Preparation method of starch-based concrete additive based on multi-element synergistic grafting and digital control

Abstract

The invention relates to the technical field of green high-performance concrete admixtures, and discloses a multi-element synergistic grafting and digital control starch-based concrete additive preparation method, which comprises: S1, pre-establishing a soft measurement model for online analysis of monomer concentration, and a kinetic model capable of predicting the dynamic change of a reaction process; s2, in the reaction process, collecting process information such as spectrums on line, and analyzing the current concentrations of various monomers in real time by using a soft measurement model; s3, taking the real-time concentration as an initial condition, and solving a control strategy taking the optimal comprehensive performance of a final product as a target on line by utilizing the kinetic model; and S4, according to the optimized control strategy, implementing dynamic closed-loop regulation and control on the reaction process by adjusting operation variables such as dropping speed and temperature. Through the soft measurement and model prediction control technology, online optimization of the reaction process facing terminal application performance is realized, and accurate regulation and control of product quality and batch stability are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of green high-performance concrete admixtures, and specifically to a method for preparing a starch-based concrete additive through multi-element synergistic grafting and digital control. Background Art

[0002] The present invention relates to the intersection of polymer chemistry and civil engineering, and specifically to a method for preparing a starch-based concrete additive with multi-component synergistic grafting and digital control. As a natural polymer material with a wide range of sources, low cost and biodegradability, starch has become an important direction for the development of green high-performance concrete admixtures by giving it specific functions through chemical modification. By selectively grafting a variety of functional monomers onto the starch macromolecular skeleton, additives with comprehensive properties such as excellent water dispersibility, long-term slump retention and strong mud resistance can be prepared. However, how to accurately control this complex liquid-phase free radical copolymerization process and stably obtain the molecular structure and final performance that meet specific engineering needs is a key technical challenge that needs to be urgently solved in this field.

[0003] In the prior art, the preparation of these starch-based additives is typically carried out in a batch or semi-batch polymerization reactor equipped with an agitator and a jacketed temperature control system. The typical process flow is as follows: First, native starch is gelatinized in an aqueous medium to form a uniform starch matrix solution. Subsequently, an initiator system (preferably a redox initiator system, such as ammonium persulfate-sodium bisulfite) is activated at a set reaction temperature, and an aqueous solution of one or more functional monomers is added dropwise according to a predetermined schedule. Throughout the reaction, the operator typically controls the addition rate of each material and the reactor temperature profile according to a pre-defined process recipe. To monitor the progress of the reaction, offline analysis is commonly employed. Samples are periodically withdrawn from the reactor and sent to the laboratory for analysis using methods such as high-performance liquid chromatography (HPLC), chemical titration, or Fourier transform infrared spectroscopy (FTIR) to determine key indicators such as the residual monomer concentration and the grafting yield of the product.

[0004] Although the existing technology can produce starch-based additives using the above-mentioned method, it still has some shortcomings in terms of process control accuracy and product performance stability. Existing control strategies mainly rely on offline detection, which leads to serious control lag. The fundamental reason is that the entire process from reactor sampling, sample pretreatment, instrument analysis to data interpretation is time-consuming. When the analysis results are fed back to the control system, the actual working conditions in the reactor have long deviated from the state at the time of sampling. This makes any control measures based on this result lose real-time performance, unable to truly achieve precise guidance of the reaction dynamic process, and even more difficult to cope with sudden disturbances such as differences between raw material batches or ambient temperature fluctuations.

