of magnesium glycine, the system comprising: The reaction module includes a reactor equipped with a stirring mechanism, a jacket, and a feed pipeline. The reactor is used to contain the reactants and provide space for the chemical reaction and mixing of the reactants. The jacket is used to introduce an external heat exchange medium to exchange heat with the reactants inside the reactor.

[0030] The sensing module includes an AC impedance detection unit installed inside the reactor and an active power calculation unit for acquiring the active power of the stirring motor, which is used to acquire the impedance phase angle of the reactants and the instantaneous active power of the stirring motor during operation.

[0031] The active power calculation unit can be deployed in the motor drive unit inside the execution module and transmit the calculated instantaneous active power to the control module.

[0032] The execution module, which includes a frequency converter, a stirring motor, and a feeding regulating valve, is used to adjust the operating frequency of the stirring motor and the feeding rate of the reaction raw materials according to the control commands output by the control module.

[0033] The control module includes a programmable logic controller (PLC) for signal processing of impedance phase angle and active power, as well as decoupling of related components. Based on the processing results, it generates stirring frequency commands and feeding control valve opening adjustment commands.

[0034] The communication module, which includes an industrial fieldbus, is used to connect the sensing module, the execution module, and the control module, enabling the transmission of detection data to the control module and the issuance of control commands to the execution module.

[0035] Reference Figure 2 This invention provides a method for intelligent regulation of the glycine magnesium synthesis process, the method comprising the following steps: Step S10: The control module synchronously acquires the impedance phase angle and instantaneous active power through the communication module, and performs filtering processing on each to obtain the baseline component of the impedance phase angle and the baseline component of the active power.

[0036] The control module calculates the amplitude and frequency of the stirring frequency disturbance based on the active power baseline component, and superimposes the corresponding frequency disturbance onto the base stirring frequency to generate a stirring frequency command and send it to the execution module; at the same time, it obtains the transient fluctuation component of active power based on the instantaneous active power and the active power baseline component.

[0037] Step S20: The control module determines the adaptive integration window based on the disturbance frequency, and performs dual-channel synchronous detection on the impedance phase angle and the transient fluctuation components of active power within the adaptive integration window.

[0038] The control module calculates the projection coefficients of the mechanically correlated components based on the linear projection relationship within the window, and uses these projection coefficients to subtract the power-frequency response component from the impedance phase response component, thereby obtaining the decoupled electrochemical response component and the decoupled response intensity.

[0039] Step S30: The control module uses the baseline components of impedance phase angle, active power, and their corresponding sliding standard deviations to calculate the macroscopic response dynamic index.

[0040] The control module further constructs a response trajectory in the complex plane based on the decoupled electrochemical response components, and calculates the response trajectory direction consistency coefficient based on the response trajectory.

[0041] Step S40: The control module performs delay differential processing on the decoupling response intensity quantity based on an adaptive integral window to obtain the delay differential component.

[0042] The control module uses the delay difference component to integrally update the basic mixing frequency and adjusts the speed of the mixing motor according to the updated basic mixing frequency. At the same time, it calculates the opening adjustment command of the feeding regulating valve by combining the macroscopic reaction dynamic index, the response trajectory direction consistency coefficient and the delay difference component, and outputs the opening adjustment command to the execution module.

[0043] Step S50: The control module performs integral calculations based on the decoupling response intensity and phase change rate to obtain the dynamic response activity index.

[0044] The control module jointly judges the dynamic response activity index and its rate of change, the macroscopic reaction dynamic index and its rate of change, and the delay difference component; when the above parameters meet the reaction endpoint determination conditions and reach the corresponding duration conditions, the control module outputs a valve closing command to close the feeding regulating valve and terminate the feeding process of the reaction raw materials.

[0045] To further clarify the implementation of each technical aspect of the present invention, the following will provide a detailed description of the implementation of each functional module involved above and its internal processing flow.

[0046] Reference Figure 3 Regarding step S10 in the above method, this embodiment further describes the process of acquiring impedance phase angle and instantaneous active power, extracting baseline components, calculating disturbance parameters, generating stirring frequency commands, and extracting transient fluctuation components of active power, specifically including steps S101 to S105.

[0047] In step S101, the control module operates according to the set discrete sampling period and synchronously reads the impedance phase angle output by the AC impedance detection unit in the sensing module through the communication module. and the instantaneous active power output by the active power calculation unit. .

[0048] After the system starts, the control module continuously acquires the impedance phase angle according to the discrete sampling period. and instantaneous active power The collected data is then written into a circular buffer unit.

[0049] When the length of historical data in the cyclic buffer unit meets the data retrieval requirements of the subsequent adaptive integral window, the control module begins to execute subsequent calculations; when the length of historical data does not yet meet the data retrieval requirements, the initial basic stirring frequency and initial feeding state are maintained.

[0050] The AC impedance detection unit continuously detects the reactants at a fixed AC excitation frequency to obtain the real and imaginary parts of the impedance at the corresponding AC excitation frequency, and then obtains the impedance phase angle accordingly. .

[0051] The discrete sampling period of the control module can be determined based on the processor's computing power and the communication module's data transmission capability; in this embodiment, it is set to 5 to 20 milliseconds.

