Preparation process optimization method and system for organic scale remover

Abstract

The invention relates to the technical field of process optimization, in particular to a preparation process optimization method and system for an organic descaling agent, and the method comprises the steps: obtaining raw material characteristic data, formula parameters and reaction condition parameters of a to-be-optimized preparation process; wherein the formula parameters comprise a plurality of reaction stages, an optimal process path of each reaction stage and a preset regulation and control node corresponding to the optimal process path, and the preset regulation and control node is set as a reaction turning point; based on the raw material characteristic data and the reaction condition parameters, parameter sensitivity distribution in the preparation process is calculated through reaction kinetics simulation, and the influence range of technological parameters is determined according to a parameter coupling model. According to the present invention, by carrying out the multi-stage process path optimization on the reaction process, the most suitable optimization measure can be adopted according to the reaction characteristic of each stage, such that the reaction efficiency of each stage is improved, and the unnecessary energy consumption and the raw material waste during the reaction process are reduced.

Description

Technical Field

[0001] This invention relates to the field of process optimization technology, specifically to a method and system for optimizing the preparation process of an organic descaling agent. Background Technology

[0002] In the preparation of organic descaling agents, the characteristics of raw materials and reaction conditions may fluctuate, which will directly affect the stability and efficiency of the reaction. Due to the complexity of the process, especially in actual production, the optimized scheme under standard experimental conditions may not be fully adapted to the actual situation. Therefore, how to ensure the stability and efficiency of the process under unstable raw materials and reaction conditions has become an urgent problem to be solved.

[0003] Currently, traditional preparation processes often suffer from excessive energy consumption and raw material waste during the reaction process. This is especially true in multi-stage reaction processes, where failure to accurately optimize reaction conditions and process parameters at each stage can lead to energy waste and unnecessary resource consumption, affecting production costs and efficiency.

[0004] Furthermore, during the process optimization process, the interaction between multiple process parameters may lead to complex coupling effects, increasing the difficulty of optimization. If the adjustment space is not scientifically and reasonably divided, or if each sub-region is not optimized independently, it is easy to cause instability in the adjustment effect and interference between parameters, thereby affecting the overall process optimization effect.

[0005] Moreover, under standard experimental conditions, process optimization through simulation models often fails to fully consider raw material variations and environmental fluctuations encountered in actual production. Traditional optimization methods fail to effectively correct for these variations by incorporating raw material characteristic data, leading to discrepancies between the laboratory and the production process and affecting the feasibility of the optimization scheme in actual production. Summary of the Invention

[0006] To achieve the above objectives, the present invention provides the following technical solution: an optimized preparation process method for an organic descaling agent, comprising: Obtain raw material characteristic data, formulation parameters, and reaction condition parameters for the preparation process to be optimized; wherein, the formulation parameters include several reaction stages, the optimal process path for each reaction stage, and the preset control nodes for the corresponding optimal process paths, and the preset control nodes are set as reaction inflection points. Based on the raw material characteristic data and reaction condition parameters, the parameter sensitivity distribution of the preparation process is calculated through reaction kinetics simulation, and the influence range of the process parameters is determined according to the parameter coupling model. Based on the influence range of the process parameters, the parameter adjustment space of the preparation process is divided into multiple interconnected and independently optimizable adjustment sub-regions, each of which includes process parameters. The execution order of each reaction stage is used as the first optimization parameter, and the core process parameters of each regulation sub-region are used as the second optimization parameter. Based on the first optimization parameter, the second optimization parameter, and the pre-trained reaction simulation model, the initial reaction efficiency of each preset control node is determined. The reaction simulation model is a mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions. The initial reaction efficiency is corrected based on the raw material characteristic data using the pre-trained influence correction model. According to the preset parameter adjustment order, sensitivity analysis and optimization calculations are performed on the process parameters of multiple adjustment sub-regions to obtain sub-region optimization parameter sets; the sub-region optimization parameter sets are coupled and verified to generate a global parameter optimization matrix; Based on the corrected reaction efficiency, the first optimized parameter, and the second optimized parameter, the first preparation time and the first raw material consumption are determined as optimization objectives. A multi-objective optimization problem is constructed by combining the global parameter optimization matrix. The multi-objective optimization problem is solved by a multi-objective optimization algorithm to obtain the optimal process parameter scheme.