[0005] In addition, the correlation established by traditional processes between process control and the performance of the final product is relatively weak. Control systems often target process variables that are easy to measure, such as monomer conversion rate or the concentration of a certain intermediate product. However, there is a complex non-linear relationship between these variables and the terminal performance indicators such as the water dispersion ability and slump retention time of the additives in concrete applications. Due to the lack of a mathematical model that directly correlates the two, the optimization of the production process often relies on a large number of experimental trials and errors, making it difficult to achieve "customized" production for specific engineering requirements. This "open-loop" control logic and the excessive reliance on the experience of operators ultimately lead to difficulty in ensuring the performance consistency between product batches. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the present invention provides a method for preparing starch-based concrete additives with multi-component synergistic grafting and digital control, which solves the problems in the prior art that due to the reliance on off-line detection and preset open-loop control strategies, the starch-based graft copolymerization reaction process is difficult to accurately regulate, the product batch performance is unstable, and it cannot be directly correlated with the terminal application performance.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for preparing starch-based concrete additives with multi-component synergistic grafting and digital control, comprising the following steps: Pre-establish a soft measurement model capable of analyzing and obtaining the concentrations of multiple monomers based on the online process information of the reaction system, and a reaction process kinetics model capable of predicting the dynamic changes in the graft copolymerization reaction process; During the progress of the graft copolymerization reaction, use the reaction process kinetics model to online obtain the process information of the reaction system, and use the previously established soft measurement model to real-time analyze and obtain the current concentrations of multiple monomers in the reaction system; Based on the current concentrations of multiple monomers obtained by this real-time analysis as the initial conditions, use the previously established reaction process kinetics model to online optimize and solve a control strategy that can optimize the preset comprehensive performance indicators of the final product; And regulate the process of the graft copolymerization reaction according to the control strategy obtained by this optimization.

[0008] The step of pre-establishing the soft measurement model includes: First, conduct a series of grafting reaction experiments covering different process conditions, and synchronously collect the online spectral data of the reaction system and samples for off-line analysis during the experiments; Secondly, determine the absolute concentrations of multiple monomers in the samples by off-line analysis methods; Finally, use the partial least squares method to perform regression analysis on the spectral data and the corresponding monomer concentrations to establish a mapping relationship between the spectral information and the actual concentrations, and its mathematical model is: ; In the formula, is the true monomer concentration matrix measured by offline analysis; is the online spectral data matrix collected synchronously; is the regression coefficient matrix of the PLS model obtained through regression training; is the residual matrix of the model.

[0009] Preferably, first of all, the present invention needs to pre - establish a "soft - sensing model" that can characterize the internal relationship of the reaction system. The establishment process of this model is as follows: in multiple reference experiments, through sensors such as online spectral probes inserted into the reaction kettle, "online process information" (such as infrared spectrum, Raman spectrum, etc.) is continuously obtained; at the same time, physical sampling is carried out at different time points, and standard chemical analysis methods such as high - performance liquid chromatography (HPLC) are used to accurately measure the "concentrations of multiple monomers" in the samples.

[0010] Finally, using chemometric algorithms such as partial least squares method, a large amount of online process information is correlated and regression - analyzed with the corresponding true values of monomer concentrations, so as to construct and calibrate the soft - sensing model.

[0011] In the actual production process of graft copolymerization reaction, the system will continuously "online obtain the process information of the reaction system" and use it as input, and transmit it to the previously established soft - sensing model in real - time. The model performs high - speed operations according to the calibrated mathematical relationship, so as to "real - time analyze and obtain the current concentrations of multiple monomers in the reaction system".

[0012] Through soft - sensing technology, the time to obtain monomer concentration data is shortened, so that the control system can "clearly see" the real chemical component changes inside the reaction kettle in real - time and accurately on a time scale matching the dynamic changes of the reaction.

[0013] The steps of pre - establishing the reaction process kinetic model include: By experimental determination and parameter separation method, a kinetic equation describing the change rate of grafting degree is constructed, and its core equation is: ; In the formula: is the grafting degree, which is measured by sampling during the reaction using the ninhydrin method; is the pre - exponential factor; is the initiator concentration, using the ammonium persulfate - sodium bisulfite redox system; is the reaction order of the initiator; is the total monomer concentration; is the reaction order of the monomer; is the apparent activation energy; is the starch concentration; is the starch concentration index; Determine each parameter in the equation through the following steps: Temperature gradient experiment: Fix unchanged, conduct the reaction in the range of 50 - 80 °C, and obtain through linear regression and ; Initiator concentration experiment: Fix the temperature and unchanged, change , and determine through the slope of = 0.5; Monomer concentration experiment: Fix the temperature and unchanged, change , and determine through the slope of = 1.0; Starch concentration experiment: Fix the temperature and unchanged, change , and determine through the slope of = -0.2; Substitute the obtained parameters into the equation, predict the graft degree change curve of the experimental group that did not participate in the parameter fitting, verify that the prediction error is less than 5%, and couple the verified kinetic equation with the material balance equation and the energy balance equation to form a complete reaction process kinetic model.