[0052] The active power calculation unit obtains the instantaneous active power based on the stator voltage and stator current during the operation of the stirring motor. The active power calculation process can be implemented by the motor drive unit using the existing motor active power calculation method.

[0053] Step S102, the control module controls the synchronously acquired impedance phase angle and instantaneous active power Digital low-pass filtering is performed separately to obtain the corresponding low-frequency baseline components.

[0054] The cutoff frequency of the digital low-pass filter is lower than the lower limit of the effective operating range of the subsequent perturbation frequency. It maintains a distance from the frequency band where the disturbance frequency is located. In this embodiment, the cutoff frequency of the digital low-pass filter is set to 0.02 Hz to 0.05 Hz to retain the components in the impedance phase angle and active power that change slowly, and to suppress the disturbance frequency band and higher frequency fluctuation components.

[0055] After filtering, the control module obtains the baseline component of the impedance phase angle. and active power baseline component .

[0056] Among them, the active power baseline component Used to characterize the slow changing trend of the load on the stirring motor; impedance phase angle baseline component Used to characterize the slow change trend of the impedance phase angle of reactants.

[0057] Step S103, the control module bases the active power baseline component Calculate the amplitude of the stirring frequency disturbance and disturbance frequency .

[0058] During the reaction, as the flow state of the reactants changes, the operating load of the stirring motor changes accordingly. Active power baseline component. To reflect changes in the operating load, the control module adjusts the disturbance parameters applied to the base mixing frequency accordingly.

[0059] The control module determines the stirring frequency disturbance amplitude according to the following calculation relationship. and disturbance frequency : The control module first analyzes the active power baseline component. Normalize the rated power to obtain the normalized power state variables. The calculation method is as follows: ;in, This refers to the rated operating power of the mixing motor.

[0060] Subsequently, the control module bases its decisions on the normalized power state variables. Calculate the amplitude of the stirring frequency disturbance and disturbance frequency : ; ; In the formula, This is the amplitude proportional gain coefficient, whose value can be calibrated based on the rated operating parameters and allowable operating load of the stirring motor. In this embodiment, the value is taken as 0.01 to 0.15. This is the frequency scaling factor, which is set to a value of 5 to 20 in this embodiment.

[0061] Normalized power state quantity When the value increases, according to the above calculation relationship, the amplitude of the stirring frequency disturbance... and disturbance frequency It increases accordingly.

[0062] The control module calculates the disturbance frequency. Amplitude limiting is applied to keep it within the effective operating range of 0.1 Hz to 1.5 Hz. The stirring motor periodically changes its operating frequency at this disturbance frequency, thereby causing the stirring effect on the reactants to change periodically.

[0063] Step S104, the control module determines the frequency of the disturbance. Generate a disturbance phase corresponding to the disturbance frequency. It also uses the perturbation phase to generate a stirring frequency command.

[0064] Considering that the control module operates using discrete sampling, the control module measures the disturbance frequency in each sampling period. The corresponding angular frequencies are discretely accumulated and combined with the phase value of the previous sampling period to update the current perturbation phase. .

[0065] The resulting perturbation phase With the current disturbance frequency Correspondingly, a sinusoidal reference signal is used to generate a signal that maintains phase continuity with the stirring frequency perturbation.

[0066] The control module will be based on the disturbance phase The generated sinusoidal disturbance is superimposed on the current base mixing frequency. Receive stirring frequency command .

[0067] Basic mixing frequency The stirring frequency is maintained at its initial value during the initial stage of the reaction and updated based on the delay difference during subsequent closed-loop adjustment. The synthesis formula is: ; The control module transmits stirring frequency commands through the communication module. The data is sent to the frequency converter in the execution module.

[0068] During the command output process, the control module issues commands regarding the stirring frequency. A frequency limiting process is performed to keep it within the allowable operating frequency range of the stirring motor. The frequency converter operates according to the received stirring frequency command. Drive the mixing motor to operate, ensuring that the motor's operating frequency revolves around the base mixing frequency. Periodic changes.

[0069] Step S105, the control module outputs a stirring frequency command. At the same time, based on the instantaneous active power currently output by the active power calculation unit and the active power baseline component obtained in step S102 Calculate the transient fluctuation components of active power. .

[0070] Specifically, the calculation formula is as follows: ; Active power transient fluctuation components This represents the change in instantaneous active power relative to the active power baseline component, thereby reducing the impact of slow changes in active power on subsequent disturbance response analysis.

[0071] The control module will control the transient fluctuation components of active power. It is stored in the internal storage unit and is called in subsequent steps to calculate the mechanical response related components and perform related component subtraction.

[0072] Reference Figure 4 Regarding step S20 in the above method, this embodiment further explains the determination of the adaptive integration window, the synchronous detection of impedance phase response and active power response, the calculation of the projection coefficient of mechanical correlation component, and the correlation component subtraction process, specifically including steps S201 to S204.

[0073] Step S201, the control module determines the disturbance frequency obtained in step S10. and the perturbation phase corresponding to the perturbation frequency Determine the adaptive integration window at the current moment. .