[0007] Preferably, based on the raw material characteristic data and reaction condition parameters, the parameter sensitivity distribution of the preparation process is calculated through reaction kinetic simulation, and the influence range of the process parameters is determined according to the parameter coupling model, including: Input the content and purity of active ingredients from the raw material characteristic data, and the initial temperature and pressure from the reaction condition parameters into the reaction kinetics simulation model; The effect of a single parameter change on the reaction rate was simulated using the controlled variable method, and the sensitivity coefficients of each process parameter were calculated. The interaction between parameters is analyzed based on the parameter coupling model. The influence range of each process parameter is determined according to the influence boundary of the parameter whose sensitivity coefficient is greater than the preset threshold.

[0008] Preferably, based on the influence range of the process parameters, the parameter adjustment space of the preparation process is divided into multiple interconnected and independently optimizable adjustment sub-regions, including: The initial region is divided according to the reaction process sequence, with the parameter baseline value of the reaction initiation stage as the starting point. Adjust the initial region boundary according to the influence range of each process parameter to ensure that the parameter influence of adjacent regions does not overlap; Correlation analysis is performed on the process parameters in each region, and strongly correlated parameters are grouped into the same adjustment sub-region, resulting in multiple adjustment sub-regions.

[0009] Preferably, sensitivity analysis and optimization calculations are performed on the process parameters of multiple regulation sub-regions to obtain a set of optimized parameters for each sub-region, including: The preset parameter adjustment order is determined based on the parameter importance ranking, with priority given to optimizing process parameters with high sensitivity coefficients; For each regulation sub-region, the first priority parameter for this round of optimization is determined, and multi-level testing is conducted using response surface methodology to obtain the parameter effect curve; The optimal range of values ​​for the first priority parameter is determined based on the parameter effect curve, forming an optimized subset of the first parameter; Determine the second priority parameters sequentially according to the adjustment order, repeat the sensitivity test and optimization process, and obtain the optimized subset of the second parameters; Integrate the optimization subsets of each parameter to generate a sub-region optimization parameter set.

[0010] Preferably, the optimization parameter set of the sub-region is coupled and verified to generate a global parameter optimization matrix, including: Establish parameter correlation matrices between sub-regions to identify key parameters with cross-regional impact; Substitute the optimized parameter sets of each sub-region into the full-process reaction simulation to detect efficiency fluctuations caused by parameter coupling; The parameter combinations that exceed the fluctuation limit are corrected, and a global parameter optimization matrix including parameter coordination coefficients is constructed based on the correction results.

[0011] Preferably, the reaction efficiency is corrected based on a pre-trained impact correction model combined with raw material characteristic data, including: The impurity content, moisture content, and active ingredient stability parameters of each reaction stage are extracted from the raw material characteristic data as correction features. The initial reaction efficiency and the correction features are input into the influence correction model, which includes component feature gating and raw material perturbation gating; The influence of raw material components on the reaction is extracted by gating based on component characteristics, and the perturbation coefficient of historical data deviation is calculated by gating based on raw material perturbation. The initial reaction efficiency is corrected based on the perturbation coefficient and feature fusion results, and the corrected reaction efficiency is obtained.

[0012] Preferably, the training method for the reaction simulation model includes: Several process parameters related to the reaction efficiency mapping of organic descaling agents are determined; wherein, the reaction efficiency is the conversion rate of raw materials into target products per unit time under standard raw material conditions; Based on all the process parameters, an initial simulation model is established to characterize the mapping relationship between process parameters and reaction efficiency; A parameter sensitivity analysis was performed on the initial simulation model to obtain the weighting factors between process parameters and reaction efficiency; Based on the weighting factor, a preset simulation model is determined to characterize the mapping relationship between process parameters, weighting factor and reaction efficiency. The target data for the weighting factors are determined using a data fitting method, and a reaction simulation model under standard raw material conditions is determined based on the target data. The actual data of process parameters under standard raw material conditions are input into the reaction simulation model to verify the deviation between the predicted and actual values ​​of the reaction efficiency output by the model.

[0013] Preferably, a multi-objective optimization problem is constructed by combining the global parameter optimization matrix, including: The objective function is set with the optimization objectives of minimizing the first preparation time and minimizing the first raw material consumption; Based on the parameter coordination coefficients in the global parameter optimization matrix, the value constraints and interaction constraints of each process parameter are determined. The reaction temperature range, raw material ratio limits, and product purity standards are treated as hard constraints and incorporated into the constraints of the multi-objective optimization problem.

[0014] Preferably, the optimal process parameter scheme is obtained by solving the multi-objective optimization problem using a multi-objective optimization algorithm, including: A non-dominated sorting genetic algorithm is used to solve the multi-objective optimization problem, and the population includes various combinations of process parameters. The objective function value of each individual in the population is calculated based on the global parameter optimization matrix, and non-dominated sorting and crowding calculation are performed. The next generation population is generated through crossover and mutation operations, while retaining the Pareto optimal solution; After iterating to a preset number of times, the parameter combination with the best overall performance is selected from the final Pareto optimal solution set as the optimal process parameter scheme.