[0014] Preferably, before starting the optimization, it is also necessary to pre - establish a reaction process kinetic model that can predict the dynamic changes in the graft copolymerization reaction process. During the reaction, the system inputs the "current concentrations of various monomers parsed from S2 in real - time as initial conditions" into this kinetic model. The current concentrations of various monomers include, but are not limited to: the current concentration of the first functional monomer (such as itaconic anhydride that confers carboxyl groups), the current concentration of the second functional monomer (such as glycidyl acrylate that confers epoxy groups), and the current concentration of the third functional monomer (such as zwitterionic monomer that confers sulfonic acid groups). Subsequently, the model predictive control (MPC) algorithm starts the "online optimization and solution" process. Specifically, with an optimization goal of "optimizing the comprehensive performance index of the preset final product", and on the premise of meeting various process constraints (such as temperature and pressure safety ranges), through repeated deduction of the kinetic model, calculate how to operate (such as adjusting the dropping rates of each material and the reaction temperature) in the next period of time to make the comprehensive performance of the final product (such as water dispersibility, slump retention, etc.) reach the best, thereby overcoming uncertainties such as raw material fluctuations and environmental changes, and improving the performance consistency and stability between product batches.

[0015] The steps of online obtaining process information and real-time parsing to obtain the current concentration include: Through an online Fourier transform infrared probe, the mid-infrared full-spectrum data of the reaction system is collected in real time as the online process information; Substitute the spectrogram data collected in real time into the pre-established soft measurement model to instantaneously calculate the current concentrations of various monomers in the reaction system.

[0016] In the steps of online optimizing and solving the control strategy, the preset comprehensive performance index of the final product is an evaluation function, which is established based on the performance of the auxiliary agent in concrete applications, and its specific form is: ; In the formula, is the comprehensive performance index to be maximized; is the water dispersion performance index; is the hydrolysis kinetic constant of the slow-release unit; is the characteristic evaluation time of the slump retention performance; is the anti-clay performance index; , , are the weight factors of each performance.

[0017] The steps of online optimizing and solving the control strategy are implemented through a model predictive control algorithm, which repeatedly executes a complete optimization process within a rolling time domain: First, based on the current concentrations of the various monomers as initial conditions, use the reaction process kinetic model to predict the evolution trajectory of the reaction system state within a future prediction time domain under different alternative control strategies; Then, solve an open-loop optimization problem with the goal of optimizing the comprehensive performance index of the final product to obtain an optimal control sequence covering multiple future steps; Finally, only execute the first control action in the optimal control sequence and repeat this optimization process at the next control moment.

[0018] The steps of solving the open-loop optimization problem include: Under the conditions of satisfying the process model, initial state, control input, and state variable constraints, find the optimal control sequence that maximizes the following objective function: ; In the formula, is the current control moment; is the length of the prediction time domain; is the predicted value of the state at the end of the prediction time domain at time ; is at time Planned control input.

[0019] The step of regulating the reaction process according to the control strategy includes: According to the control strategy obtained by the optimization, at least one or more process variables selected from the following are adjusted in real time: The dropping flow rate of each monomer solution, the dropping flow rate of the initiator solution, and the heating or cooling power of the reaction kettle jacket.