[0074] Specifically, the control module uses a perturbation phase As the basis for identifying the perturbation period, the system searches backward along the historical data cache from the data position corresponding to the current moment to determine the position where the phase difference reaches the required number of complete periods. The time corresponding to this position is then used as the starting time of the current integration window. The time length between the current moment and the starting time is the adaptive integration window. .

[0075] Therefore, adaptive integral window With disturbance frequency The adjustment is made according to the changes and covers the entire disturbance period to reduce the impact of the inconsistency between the integration window and the current disturbance period on subsequent synchronous detection results.

[0076] In discrete data processing, the control module uses an adaptive integral window. The correspondence between the sampling period and the sampling period determines the number of sampling points contained in the current integration window, and the samples required for subsequent calculations are extracted from the historical data cache accordingly.

[0077] In step S202, the control module determines the adaptive integral window based on the one determined in step S201. Retrieve impedance phase angle from historical data cache and active power transient fluctuation components And using perturbation phase A reference signal is generated, and the above signal is synchronously detected.

[0078] The control module based on the disturbance phase Sine and cosine reference signals are generated separately. The impedance phase angle is multiplied by the sine and cosine reference signals respectively, and then processed within an adaptive integration window. Integrating within the inner loop yields the impedance phase in-phase response component. and impedance phase orthogonal response components .

[0079] At the same time, the control module will control the transient fluctuation component of active power. Multiplied by a sinusoidal reference signal and within an adaptive integration window Integrating within the range yields the power response component at the same frequency. .

[0080] The calculation formula is as follows: ; ; ; In the formula, The impedance phase in-phase response component, These are the impedance phase quadrature response components. The power-frequency response component, This is the time integration variable.

[0081] When the control module performs the above integral operation using discrete sampling, it reads the adaptive integral window from the historical data cache. For each sampling point, the signal value of each sampling point is multiplied by the corresponding sine or cosine reference value, the resulting product terms are accumulated, and the corresponding integral result is obtained by combining the sampling period.

[0082] Step S203, the control module determines the impedance phase in-phase response component. Impedance phase quadrature response components and power frequency response components Calculate the projection coefficients of the mechanically related components. and .

[0083] Among them, the power frequency response component This is used to characterize the same-frequency response of the stirring load to an applied frequency disturbance. The control module calculates respectively. , Compared to The window linear projection relationship is used to determine the components in the impedance phase response that are related to the power frequency response.

[0084] The formula for calculating the projection coefficients of the mechanically related components is: ; ; In the formula, and Adaptive integral window The inner correspondence and The projection coefficients of the mechanically related components; It is a very small positive constant, used for power-frequency response components. To keep the denominator stable when it is small, we take 10 in this embodiment. -8 Up to 10 -6 .

[0085] In step S204, the control module uses the projection coefficients of the mechanical correlation components obtained in step S203. and From the impedance phase in-phase response components respectively and impedance phase orthogonal response components Subtracting the power-frequency response component The linearly correlated components yield the decoupled electrochemical response components. and .

[0086] The calculation formula is: ; ; In the formula, and These are the decoupled electrochemical response components after deducting the power-frequency response correlation components.

[0087] The control module further determines, and Calculate the decoupling response intensity : ; Calculated decoupled electrochemical response components , and decoupling response intensity It is stored in the internal storage area of ​​the control module for subsequent calculation of reaction state parameters, adjustment of basic stirring frequency, and control of the feeding regulating valve.

[0088] Reference Figure 5 Regarding step S30 in the above method, this embodiment further explains the calculation process of the macroscopic response dynamic index, the decoupled response vector, and the response trajectory direction consistency coefficient, specifically including steps S301 to S303.

[0089] Step S301: The control module retrieves the impedance phase angle baseline component obtained in step S10. and active power baseline component The macroscopic response dynamic index is calculated based on the changes in the two baseline components and their sliding standard deviations. .

[0090] Impedance phase angle baseline component With active power baseline component They have different numerical ranges. To reduce the impact of numerical scale differences on the combined calculation results, the control module uses the corresponding moving standard deviation to scale the first derivatives of the two baseline components.

[0091] Among them, the sliding standard deviation and All are based on the current adaptive integration window Data calculation within.

[0092] The formula for calculating the macroscopic response dynamic index is: ; In the formula, It is a dynamic index of macroscopic response; and These are the first-order time derivatives of the impedance phase angle baseline component and the active power baseline component, respectively. The control module uses a discrete-time first-order backward difference method to approximate the first-order time derivatives. and They are respectively and In the current adaptive integration window The sliding standard deviation within; In this embodiment, 10 is chosen as the smallest positive constant used to keep the denominator stable. -6 Up to 10 -4 .

[0093] when or When the fluctuation within the current adaptive integration window is small, the corresponding sliding standard deviation decreases. Used to reduce the impact of a denominator approaching zero on the calculation results. Macroscopic reaction dynamic index This is used to represent the combined result of the relative changes in the baseline components of the impedance phase angle and the active power.

[0094] Step S302: The control module retrieves the decoupled electrochemical response components obtained in step S20. and ,by As the actual part, As the imaginary part, a decoupled response vector is constructed in the complex plane. .