[0015] An optimization system for the preparation process of an organic descaling agent, applicable to the aforementioned optimization method for the preparation process of an organic descaling agent, comprising: The parameter acquisition unit is used to acquire raw material characteristic data, formulation parameters, and reaction condition parameters of the preparation process to be optimized; wherein, the formulation parameters include several reaction stages, the optimal process path for each reaction stage, and the preset control nodes corresponding to the optimal process path, and the preset control nodes are set as reaction inflection points. The parameter coupling unit is used to calculate the parameter sensitivity distribution of the preparation process through reaction kinetics simulation based on the raw material characteristic data and reaction condition parameters, and to determine the influence range of process parameters according to the parameter coupling model. The range division unit is used to divide the parameter adjustment space of the preparation process into multiple interconnected and independently optimizable adjustment sub-regions according to the influence range of the process parameters. Each adjustment sub-region includes the process parameters. The efficiency analysis unit is used to take the execution order of each reaction stage as the first optimization parameter and the core process parameters of each regulation sub-region as the second optimization parameter; based on the first optimization parameter, the second optimization parameter and the pre-trained reaction simulation model, the initial reaction efficiency of each preset control node is determined, wherein the reaction simulation model is a mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions; the initial reaction efficiency is corrected based on the raw material characteristic data using the pre-trained influence correction model. The parameter optimization unit is used to perform sensitivity analysis and optimization calculations on the process parameters of multiple adjustment sub-regions according to a preset parameter adjustment order to obtain a sub-region optimized parameter set; and to perform coupling verification on the sub-region optimized parameter set to generate a global parameter optimization matrix. The process optimization unit is used to determine the first preparation time and the first raw material consumption as optimization objectives based on the corrected reaction efficiency, the first optimization parameter, and the second optimization parameter, and to construct a multi-objective optimization problem by combining the global parameter optimization matrix; and to solve the multi-objective optimization problem by using a multi-objective optimization algorithm to obtain the optimal process parameter scheme.

[0016] Compared with the prior art, the beneficial effects of the present invention are: (1) By acquiring raw material characteristic data and reaction condition parameters, this invention provides accurate basic data for process optimization, which helps to accurately grasp the dynamic changes of the reaction process and avoid process efficiency instability caused by raw material fluctuations or environmental changes. Furthermore, by calculating the sensitivity coefficient of each process parameter through reaction kinetic simulation, the influence of each parameter on the reaction efficiency can be quantified, which provides a quantitative basis for subsequent parameter optimization, effectively avoiding the situation of blindly adjusting parameters and improving the stability and efficiency of the process.

[0017] (2) By optimizing the process path of the reaction process in multiple stages, the present invention can take the most appropriate optimization measures for the reaction characteristics of each stage, which not only improves the reaction efficiency of each stage, but also reduces unnecessary energy consumption and raw material waste in the reaction process. The process parameter adjustment space is divided into multiple independently optimizable adjustment sub-regions, and sensitivity analysis and optimization are performed for different regions. This ensures that the process parameters in each region are adjusted most effectively, avoids mutual interference and coupling effects between parameters, and improves the controllability and optimization space of the process.

[0018] (3) By combining raw material characteristic data and a preset influence correction model, the present invention corrects the reaction efficiency, which can more accurately reflect the raw material variations and environmental fluctuations that may be encountered in actual production, avoid deviations under standard experimental conditions, and make the optimization scheme more feasible in actual production.

[0019] (4) By using a non-dominated sorting genetic algorithm for multi-objective optimization, this invention can balance multiple objectives and ensure that the optimization scheme achieves the best performance in many aspects, thereby further improving the economy and sustainability of the process. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of the overall method in one embodiment of the present invention; Figure 2 This is a schematic diagram of the overall system architecture in one embodiment of the present invention.

[0021] In the diagram: 1. Parameter acquisition unit; 2. Parameter coupling unit; 3. Range division unit; 4. Efficiency analysis unit; 5. Parameter optimization unit; 6. Process optimization unit. Detailed Implementation

[0022] The technical solutions of 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.