[0020] The method further includes a reaction end determination step, which is judged based on the result of the online optimization solution. Specifically: When the future gain of the predicted comprehensive performance index of the final product is lower than the preset threshold, it is determined that the reaction reaches the optimal end point and the reaction is terminated; The definition of the preset threshold is: Within three consecutive control cycles, the predicted Increment Is less than or equal to 0.5% of the current Value, where Is the predicted increment of the comprehensive performance index between adjacent control times; Is at time The predicted next moment Comprehensive performance index; Is the moment Current comprehensive performance index.

[0021] The present invention provides a method for preparing a starch-based concrete admixture with multi-component synergistic grafting and digital control. It has the following beneficial effects: 1. By combining the online spectral sensing technology with the soft measurement model, the present invention breaks through the lag limitation of traditional off-line sampling analysis, resolves the changes of various monomer concentrations in real time, and provides dynamic data support with high spatio-temporal resolution for the reaction process. Combining the multi-step prediction ability of the reaction kinetics model enables the control strategy to prospectively compensate for interference factors such as material consumption and temperature fluctuations, and improves the control accuracy of the grafting degree and batch consistency.

[0022] 2. By inversely mapping the terminal performance indexes of the concrete admixture (water dispersibility, slump retention, and anti-sludging property) to the reaction process control target, the present invention realizes the closed-loop linkage between process parameters and product performance through the comprehensive performance index function. This design enables the production process to dynamically adjust and optimize the weights according to different engineering requirements, and flexibly adapt to diverse concrete mix scenarios.

[0023] 3. Based on the model predictive control (MPC) framework, the present invention utilizes the predictive ability of the kinetic model for the reaction trend to adjust the operating parameters in advance when external disturbances (such as raw material fluctuations and environmental temperature changes) have not significantly affected the grafting degree. This feedforward-feedback composite control mechanism effectively suppresses the overshoot and oscillation phenomena commonly seen in traditional PID control. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the specification of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0026] ]>Please refer to the attached Figure 1 , the embodiments of the present invention provide a method for preparing a starch-based concrete admixture with multi-component synergistic grafting and digital control, including the following steps: S1. First, establish a soft sensor model capable of analyzing and obtaining various monomer concentrations based on the online process information of the reaction system, and a reaction process kinetic model capable of predicting the dynamic changes in the graft copolymerization reaction process; S1 constructs the mapping relationship between spectral information and monomer concentration through experimental design and multivariate calibration algorithms.

[0027] First, in a laboratory-scale reaction device, design graft copolymerization experiments covering different temperatures, material ratios, and reaction times to ensure that the process conditions cover the operating ranges that may occur in actual production. During the reaction process, real-time collect the mid-infrared full-spectrum data of the reaction system through an online Fourier transform infrared (FTIR) spectrometer to form a spectral data matrix , where each row corresponds to an experimental sample and each column corresponds to the absorbance value at a specific wavenumber.

[0028] At the same time, at the synchronous time point of spectral collection, sample the reaction solution and accurately measure the concentrations of three functional monomers (itaconic anhydride, glycidyl acrylate, sulfobetaine) in the sample through offline analysis methods such as high performance liquid chromatography (HPLC) to form a concentration data matrix . Use the partial least squares regression (PLS) algorithm to model the spectral data and the concentration data, and its mathematical expression is: ; In the formula, is the true monomer concentration matrix measured through offline analysis; is the online spectral data matrix collected synchronously; is the regression coefficient matrix of the PLS model obtained through regression training; is the residual matrix of the model.

[0029] Based on the principles of chemical reaction engineering and experimental data fitting, a state - space model describing the dynamic behavior of the graft copolymerization reaction is established. The core of the kinetic model is the rate equation of the degree of grafting, and its form is: ; In the formula: is the degree of grafting, which is measured by sampling during the reaction using the ninhydrin method; is the pre - exponential factor; is the initiator concentration, using the ammonium persulfate - sodium bisulfite redox system; is the reaction order of the initiator; is the total monomer concentration; is the reaction order of the monomer; is the apparent activation energy; is the starch concentration; is the starch concentration exponent.