[0095] Its mathematical expression is: ; In the formula, To decouple the response vector, The imaginary unit satisfies .

[0096] The control module uses the decoupling response vector Calculate the corresponding phase : ; In the formula, It is the argument operation function for complex numbers.

[0097] In calculating phase At this time, the control module can use four-quadrant arctangent calculation to determine the decoupling response vector. The phase corresponding to the quadrant in which it is located.

[0098] The control module compares the phase values ​​of adjacent sampling times; when the adjacent phase difference crosses... At the boundary, by adding or subtracting 2π, the phase sequence is expanded to obtain a phase that changes continuously along time. .

[0099] Step S303: The control module extracts the current adaptive integral window. Inner continuity The decoupling response vector corresponding to each sampling time. The response trajectory direction consistency coefficient is calculated based on the trajectory formed by the decoupled response vector in the complex plane. .in, Indicates the current adaptive integration window The number of corresponding sampling points.

[0100] Response trajectory direction consistency coefficient By comparing the total displacement of the response trajectory within the window with the actual path length, the degree to which the decoupled response vector changes in a consistent direction within the window is characterized.

[0101] The formula for calculating the consistency coefficient of the response trajectory direction is: ; In the formula, This is the consistency coefficient of the response trajectory direction within the current adaptive integration window; and These represent the first and second digits within the adaptive integral window, respectively. The decoupled response vectors corresponding to each sampling time and the previous sampling time; the numerator represents the total displacement of the response trajectory within the window, and the cumulative term in the denominator represents the cumulative path length of the response trajectory; In this embodiment, 10 is chosen as the smallest positive constant used to keep the denominator stable. -8 .

[0102] when When the value is close to 1, it indicates that the total displacement and cumulative path length of the response trajectory within the current window are close, and the direction consistency of the trajectory is high; when When the value decreases, it indicates that the cumulative path length increases relative to the overall displacement, and the directional consistency of the response trajectory decreases.

[0103] To reduce the impact of detection fluctuations on the consistency coefficient of the response trajectory direction when the decoupling response intensity is low, the control module also compares the current decoupling response intensity. The average level of the decoupled response intensity within the current adaptive integral window.

[0104] when When continuously in a low response state, the current calculation cycle yields... The control module does not participate in the feeding adjustment and continues to use the data corresponding to the most recent valid calculation cycle. ;when After restoring to a response state capable of trajectory determination, the result obtained in the current calculation cycle is then used. .

[0105] The control module will macroscopically react to the dynamic index. Consistency coefficient with response trajectory direction Stored in the internal storage area for subsequent basic mixing frequency adjustment and feeding control valve processes.

[0106] Reference Figure 6 Regarding step S40 in the above method, this embodiment further explains the calculation process of the delay difference of the decoupled response intensity, the integral update of the basic stirring frequency, and the opening of the feeding regulating valve, specifically including steps S401 to S403.

[0107] Step S401: The control module obtains the decoupling response intensity for the current calculation cycle. And based on the current adaptive integration window The corresponding historical decoupling response intensity is used to calculate the delay difference component. .

[0108] The control module continuously stores historical decoupled response intensity values ​​according to the sampling time sequence through a circular data buffer queue. The capacity of the circular data buffer queue covers the amount of historical data corresponding to the maximum achievable length of the adaptive integration window, where the maximum length of the adaptive integration window is related to the lower limit of the perturbation frequency in step S10. Correspondingly.

[0109] When calculating the current delay difference component, the control module reads an adaptive integral window from the circular data buffer queue, which is one time interval from the current time. Historical decoupling response intensity and the current decoupling response intensity The difference is calculated between the intensity of the decoupled response and the historical intensity.

[0110] when When the corresponding time point falls between two adjacent historical sampling times, the control module performs linear interpolation based on the two adjacent historical sampling values ​​to obtain... .

[0111] The calculation formula is: ; Delay difference component This represents the net change in the current decoupling response strength compared to the decoupling response strength one adaptive integration window prior. This is due to the difference between the used delay length and the adaptive integration window. Consistent, the differential result corresponds to the time window used in the current synchronous detection.

[0112] During the data accumulation phase after system startup, when there is no complete adaptive integral window in the circular data buffer queue... When the data is received, the control module does not perform delay differential adjustment and maintains the current basic mixing frequency and current feeding status; once the cached data meets the data retrieval requirements of the current adaptive integral window, it begins to calculate the delay differential component. .

[0113] Step S402, the control module uses the delay difference component As an input for adjusting the basic stirring frequency, the basic stirring frequency is... Update points.

[0114] To mitigate the impact of small fluctuations in the decoupling response intensity on the continuous cumulative change in the basic stirring frequency, the control module adjusts the frequency based on the current adaptive integral window. The fluctuation range determines whether to perform an integral update. When the current... When the fluctuation range is within the specified range, maintain the basic mixing frequency unchanged; when the current When the fluctuation range is exceeded, the basic mixing frequency is updated according to the following calculation relationship.