[0023] Example 1, please refer to Figure 1 This invention provides a technical solution: an optimized preparation process method for an organic descaling agent, comprising: S1. Obtain the raw material characteristic data, formulation parameters and reaction condition parameters of the preparation process to be optimized; wherein, the formulation parameters include several reaction stages, the optimal process path of each reaction stage and the preset control nodes of the corresponding optimal process path, and the preset control nodes are set as reaction inflection points. S2. Based on raw material characteristic data and reaction condition parameters, the parameter sensitivity distribution of the preparation process is calculated through reaction kinetics simulation, and the influence range of process parameters is determined according to the parameter coupling model; S3. Based on the influence range of process parameters, the parameter adjustment space of the preparation process is divided into multiple interconnected and independently optimizable adjustment sub-regions. Each adjustment sub-region includes process parameters, including but not limited to reaction temperature, stirring rate, and raw material ratio. S4. The execution order of each reaction stage is used as the first optimization parameter, and the core process parameters of each regulation sub-region are used as the second optimization parameter. Based on the first optimization parameter, the second optimization parameter, and the pre-trained reaction simulation model, the initial reaction efficiency of each preset control node is determined. The reaction simulation model is a mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions. The initial reaction efficiency is corrected based on the raw material characteristic data using the pre-trained influence correction model. S5. According to the preset parameter adjustment order, perform sensitivity analysis and optimization calculation on the process parameters of multiple adjustment sub-regions to obtain the sub-region optimization parameter set; perform coupling verification on the sub-region optimization parameter set to generate a global parameter optimization matrix; S6. Based on the corrected reaction efficiency, the first optimized parameter, and the second optimized parameter, determine the first preparation time and the first raw material consumption as optimization objectives, and construct a multi-objective optimization problem by combining the global parameter optimization matrix; solve the multi-objective optimization problem by using a multi-objective optimization algorithm to obtain the optimal process parameter scheme.

[0024] It should be noted that detailed data collection is conducted on the raw materials, formulation, and reaction conditions used in the production process. Raw material characteristic data includes the physicochemical properties of all reactants, such as purity, reactivity, and solubility. For example, in organic descaling agents, raw materials such as organic acids and surfactants may be used, and the characteristics of these raw materials directly affect the quality of the final product and the reaction efficiency. Formulation parameters involve each stage and process route in the reaction process; this refers to the process flow and steps that need to be followed at different stages of production. For example, there may be several reaction stages, such as a pretreatment stage, a reaction synthesis stage, and a post-treatment stage. Reaction condition parameters include temperature, stirring rate, reaction time, and pressure. Different conditions affect the descaling agent's... The reaction efficiency and the properties of the final product are directly affected. By using a reaction kinetic model, combined with raw material characteristics and reaction condition parameters, reaction kinetic simulations are performed to obtain the sensitivity of reaction efficiency to various parameters under different process conditions. This helps to understand which process parameters have the greatest impact on the final product; for example, reaction temperature may have a greater impact on the reaction rate, while stirring rate may have a smaller impact. Through simulation calculations, sensitivity maps can be plotted to show the impact of different parameters on efficiency at different reaction stages. In actual processes, process parameters may influence each other; for example, reaction temperature and stirring rate may have a coupled relationship, therefore, the combined effect of these parameters needs to be considered comprehensively. Based on the calculated influence range of each process parameter, the entire process can be analyzed. The process control space is divided into multiple independently optimizable sub-regions; for example, reaction temperature may be divided into low-temperature, medium-temperature, and high-temperature zones; stirring rate may be divided into low-speed, medium-speed, and high-speed zones; each sub-region contains a specific set of process parameters; for example, in the low-temperature zone, the reaction temperature is between 30°C and 50°C; during process optimization, a first optimization parameter and a second optimization parameter are set: the first optimization parameter refers to the execution order of different reaction stages, such as whether heating is performed first, followed by stirring, or vice versa; the second optimization parameter refers to the core process parameters of each control sub-region, such as reaction temperature, stirring rate, and raw material ratio; at this stage, a pre-trained reaction simulation model is used to estimate the reaction effect. The model already includes the mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions. For example, assuming that at a certain reaction stage, the reaction temperature is 60°C, the stirring rate is 200 rpm, and the raw material ratio is 1:1, the initial reaction efficiency is 85% according to the simulation model. Then, the influence correction model is used, combined with actual raw material characteristic data, to correct the reaction efficiency. If a certain raw material has a high moisture content, it may affect the reaction efficiency. In this case, the initial reaction efficiency is adjusted through the correction model. Sensitivity analysis is performed on multiple sub-regions to identify which process parameters have the greatest impact on the reaction efficiency, and then optimization calculations are performed. The purpose of optimization is to improve reaction efficiency, reduce raw material consumption, and shorten reaction time.For example, in the temperature regulation sub-region, it might be observed that increasing the reaction temperature from 50°C to 70°C significantly improves reaction efficiency, but increases raw material consumption. Therefore, the optimization algorithm comprehensively considers both reaction efficiency and raw material consumption to provide the optimal temperature regulation range. After optimizing the parameters of each sub-region, they are coupled and verified to ensure that the optimized parameters of each sub-region can work collaboratively. Finally, a global parameter optimization matrix is ​​generated, which covers all optimized parameters throughout the entire production process. Based on the corrected reaction efficiency and optimized parameters, optimization objectives are defined, such as: first preparation time: minimize production time; first raw material consumption: minimize raw material usage and reduce costs. By constructing a multi-objective optimization problem, multi-objective optimization algorithms such as genetic algorithms and particle swarm optimization are used to solve it, obtaining the optimal process parameter scheme to satisfy reaction efficiency, time, and cost. Multiple objectives are considered; for example, suppose an organic descaling agent is being produced, and the reaction process includes two stages: Stage 1: raw material mixing and initial reaction; Stage 2: heating to a specific temperature and continuing the reaction; The optimization objective is to maximize reaction efficiency (i.e., achieve the best descaling effect) and minimize raw material consumption; Through reaction kinetic simulation, it was found that reaction temperature has a significant impact on reaction efficiency, with efficiency increasing by 10% for every 10°C increase in temperature; However, excessively high temperatures lead to increased raw material consumption; Therefore, an optimal reaction temperature range, such as 60°C to 70°C, is obtained through optimization calculations; With a higher stirring rate, the reaction time can be shortened, but this increases energy consumption, thus requiring a balance; Finally, through a multi-objective optimization algorithm, a set of optimal reaction temperature, stirring rate, and reaction time can be obtained, maximizing reaction efficiency while minimizing raw material consumption.