[0030] Temperature gradient experiments are carried out in a constant - temperature reaction kettle. The reaction is carried out in the range of 50 - 80 °C, with other conditions fixed. By linear regression, and ; By changing the monomer concentration experiment, using the slope to determine = 0.5; By changing the monomer concentration experiment, using the slope to determine = 1.0; By changing the starch concentration experiment, using the slope to determine = - 0.2; Substitute the calibrated parameters into the kinetic equation, predict the degree - of - grafting change curve of the independent experimental group, verify that the prediction error is less than 5%. Couple the verified degree - of - grafting kinetic equation with the following auxiliary equations to form a complete kinetic model of the reaction process: Material balance equation: Describes the consumption and replenishment dynamics of each monomer and initiator in the reaction system; Energy balance equation: Reflects the influence of the heat of reaction and the external heating / cooling power on the system temperature; State - space expression: Integrates the above equations to describe the system dynamics in the form of the system state vector and the control input vector ​

[0031] The material balance equation and the energy balance equation are prior art and will not be elaborated further in this article.

[0032] S2. During the graft copolymerization reaction, use the reaction process kinetic model to obtain the process information of the reaction system online, and use the soft sensor model established above to analytically obtain the current concentrations of various monomers in the reaction system in real time; In this embodiment, the technical solution of obtaining process information online and analytically obtaining monomer concentrations in real time is realized through the synergistic effect of online spectroscopic sensing technology and the soft sensor model. The specific implementation process is as follows: During the graft copolymerization reaction, through an online Fourier transform infrared (FTIR) spectroscopic probe integrated in the reaction kettle, mid-infrared full-spectrum data of the reaction system are continuously collected at a preset sampling frequency. The spectral data cover the absorption band ranges of characteristic functional groups (for example: characteristic absorption peaks of carbonyl, epoxy, and sulfonic acid groups), forming a real-time spectral vector . The spectral acquisition system transmits data to the control unit through optical fibers to ensure real-time performance and anti-interference ability.

[0033] Embed the pre-trained partial least squares (PLS) soft sensor model into the real-time control system. For each sampling moment , the system performs the following operations: Data preprocessing: Perform standardization processing on the real-time collected spectral vector to eliminate baseline drift and noise interference, specifically including Savitzky-Golay smoothing filtering and multiplicative scatter correction (MSC); Concentration analysis: Input the preprocessed spectral data into the PLS model, and calculate the estimated values of the concentrations of each monomer in real time through matrix operations: ; In the formula, is the regression coefficient matrix obtained in the model training stage, and its dimension matches the number of spectral variables and the types of monomers to be measured; is the real-time collected spectral vector; Result output: The analytically obtained concentration vector contains the real-time concentration values of each monomer and serves as the input parameter for subsequent optimal control.

[0034] To ensure the reliability of the soft sensor results, the system sets the following verification mechanisms: Spectral quality assessment: Real-time calculate the signal-to-noise ratio (SNR) and absorbance dynamic range of the collected spectrum. When the SNR is lower than the threshold, trigger an alarm and start resampling; Concentration rationality verification: Compare the current concentration estimate with the historical data of the previous time period. If a mutation occurs (e.g., the single change rate exceeds 10%), start the redundant calculation or manual intervention process.

[0035] The real-time concentration data obtained by parsing Will be input into the reaction process kinetics model as the initial condition. Specifically, at the beginning of each optimization cycle of model predictive control (MPC), the system substitutes the monomer concentrations into the state vector And update the following key parameters: Consumption of initiator: Calculate the current effective concentration of the initiator according to the initiator concentration decay kinetics equation; Reaction progress index: Update the cumulative grafting degree by integral form to provide a basis for end point determination.

[0036] All parsing results and the original spectral data are stored in the distributed time series database, and the concentration change curve, spectral quality index and system alarm status are dynamically displayed through the human-machine interface (HMI) to provide process monitoring support for operators.