[0115] The control module uses the following discrete integral control law to calculate the new basic stirring frequency. : ; In the formula, This is the base stirring frequency for the previous control cycle; This refers to the sampling period of the control module; The integral speed control gain is set to 0.5 to 2.0 in this embodiment.

[0116] According to the above calculation relationship, when When it is negative, in When the value is positive and the basic mixing frequency has not reached the limit boundary, the current basic mixing frequency increases relative to the previous control cycle; when When the value is positive, the basic mixing frequency decreases accordingly.

[0117] Updated base mixing frequency Used for calculating the stirring frequency command in step S104, it together with the stirring frequency disturbance term forms the stirring frequency command output to the frequency converter.

[0118] The control module controls the basic mixing frequency. Limit the amplitude to keep it within the allowable operating frequency range of the stirring motor.

[0119] When the basic mixing frequency reaches the corresponding amplitude limit boundary, and the delay difference component If the integral update result continues to change in the direction of the limiting amplitude, the control module stops the integral accumulation in that direction; when the delay difference component... When the direction is changed so that the basic mixing frequency changes in a direction away from the limit boundary, the integral update is resumed.

[0120] Step S403, the control module determines the dynamic index of the macroscopic reaction. Consistency coefficient of response trajectory direction In addition to the delay difference component, the opening adjustment command of the feed regulating valve is calculated. .

[0121] Among them, the delay difference component Decomposed into positive change parts and the absolute part of negative change Consistency coefficient of response trajectory direction Used to adjust the effect of the positive change on the valve opening.

[0122] The control module uses the following calculation relationship to determine the opening adjustment command of the feeding regulating valve: ; in: ; ; In the formula, This is to output the opening adjustment command to the feed regulating valve; The valve opening is maintained at a certain level, and its value is calibrated based on the basic production load of the reactor. In this embodiment, it is 20% to 40% of the full-scale valve opening. The proportionality coefficient corresponding to the macroscopic response dynamic index is taken as 0.5 to 3.0 in this embodiment; This represents the adjustment gain corresponding to the positive delay difference component. The adjustment gain corresponding to the negative delay difference component is shown in this embodiment. and All values ​​range from 1.0 to 5.0; Represents the delay difference component The positive change portion, Represents the delay difference component The absolute part of the negative change.

[0123] In the above calculation relationship, The basic elements constituting valve opening degree; According to the macro-reaction dynamic index The value and direction of the valve opening are adjusted.

[0124] when hour, Participate in valve opening calculation. As a term that reduces the opening degree; the consistency coefficient of the response trajectory direction The larger the value, the greater its effect on reducing the valve opening, assuming other parameters remain constant.

[0125] when hour, Participate in valve opening calculation. As a term for reducing the opening, the control module updates the basic stirring frequency according to the calculation relationship in step S402, so that the stirring frequency adjustment and feeding adjustment are executed according to the current delay differential state.

[0126] The control module will calculate the results. The opening of the feed regulating valve is limited to 0% to 100%, and the limited opening regulation command is converted into a 4-20mA analog current signal and output to the feed regulating valve in the execution module.

[0127] The electric positioner of the feed regulating valve adjusts the valve core position according to a 4-20mA analog current signal, thereby changing the feed rate of the reaction raw materials.

[0128] Reference Figure 7 Regarding step S50 in the above method, this embodiment further illustrates the calculation of the dynamic response activity index and the joint determination process of the response endpoint, specifically including steps S501 and S502.

[0129] Step S501: The control module acquires the decoupling response strength. and the phase after phase expansion processing And calculate the dynamic response activity index based on the two. .

[0130] The control module utilizes decoupling response strength. It reflects the numerical magnitude of the current decoupling response and utilizes the phase. The rate of change over time reflects the degree to which the phase of the decoupled response vector changes over time.

[0131] The control module will decouple the response intensity. The square and phase The combination of time rates of change, and in the current adaptive integral window Integrating within the timeframe yields the dynamic response activity index. It is used to characterize the overall state of the decoupling response intensity and its phase change within the current window.

[0132] The calculation formula is: ; In the formula, This is a dynamic response activity indicator; This represents the intensity of the decoupling response at the integration time τ. For time integration variables; For phase Rate of change over time.

[0133] When the control module calculates the phase change rate using discrete sampling, it obtains the corresponding discrete approximation value based on the phase difference between two adjacent sampling times and the sampling period.

[0134] Before calculating the phase change rate, the control module uses the phase expansion result in step S302 to reduce the impact of discontinuous jumps caused when the phase crosses the ±π boundary on the difference result.

[0135] In the discrete implementation, the control module uses an adaptive integral window. The product terms corresponding to each sampling time are accumulated and combined with the sampling period to obtain... .

[0136] Step S502, the control module determines the dynamic response activity index. The time change rate of dynamic response activity index Macroeconomic Reaction Dynamic Index The time rate of change of the macro-reaction dynamic index and delay difference components The reaction endpoint is determined jointly.

[0137] The control module continuously calculates the above parameters and determines whether each parameter meets the corresponding endpoint constraint conditions.