[0025] In an optional embodiment, based on raw material characteristic data and reaction condition parameters, the parameter sensitivity distribution of the preparation process is calculated through reaction kinetic simulation, and the influence range of process parameters is determined according to a parameter coupling model, including: Input the content and purity of active ingredients from the raw material characteristic data, and the initial temperature and pressure from the reaction condition parameters into the reaction kinetics simulation model; The effect of a single parameter change on the reaction rate was simulated using the controlled variable method, and the sensitivity coefficients of each process parameter were calculated. The interaction between parameters is analyzed based on the parameter coupling model. The influence range of each process parameter is determined according to the influence boundary of the parameter whose sensitivity coefficient is greater than the preset threshold.

[0026] It should be noted that the active ingredient content refers to the proportion of reactive components in the raw material; for example, in the production process of organic descaling agents, the raw material may contain different concentrations of organic acids, which have different activities; purity refers to the purity of the effective components contained in the raw material; for example, if the purity of the organic acid in the raw material is 90%, it means that its reactivity is lower than that of a raw material with 100% purity; reaction condition parameters: the temperature at the start of the reaction; the reaction rate usually increases with increasing temperature, which is predicted according to the Arrhenius equation; the gas pressure during the reaction also affects the reaction rate, especially in gas-liquid reactions, where pressure usually accelerates gas dissolution, thereby increasing the reaction rate; after inputting these data into the reaction kinetics simulation model, The model predicts reaction rates based on known reaction mechanisms and provides a basis for further optimization. The controlled variable method involves changing a single process parameter while keeping other parameters constant, observing the effect of that parameter change on the reaction rate. This helps understand the independent influence of each parameter on the reaction rate. For example, reaction temperature: assuming the initial temperature is increased from 30°C to 60°C while keeping other parameters constant, simulation using the controlled variable method shows that the reaction rate increases with increasing temperature. This indicates that temperature has a significant impact on the reaction rate. If the purity of the raw materials is increased from 80% to 95% while keeping other conditions constant, the simulation results may show an increase in the reaction rate because the purer raw materials provide more active ingredients. One method can be used to determine the impact of each process parameter on the reaction rate and calculate the sensitivity coefficient. The sensitivity coefficient represents the degree of influence of a parameter change on the reaction rate, and is usually defined as the ratio of the change in reaction rate to the change in the parameter. For example, if the sensitivity coefficient for reaction temperature is 2, it means that for every 1°C increase in temperature, the reaction rate will increase by 2 times; if the sensitivity coefficient for purity is 1.5, it means that for every 1% increase in raw material purity, the reaction rate will increase by 1.5 times. Some process parameters may not affect the reaction rate independently; there may be interactions between them. Parameter coupling models help analyze the effects of these interactions, usually represented by a coupling matrix; for example, reaction temperature and stirring rate may be coupled. Increasing temperature may accelerate the reaction, but if the stirring rate is too low, the contact efficiency of the reactants will be low, and the reaction rate will be limited. The interaction between temperature and pressure may also affect the reaction rate. High temperature and high pressure may jointly promote the reaction, but if the pressure is too high, the reaction system may become unstable. By analyzing the coupled model, the optimal process conditions under different parameter combinations can be found. Based on the sensitivity coefficient, the influence boundary of each process parameter on the reaction rate can be determined. For example, if the sensitivity coefficient is greater than a certain preset threshold (e.g., 2), it indicates that temperature has a significant impact on the reaction rate, and the temperature adjustment range should be between 50°C and 80°C to ensure that the reaction rate is maximized. If the sensitivity coefficient of pressure is small (e.g., 0), the reaction rate will be significantly affected.5) If the pressure is below the threshold, it indicates that the pressure has a relatively small impact on the reaction rate, and excessive optimization is unnecessary, or it can be fixed within a reasonable range. Based on these analyses, optimization ranges for each process parameter can be set to avoid unnecessary costs or risks caused by over-adjustment.