[0037] S3. Based on the current concentrations of multiple monomers obtained by the real-time parsing as the initial conditions, use the reaction process kinetics model established above to online optimize and solve a control strategy that can optimize the preset comprehensive performance index of the final product; In this embodiment, the technical solution of online optimizing and solving the control strategy is realized through the collaborative optimization of the model predictive control (MPC) algorithm and the reaction process kinetics model. The specific implementation process is as follows: Based on the comprehensive performance index , construct an optimization objective function oriented to the application performance of concrete additives. The function integrates the quantitative indexes of water dispersion performance, slump retention performance and anti-clay performance, and its mathematical form is: ; In the formula, Is the comprehensive performance index to be maximized; Is the water dispersion performance index; Is the hydrolysis kinetic constant of the slow-release unit; Is the characteristic evaluation time of the slump retention performance; Is the anti-clay performance index; , , Are the weight factors of each performance.

[0038] Adopt a rolling horizon optimization strategy and perform the following operations in each control cycle Execute the following: State initialization: The real-time monomer concentration parsed in step S2 As the initial state vector , and load the reaction process kinetic model; Predict the evolution trajectory of the system state within the future time domain under different control input sequences , where , in the formula, is the control input vector at the discrete time step ; is the length of the prediction time domain; is the predicted value of the state at the future time at the control moment ; is the current control moment; is the planned control input at the moment ; Open-loop optimization: Solve the following constrained optimization problem: ; , in the formula, is the current control moment; is the length of the prediction time domain; is the predicted value of the state at the end of the prediction time domain at the moment ; is the planned control input at the moment ; Control execution: Select the first control action in the optimized optimal control sequence , and convert it into an actual execution instruction (such as adjusting the metering pump speed, adjusting the jacket temperature set value).

[0039] The optimization algorithm is also implemented through the following technologies: Discretization processing: Convert the continuous-time kinetic model into a discrete-time state-space model, and use the fourth-order Runge-Kutta method for numerical integration. The time step is synchronized with the control system sampling period; Gradient calculation: Calculate the gradient of the objective function with respect to the control input through the adjoint equation method or the finite difference method to accelerate the optimization convergence; Real-time solver: Use the interior point method or the sequential quadratic programming (SQP) algorithm to solve the constrained nonlinear optimization problem, and embed a pre-compiled optimization library (such as IPOPT) to ensure the calculation efficiency.

[0040] The above fourth-order Runge-Kutta method, adjoint equation method or finite difference method, interior point method or sequential quadratic programming algorithm, and pre-compiled optimization library are all existing technologies and will not be elaborated too much in this article.

[0041] To meet the requirements of different concrete engineering (such as high water reduction rate priority or long slump retention time priority), a dynamic configuration interface for the weight factor is set. Operators can adjust it in real time through the human-machine interface (HMI) , , For the value of , the system will automatically update the objective function and trigger re-optimization.

[0042] S4. Regulate the process of the graft copolymerization reaction according to the control strategy obtained from this optimization; In this embodiment, the technical solution for dynamically regulating the reaction process according to the optimization strategy is achieved through the collaborative action of a closed-loop control architecture and multiple actuators. The specific implementation process is as follows: Convert the optimal control sequence output by the model predictive control (MPC) algorithm into executable operation instructions. The control vector includes the following key operating variables: The flow rate set value of each monomer solution metering pump ; The flow rate set value of the initiator solution metering pump ; The set value of the reaction kettle jacket temperature controller .

[0043] Send the set value instructions to the corresponding actuators through an industrial fieldbus (such as PROFIBUS or Modbus) to ensure the real-time and reliability of instruction transmission.

[0044] Flow control subsystem: Each metering pump is equipped with a high-precision stepper motor and a flow sensor, and the proportional-integral-derivative (PID) control algorithm is used to adjust the rotation speed in real time so that the actual flow rate quickly tracks the set value . The flow feedback signal eliminates pulsating noise through Kalman filtering to ensure control stability.

[0045] Temperature control subsystem: The jacket temperature controller dynamically adjusts the heater power and the opening degree of the cooling water valve according to the deviation between the set value and the actual temperature of the reaction system .