[0138] The control module determines that the reaction endpoint condition has been reached when the following conditions are met simultaneously: ; ; ; ; ; In the formula, Used to limit the dynamic response activity index The value range is defined in this embodiment as the effective reaction stage of this batch. 1% to 5% of the maximum value; Used to define the macroscopic reaction dynamic index The value range is defined in this embodiment as the effective reaction stage of this batch. 1% to 5% of the maximum absolute value.

[0139] Used to limit the dynamic response activity index The degree of time variation, in this embodiment, is taken as the effective reaction stage of this batch. The absolute value of the rate of change is 0.5% to 2% of its maximum value; Used to define the macroscopic reaction dynamic index The degree of time variation, in this embodiment, is taken as the effective reaction stage of this batch. The absolute value of the rate of change is 0.5% to 2% of the maximum value.

[0140] This is used to limit the net change in the decoupled response intensity within a time range corresponding to an adaptive integral window. In this embodiment, the delay difference component of the effective response stage of this batch is taken. 1% to 3% of the maximum value.

[0141] To mitigate the impact of insufficient historical data during system startup on endpoint determination, the control module must complete the effective calculation of at least one complete adaptive integral window during this batch of operation. , or The above-mentioned joint determination of the reaction endpoint is performed only after at least one quantity in the reaction has undergone a identifiable change relative to its initial state.

[0142] When all the above endpoint determination conditions are continuously satisfied for at least one complete adaptive integral window, the control module confirms that the reaction endpoint conditions have been met. In this embodiment, the conditions can be maintained continuously for two to three complete disturbance cycles to reduce the impact of instantaneous fluctuations on the endpoint determination results.

[0143] After the control module confirms that the reaction endpoint conditions have been met, it will issue an adjustment command to the opening of the feed regulating valve. Set to 0%, close the feeding regulating valve and terminate the supply of reaction raw materials, and at the same time output a status signal indicating that the reaction has reached its endpoint through the communication module.

[0144] After feeding is terminated, the stirring motor can continue to run at the current basic stirring frequency for a holding time to keep the materials in the reactor stirred and mixed. After the holding time is over, the control module controls the stirring motor to reduce speed or stop running according to the subsequent production process.

[0145] To enable those skilled in the art to more clearly understand the purpose, technical solution, and advantages of this invention, the present invention will be further described in detail below with reference to specific application embodiments, real experimental test data, and corresponding drawings. It should be noted that the embodiments described herein are only for explaining the present invention and are not intended to limit the scope of protection of the present invention.

[0146] This embodiment applies to a 500L glycine magnesium synthesis reactor, with the stirring motor's rated operating power... =5kW. The discrete sampling period of the control module is set to... =10 ms, or 0.01 seconds. This occurs at a certain point in the middle to late stage of the reaction (high viscosity stage). The following control logic will be implemented: The current active power baseline component is extracted after low-pass filtering. kW.

[0147] Control module execution normalization: .

[0148] Set amplitude ratio gain Frequency proportionality coefficient The calculation yields: Amplitude of stirring frequency disturbance Hz; Disturbance frequency Hz.

[0149] Current basic mixing frequency If the frequency is Hz, then the stirring frequency command output to the inverter is: .

[0150] The current disturbance frequency is 1.05Hz, and the control module determines the current adaptive integral window length. . s corresponds to approximately 95 sampling points.

[0151] Within this window, the minimum normal values ​​are obtained through dual-channel synchronous detection and integral projection. The control module calculates the projection coefficients of the mechanically related components. .

[0152] The original impedance in-phase and quadrature components at the current moment are X=1.8, Y=0.6, and the power in-frequency response component is M=1.2. Perform subtraction calculation: Decoupled in-phase components ; Decoupled orthogonal components ; The decoupling response intensity was calculated. .

[0153] The current macroeconomic reaction dynamic index is calculated using standardized derivatives. .

[0154] The consistency coefficient of the response trajectory direction was calculated by comparing the displacement and path of 95 historical sampling points in the complex plane. This indicates that the current reaction is in a stable state of propulsion.

[0155] The control module extracts the historical decoupling response intensity from the circular buffer queue up to 0.952 seconds ago. Calculate the delay difference component: .

[0156] because This indicates that the system response is continuously improving, and its positive change component can be extracted. .

[0157] Basic stirring frequency integral update (assuming) ): (Hz slightly reduced to maintain stable shear force). The feed control valve command is calculated, assuming a basic maintenance opening. ; Positive driving factors: ; Negative oversaturation suppression term: ; Final output opening: .

[0158] Because the system exhibits a local supersaturation tendency, the negative inhibition term actively reduces the opening command, preventing the occurrence of side reactions.

[0159] The following verification is illustrated with specific application scenarios and comparative test data. Multiple batches of comparative experiments were conducted using a 500L synthesis reactor of the same specifications. The experimental subject was the chelation reaction of glycine and magnesium oxide powder in an aqueous solvent.

[0160] The comparison group settings are as follows: Experimental group: The intelligent control system and method for the synthesis of magnesium glycine provided in this invention were used. The system automatically calculates the opening degree of the feeding regulating valve and the stirring frequency, and automatically outputs a valve closing command based on a joint judgment criterion. Control group: A constant feeding rate (35% valve opening) and a constant rotation speed (30Hz) were maintained, combined with basic PID temperature control.