[0027] In an optional embodiment, based on the influence range of process parameters, the parameter adjustment space of the preparation process is divided into multiple interconnected and independently optimizable adjustment sub-regions, including: The initial region is divided according to the reaction process sequence, with the parameter baseline value of the reaction initiation stage as the starting point. Adjust the initial region boundary according to the influence range of each process parameter to ensure that the parameter influence of adjacent regions does not overlap; Correlation analysis is performed on the process parameters in each region, and strongly correlated parameters are grouped into the same adjustment sub-region, resulting in multiple adjustment sub-regions.

[0028] It should be noted that the reaction process can usually be divided into multiple stages, each with different process parameters. The reaction start-up stage refers to the period from the beginning of the reaction to its initial stabilization, usually accompanied by changes in temperature, pressure, and reactant concentration. These changes can be divided into different regions according to their time sequence, each region corresponding to a specific range of process parameters. The division of the initial regions is based on the baseline parameter values ​​of the reaction start-up stage. For example, based on the time sequence of the reaction process, the reaction may be divided into the following stages and different regions: Region 1: Reaction start-up stage (0-30 minutes), temperature and pressure are near the baseline values; Region 2: Reaction heating stage (30 minutes-1 hour), temperature gradually rises, pressure... The temperature begins to rise; Region 3: Stabilization phase (after 1 hour), where temperature and pressure reach the expected stable values; During the reaction, different process parameters (such as temperature, pressure, concentration, etc.) will affect the reaction rate and the formation of reaction products; To ensure that the effects between different regions do not overlap, the boundaries of the initial regions need to be adjusted according to the influence range of each process parameter; The influence range refers to the effective range of change of each process parameter during the reaction process. Parameter changes outside this range may not have a significant impact on the reaction; For example: Suppose that when the reaction temperature changes from 30°C to 80°C, the reaction rate changes significantly, but if the temperature exceeds 80°C, the change in reaction rate tends to stabilize, at which point temperature no longer affects the reaction rate. The reaction has a significant impact; at low pressure, the reaction rate is slow; at high pressure, the reaction rate is fast, but beyond a certain pressure, the influence of pressure on the reaction rate gradually decreases. Therefore, the influence range of each parameter during the reaction process is different, and the boundaries of each region must be adjusted according to these ranges to ensure that changes in process parameters within each region do not overlap with adjacent regions. For example, when the temperature changes from 30°C to 80°C, it can be divided into region 1 (30-50°C), region 2 (50-65°C), and region 3 (65-80°C), and the boundaries of these regions should avoid overlapping. During the reaction process, different process parameters may have certain correlations; for example, temperature and reactant concentration may simultaneously affect the reaction rate. The reaction rate, or pressure and stirring rate together, determines the efficiency of the reaction. To simplify process control, correlation analysis can be performed on the process parameters in each region, grouping those parameters with strong correlations into the same control sub-region. A control sub-region refers to a set of process parameters that have significant interactions, and the reaction can be optimized by adjusting these parameters together. Through correlation analysis, it can be determined which parameters need to be adjusted together. For example: Control sub-region 1: temperature and reactant concentration; these two parameters jointly affect the reaction rate, so they can be adjusted as a sub-region; Control sub-region 2: pressure and stirring rate; there may be a strong coupling relationship between pressure and stirring rate, which can be considered together during adjustment.