[0046] Preferably, a feedforward-feedback composite control strategy is adopted to predict the heat load change through a reaction heat model and compensate for temperature fluctuations in advance.

[0047] After the control instructions are executed, verify the regulation effect through the following mechanism: Flow deviation monitoring: Calculate the relative deviation between the actual flow rate of each material and the set value in real time. When the deviation continuously exceeds 5%, trigger an alarm and start the flow calibration program; Temperature gradient monitoring: Detect the uniformity of the internal temperature distribution of the reaction system through a distributed temperature sensor array. If the maximum temperature difference exceeds 2°C, automatically adjust the stirring rate or the jacket partition temperature control strategy.

[0048] After each control action is executed, the spectral data of the reaction system is recollected. , and the estimated value of the monomer concentration is updated through the soft sensor model. The updated concentration data is fed back to the MPC optimizer to form a closed-loop control loop of "measurement - estimation - optimization - execution" to achieve adaptive regulation under dynamic working conditions.

[0049] When the system detects the following abnormal conditions, it automatically switches to the safe operation mode: Material shortage warning: When the tank level sensor detects that the reserve of any raw material is lower than the safety threshold, feeding is paused and replenishment is notified. Runaway reaction warning: When the temperature rising rate exceeds the safety limit or the pressure sensor fluctuates abnormally, an emergency cooling and pressure relief procedure is started. Model mismatch detection: When the prediction residual of the soft sensor model continuously increases, online model update is automatically triggered or the standby control strategy is switched to.

[0050] All control instructions, actuator states, process variables, and alarm events are stored in the industrial database in the form of timestamps, supporting multi-dimensional retrieval and analysis by batch, time range, or abnormal type, providing a data basis for process optimization and fault diagnosis.

[0051] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A preparation method of a starch-based concrete aid with multi-component synergistic grafting and digital control, characterized in that It includes the following steps: S1. Pre - establish a soft - sensing model capable of parsing multiple monomer concentrations from the online process information of the reaction system, and a reaction - process kinetic model capable of predicting the dynamic changes in the graft copolymerization reaction process; S2. During the graft copolymerization reaction process, use the reaction - process kinetic model to online obtain the process information of the reaction system, and use the previously established soft - sensing model to real - time analyze and obtain the current concentrations of multiple monomers in the reaction system; S3. Based on the current concentrations of multiple monomers obtained by real - time analysis as the initial conditions, use the previously established reaction - process kinetic model to online optimize and solve a control strategy that can optimize the preset comprehensive performance index of the final product; S4. Regulate the process of the graft copolymerization reaction according to the optimized control strategy obtained.

2. The preparation method of the starch-based concrete admixture with multi-component synergistic grafting and digital control according to claim 1, wherein, The step of pre - establishing the soft - sensing model includes: First, conduct a series of graft reaction experiments covering different process conditions, and synchronously collect the online spectral data of the reaction system and samples for off - line analysis during the experiments; Second, determine the absolute concentrations of multiple monomers in the samples by off - line analysis methods; Finally, use the partial least - squares method to perform regression analysis on the spectral data and the corresponding monomer concentrations to establish a mapping relationship between spectral information and actual concentrations, and its mathematical model is: ； In the formula, is the true monomer concentration matrix measured by offline analysis; is the online spectral data matrix collected synchronously; is the regression coefficient matrix of the PLS model obtained through regression training; is the residual matrix of the model.