[0161] The comparative test data and effect analysis are as follows: according to Figure 8It can be seen that in the initial stage of the reaction, the material mixing process is accompanied by random fluid fluctuations, and the trajectory of the decoupled response vector in the complex plane exhibits a divergent and disordered random free state, with the corresponding response trajectory direction consistency coefficient at a low level. When the reaction enters the middle stage, under the adaptive application of perturbation commands and the action of dual-channel detection decoupling by the control module, the response trajectory in the complex plane exhibits a center-oriented contraction characteristic, and the corresponding response trajectory direction consistency coefficient approaches 1. This result verifies that the present invention can accurately separate mechanical disturbances and extract the pure electrochemical state evolution characteristics of the reaction system under complex stirring environments.

[0162] according to Figure 9 It can be seen that the product yield curve of the control group showed a significant slowdown in growth after 80 minutes of reaction, with the final product yield remaining at around 86.5%. Simultaneously, due to localized deterioration of mass transfer and localized accumulation of unreacted raw materials, the byproduct content shown on the secondary ordinate exhibited a non-linear surge during this stage, ultimately reaching 4.2%. The product yield curve of the experimental group maintained a stable growth range for a longer period, eventually rising to 97.2%, and the byproduct content was consistently suppressed below 0.3% throughout the reaction. Furthermore, the experimental group accurately output a valve closing command at 102 minutes based on the aforementioned reaction endpoint determination conditions, shortening the production cycle compared to the average reaction time of 135 minutes for the control group. This comparative data demonstrates that the present invention, by dynamically coordinating the stirring frequency and feeding opening, eliminates the localized mass transfer encapsulation phenomenon caused by increased material viscosity in the later stages of the reaction, reducing the localized supersaturation concentration conditions for side reactions, and improving product yield and purity while shortening the reaction time.

[0163] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be subject to the final approved scope of protection.

Claims

1. A method for intelligent regulation of the glycine magnesium synthesis process, characterized in that, Includes the following steps: Obtain the impedance phase angle and instantaneous active power and filter them to obtain the baseline components of the impedance phase angle and active power. The disturbance frequency is determined based on the active power baseline component to generate a stirring frequency command, and the active power transient fluctuation component is extracted. Within the adaptive integration window determined based on the disturbance frequency, the impedance phase angle and the transient fluctuation component of the active power are subjected to dual-channel synchronous detection. The projection coefficient of the mechanically related component is calculated to subtract the response component with the same frequency as the power, and the decoupled electrochemical response component and the decoupled response intensity are obtained. The macroscopic reaction dynamic index is calculated using the baseline components of the impedance phase angle and the active power; a decoupled response vector is constructed based on the decoupled electrochemical response components to form a response trajectory, and the response trajectory direction consistency coefficient is calculated. The decoupled response intensity is processed by delay difference to obtain delay difference component. The delay difference component is used to update the basic stirring frequency. The macroscopic reaction dynamic index, the response trajectory direction consistency coefficient and the delay difference component are combined to generate the opening adjustment command. Based on the decoupling response intensity and the phase after phase expansion, the dynamic response activity index is calculated; the dynamic response activity index, the macroscopic reaction dynamic index, and the delay difference component are jointly judged, and a valve closing command is output when the reaction endpoint determination condition is met.

2. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The step of determining the disturbance frequency based on the active power baseline component to generate a stirring frequency command, and extracting the active power transient fluctuation component, specifically includes: After normalizing the active power baseline component to rated power, it is multiplied by the set amplitude proportional gain coefficient and frequency proportional coefficient to obtain the stirring frequency disturbance amplitude and the disturbance frequency within the effective operating range. The angular frequencies corresponding to the disturbance frequencies are discretely accumulated within the discrete sampling period to generate a continuous disturbance phase; A sinusoidal reference signal is generated using the perturbation phase. The sinusoidal reference signal is multiplied by the amplitude of the stirring frequency perturbation and then superimposed onto the current base stirring frequency to obtain the stirring frequency command. The transient fluctuation component of the active power is obtained by subtracting the instantaneous active power from the baseline component of the active power.

3. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The dual-channel synchronous detection specifically includes: The time length between the current moment and the position in historical data where the perturbation phase difference reaches the number of complete cycles is used as the adaptive integration window; Based on the disturbance phase, a sinusoidal reference signal and a cosine reference signal are generated. The impedance phase angle is multiplied by the sinusoidal reference signal and the cosine reference signal respectively and integrated within the adaptive integration window to obtain the impedance phase in-phase response component and the impedance phase quadrature response component. The transient fluctuation component of active power is multiplied by a sinusoidal reference signal and integrated within the adaptive integration window to obtain the power frequency response component.