[0029] In an optional embodiment, sensitivity analysis and optimization calculations are performed on the process parameters of multiple adjustment sub-regions to obtain a set of optimized parameters for each sub-region, including: The preset parameter adjustment order is determined based on the parameter importance ranking, with priority given to optimizing process parameters with high sensitivity coefficients; For each regulation sub-region, the first priority parameter for this round of optimization is determined, and multi-level testing is conducted using response surface methodology to obtain the parameter effect curve; The optimal range of values ​​for the first priority parameter is determined based on the parameter effect curve, forming an optimized subset of the first parameter; Determine the second priority parameters sequentially according to the adjustment order, repeat the sensitivity test and optimization process, and obtain the optimized subset of the second parameters; Integrate the optimization subsets of each parameter to generate a sub-region optimization parameter set.

[0030] It is important to note that it is necessary to analyze the importance of each process parameter. This process typically involves sensitivity analysis to determine the impact of each parameter on the reaction process. Parameters with high sensitivity coefficients have a greater impact on the reaction rate or product quality and are therefore prioritized for optimization. The sensitivity coefficient indicates the degree to which a parameter change affects the reaction result; a higher sensitivity coefficient means a greater impact of the parameter on the final result, and vice versa. For example, assuming the following process parameters: temperature: sensitivity coefficient 0.8; pressure: sensitivity coefficient 0.6; reactant concentration: sensitivity coefficient 0.4, based on the magnitude of the sensitivity coefficients, temperature is listed as the highest priority parameter for optimization, followed by pressure, and lastly reactant concentration. Within each control sub-region, the first... First, determine the primary parameter to be optimized. To determine the optimal value, response surface methodology (RSM) can be used. This method can establish a mathematical model between the reaction and the parameters using experimental data, thereby predicting the impact of parameter changes on the reaction. Experimental data is obtained by conducting multi-level tests on the parameters, and these data are used to plot parameter-effect curves. These curves show how parameter changes affect the reaction output, such as reaction rate, product yield, or quality. For example, when optimizing temperature, different temperature values ​​(e.g., 30°C, 40°C, 50°C) are set for testing, and the corresponding product amounts are recorded. Then, through data analysis, a temperature-effect curve is obtained, showing the trend of product amount changes at different temperatures. The parameter effect curve obtained through response surface methodology helps identify the optimal range of values ​​for a parameter; this range refers to the value at which the reaction effect is best. For example, from the temperature effect curve, it may be found that the product yield is highest between 45°C and 60°C; therefore, this temperature range is determined as the optimal temperature range, forming the first parameter optimization subset. Next, according to a preset adjustment sequence, the second priority parameter for optimization is determined, and the sensitivity test and optimization process is repeated. This process is similar to optimizing the first priority parameter, but this time the focus is on the second priority parameter; for example, when optimizing pressure, assuming that response surface methodology shows that the reaction rate increases fastest between 1 atm and 2 atm, the reaction rate is higher than... After 2 atm, the reaction rate change becomes smaller; at this point, the optimal pressure range is determined to be 1 atm to 2 atm, forming the second parameter optimization subset; finally, the optimization subsets of each sub-region are integrated to form a sub-region optimization parameter set; this set contains the optimized process parameters and their optimal values ​​in each regulation sub-region; for example, assuming that temperature and pressure are optimized in the first sub-region, the optimal parameters obtained are: temperature: 45°C-60°C; pressure: 1 atm-2 atm; in the second sub-region, after optimization, the parameters are: temperature: 50°C-65°C; pressure: 2 atm-3 atm; through these optimization results, the final sub-region optimization parameter set can be generated to provide guidance for actual production.

[0031] In an optional embodiment, coupled verification of the sub-region optimization parameter set is performed to generate a global parameter optimization matrix, including: Establish parameter correlation matrices between sub-regions to identify key parameters with cross-regional impact; Substitute the optimized parameter sets of each sub-region into the full-process reaction simulation to detect efficiency fluctuations caused by parameter coupling; The parameter combinations that exceed the fluctuation limit are corrected, and a global parameter optimization matrix including parameter coordination coefficients is constructed based on the correction results.