3. The preparation method of the starch-based concrete admixture with multi-component synergistic grafting and digital control according to claim 2, characterized in that, The step of pre - establishing the reaction - process kinetic model includes: Construct a kinetic equation describing the change rate of grafting degree through experimental determination and parameter separation method, and its core equation is: ； In the formula: is the grafting degree, which is measured by sampling during the reaction using the ninhydrin method; is the pre-exponential factor; is the initiator concentration, and an ammonium persulfate-sodium bisulfite redox system is adopted; is the reaction order of the initiator; is the total monomer concentration; is the reaction order of the monomer; is the apparent activation energy; is the starch concentration; is the starch concentration index; Determine each parameter in the equation through the following steps: Temperature gradient experiment: Fix unchanged, and carry out the reaction in the range of 50 - 80 °C. Obtain through linear regression and ; Initiator concentration experiment: Fix the temperature, remain unchanged, and change , and determine through the slope of ; Monomer concentration experiment: Fix the temperature, remain unchanged, change , and determine through the slope of ; Starch concentration experiment: Fix the temperature, remain unchanged, change , and determine through the slope of ; Substitute the obtained parameters into the equation, predict the grafting - degree change curve of the experimental group not participating in parameter fitting, verify that the prediction error is less than the first preset value, and couple the verified kinetic equation with the material balance equation and the energy balance equation to form a complete reaction - process kinetic model.

4. The preparation method of the starch-based concrete admixture with multi-component synergistic grafting and digital control according to claim 3, characterized in that, The step of real - time analyzing and obtaining the current concentrations of multiple monomers in the reaction system includes: Real - time collect the mid - infrared full - spectrum data of the reaction system through an online Fourier transform infrared probe as the online process information; Substitute the real - time collected spectral data into the previously established soft - sensing model to instantaneously calculate the current concentrations of multiple monomers in the reaction system.

5. The preparation method of the starch-based concrete aid with multi-component synergistic grafting and digital control according to claim 4, characterized in that, In the step of online optimizing and solving the control strategy, the preset comprehensive performance index of the final product is an evaluation function, which is established based on the performance of the additive in concrete applications, and its specific form is: ； In the formula, is the comprehensive performance index to be maximized; is the water dispersion performance index; is the hydrolysis kinetic constant of the sustained release unit; is the characteristic evaluation time of slump retention performance; is the anti-clay performance index; , , are the weight factors of each performance.

6. The preparation method of the starch-based concrete aid with multi-component synergistic grafting and digital control according to claim 5, wherein, The step of online optimizing and solving the control strategy is realized through a model predictive control algorithm, and this algorithm executes the complete optimization process in a rolling time domain: First, based on the current concentrations of multiple monomers as the initial conditions, use the reaction - process kinetic model to predict the evolution trajectory of the reaction - system state within a future prediction time domain under different alternative control strategies; Then, solve an open - loop optimization problem with the goal of optimizing the comprehensive performance index of the final product to obtain an optimal control sequence covering multiple future steps; Finally, only execute the first control action in the optimal control sequence and repeat this optimization process at the next control moment.

7. The preparation method of the starch-based concrete aid with multi-component synergistic grafting and digital control according to claim 6, characterized in that, The steps of solving an open-loop optimization problem with the goal of maximizing the comprehensive performance index of the final product include: Under the conditions of satisfying the process model, initial state, control input, and state variable constraints, find the optimal control sequence that maximizes the following objective function: ； wherein, is the current control time; is the length of the prediction time domain; at time is the predicted value of the state at the end of the prediction time domain; at time is the planned control input.

8. The preparation method of the starch-based concrete admixture with multi-component synergistic grafting and digital control according to claim 7, wherein, The steps of regulating the graft copolymerization process according to the control strategy obtained by the optimization solution include: According to the control strategy obtained by the optimization solution, adjust at least one or more of the following process variables in real time: The dropping flow rate of each monomer solution, the dropping flow rate of the initiator solution, and the heating or cooling power of the reactor jacket.

9. The preparation method of the starch-based concrete aid with multi-component synergistic grafting and digital control according to claim 8, characterized in that, The method further includes a reaction end determination step, which is judged based on the result of the online optimization solution. Specifically: When the future gain of the predicted comprehensive performance index of the final product is lower than the preset threshold, it is determined that the reaction reaches the optimal end point and the reaction is terminated; The definition of the preset threshold is: Within three consecutive control cycles, the predicted increment is less than or equal to the current value. Wherein, is the predicted increment of the comprehensive performance index between adjacent control instants; is the predicted comprehensive performance index at the next instant at time ; is the current comprehensive performance index at time .

Images