4. The intelligent control method for the synthesis process of glycine magnesium according to claim 3, characterized in that, The calculation of the mechanically correlated component projection coefficients, after subtracting the power-frequency response component, yields the decoupled electrochemical response component and the decoupled response intensity, specifically including: The window linear projection relationship between the impedance phase in-phase response component and the impedance phase quadrature response component and the power frequency response component is calculated respectively to obtain the corresponding mechanical correlation component projection coefficients; Using the projection coefficients of the mechanically correlated components, the parts linearly related to the power-frequency response component are subtracted from the impedance-phase in-phase response component and the impedance-phase quadrature response component, respectively, to obtain the decoupled electrochemical response components as real and imaginary parts; The decoupling response intensity is obtained by taking the root of the square of the real and imaginary parts.

5. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The calculation of the macroscopic response dynamic index using the baseline component of the impedance phase angle and the baseline component of the active power specifically includes: Calculate the time first derivatives of the baseline component of the impedance phase angle and the baseline component of the active power at the current sampling time; The time first derivatives of the impedance phase angle baseline component and the active power baseline component are scaled by dividing them by their sliding standard deviations within the adaptive integration window, respectively. The two values ​​after scale standardization are added together to obtain the macroscopic response dynamic index, which characterizes the combined result of the relative change of the baseline.

6. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The construction of decoupled response vectors to form response trajectories and the calculation of response trajectory direction consistency coefficients specifically include: Extract the decoupled response vector sequence corresponding to the continuous sampling time within the current adaptive integration window; The difference magnitude of the first and last decoupled response vectors within the adaptive integral window is calculated as the overall displacement of the response trajectory, and the cumulative difference magnitude of the decoupled response vectors at adjacent sampling times is calculated as the cumulative path length. The total position is removed by the cumulative path length to obtain the response trajectory direction consistency coefficient, which characterizes the degree of variation of the pure electrochemical response along a consistent direction.

7. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The step of performing delay differential processing on the decoupling response intensity to obtain a delay differential component, and using the delay differential component to update the basic stirring frequency, specifically includes: The historical decoupling response intensity, which is separated from the current time by one adaptive integral window, is read by a circular data buffer queue, and the current decoupling response intensity is subtracted from the historical decoupling response intensity to obtain the delay difference component that characterizes the degree of net change in the response. When the delay difference is negative or positive, the discrete integral accumulation update of the basic stirring frequency is performed in the opposite direction in combination with the integral speed regulation gain, and the integral accumulation in the corresponding direction is stopped when the operating frequency limit boundary is reached.

8. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The process of generating an opening adjustment command by combining the macroscopic response dynamic index, the response trajectory direction consistency coefficient, and the delay difference component specifically includes: Extract the positive change portion and the absolute negative change portion of the delay difference component; Using the preset baseline opening degree as the base term, the macroscopic response dynamic index multiplied by the proportional coefficient is used as the positive driving term; The positive change portion is multiplied by the adjustment gain corresponding to the response trajectory direction consistency coefficient and the positive delay difference component as the first negative suppression term, and the absolute portion of the negative change is multiplied by the adjustment gain corresponding to the negative delay difference component as the second negative suppression term. After adding the basic term to the positive drive term, subtracting the first negative suppression term and the second negative suppression term, the opening adjustment command is generated.

9. The intelligent control method for the synthesis process of glycine magnesium according to claim 1, characterized in that, The dynamic response activity index is calculated based on the decoupling response intensity and the phase after phase expansion. The joint judgment of the dynamic response activity index, the macroscopic response dynamic index, and the delay difference component specifically includes: Within the adaptive integration window, the product of the square of the decoupling response intensity and the time rate of change of the phase is integrated to obtain the dynamic response activity index. Obtain the time rate of change of the dynamic response activity index and the time rate of change of the macro-response dynamic index; When the absolute values ​​of the dynamic response activity index and time change rate, the macroscopic response dynamic index and time change rate, and the delay difference component are all synchronously less than or equal to the threshold corresponding to the endpoint limitation condition, and the duration meets the set number of disturbance cycles, the reaction endpoint determination condition is confirmed to be met.

10. A smart control system for the synthesis of magnesium glycine, applied to the smart control method for the synthesis of magnesium glycine as described in any one of claims 1-9, characterized in that, include: The reaction module includes a reaction vessel equipped with a stirring mechanism, a jacket, and a feed pipeline. The reaction vessel is used to contain the reactants and provide space for the chemical reaction and mixing of the reactants. The jacket is used to introduce an external heat exchange medium to exchange heat with the reactants inside the reaction vessel. The sensing module includes an AC impedance detection unit installed inside the reactor and an active power calculation unit for acquiring the active power of the stirring motor. The active power calculation unit is used to acquire the impedance phase angle of the reactants and the instantaneous active power during the operation of the stirring motor. The active power calculation unit can be deployed in the motor drive unit inside the execution module and transmit the calculated instantaneous active power to the control module. The execution module, including a frequency converter, a stirring motor, and a feeding regulating valve, is used to adjust the operating frequency of the stirring motor and the feeding rate of the reaction raw materials according to the control commands output by the control module. The control module, including a programmable logic controller, is used to perform signal processing and related component decoupling processing on impedance phase angle and active power, and generate stirring frequency command and feeding regulating valve opening adjustment command based on the processing results. The communication module, including the industrial fieldbus, is used to connect the sensing module, the execution module, and the control module, enabling the transmission of detection data to the control module and the issuance of control commands to the execution module.