[0032] It should be noted that during the optimization process across multiple sub-regions, the optimal parameter set for each sub-region is obtained. However, the parameters between these sub-regions are often not independent; some parameters may influence each other in different sub-regions. To identify these influence relationships, a parameter correlation matrix needs to be established. The parameter correlation matrix is ​​a two-dimensional matrix where each element represents the degree of correlation or influence between parameters in two sub-regions. This can be obtained through statistical analysis, sensitivity analysis, and other methods. Key parameters refer to those parameters that repeatedly appear across multiple sub-regions and have a strong influence. The coupling of these parameters between different sub-regions may affect the entire process flow. For example, suppose there are two sub-regions, A and B, involving temperature and pressure, respectively. Parameters such as reactant concentration are considered. When establishing the parameter correlation matrix, it may be found that temperature has a significant impact on pressure and concentration when optimizing sub-region A. This relationship can be reflected in the matrix. In this matrix, temperature has a stronger impact on sub-region A, while concentration has a stronger impact on sub-region B. Through this matrix analysis, key parameters with cross-regional influence can be identified. The optimized parameters of all sub-regions are then substituted into a full-process reaction simulation model for simulation calculation. The purpose of this step is to detect whether these optimized parameters cause efficiency fluctuations or instability in the overall process. A full-process reaction simulation is usually a mathematical or computer model that can simulate the entire process from reactants to products. During the simulation, the effects of different parameter combinations on... The impact on system efficiency (such as output, reaction time, energy consumption, etc.); efficiency fluctuations refer to instability or performance fluctuations during the reaction process. For example, certain parameter combinations may lead to irregular fluctuations in reaction rate or unstable product quality. Suppose that during simulation, the temperature and pressure combination in sub-regions A and B causes significant fluctuations in reaction rate, potentially resulting in short-term excessively high or low output. Such fluctuations can be calculated using the model. When certain parameter combinations are found to cause excessive efficiency fluctuations or system instability, these fluctuations need to be corrected. This step involves adjusting the parameter value ranges and optimizing their synergistic effects to solve the problem. By adjusting key parameters, the coupling between them is improved. A more ideal combined effect can be achieved, thereby reducing efficiency fluctuations. The correction process can rely on numerical optimization algorithms, empirical data, or experimental results. The purpose of correction is to ensure that all parameters can work synergistically throughout the entire process, avoiding the impact of optimizing parameters in one sub-region on the overall efficiency of other regions. For example, if certain combinations of temperature and pressure lead to unstable yields, the temperature range can be adjusted, or the reactant concentration can be adjusted simultaneously with pressure changes to balance this fluctuation. After correcting for efficiency fluctuations, a global parameter optimization matrix can be constructed. This matrix not only contains the optimal value of each parameter but also the parameter synergy coefficient. This matrix is ​​used to describe the synergistic effect between various parameters, helping to find the optimal parameter combination in the overall process flow.The parameter synergy coefficient refers to the degree of mutual influence between two parameters when used together. A higher synergy coefficient indicates a closer cooperation between the two parameters, resulting in better optimization of the reaction effect. The final global parameter optimization matrix provides the optimal parameter configuration for each sub-region and the entire process. For example, assuming the corrected results show a synergy coefficient of 0.95 for temperature and pressure, and 0.85 for temperature and concentration, this can be represented in the global parameter optimization matrix to account for the interaction between these parameters in future optimizations.

[0033] In an optional embodiment, the reaction efficiency is corrected based on a pre-trained impact correction model combined with raw material characteristic data, including: Impurity content, moisture content, and active ingredient stability parameters at each reaction stage are extracted from raw material characteristic data as correction features. The initial reaction efficiency and correction features are input into the influence correction model, which includes component feature gating and raw material perturbation gating. The influence of raw material components on the reaction is extracted by gating based on component characteristics, and the perturbation coefficient of historical data deviation is calculated by gating based on raw material perturbation. The initial reaction efficiency is corrected based on the perturbation coefficient and feature fusion results, and the corrected reaction efficiency is obtained.

[0034] It should be noted that some key features need to be extracted from the raw material characteristic data. These features will be used for subsequent reaction efficiency correction. Common features include: impurity content: the proportion of impurities in the raw materials; impurities may affect the reaction rate, product quality, etc., so they need special attention; moisture content: moisture affects the heat effect and solubility of the reaction, which may lead to instability in reaction efficiency; stability of active ingredients: the impact of the stability of active ingredients (such as catalysts and reactants) on the reaction process; degradation or instability of active ingredients will lead to a decrease in reaction efficiency; changes in these features at each stage of the reaction may affect the reaction efficiency, so they need to be input as correction features into the subsequent model. For example, suppose a chemical synthesis reaction is underway, and the raw materials... The moisture content, impurity ratio, and stability of active ingredients (such as catalyst stability) of component A will vary at different stages. Data on these characteristics can be recorded at the initial, intermediate, and final stages of the reaction. Next, the initial reaction efficiency and correction features are input into the influence correction model, which aims to correct the reaction efficiency based on the input data. The specific effects of these correction features are achieved through two gating mechanisms: component feature gating is used to extract the influence of raw material components on the reaction. These features include impurity content, moisture content, and the stability of active ingredients, which affect the reaction rate, product quality, and overall reaction efficiency. Through the gating mechanism, the model can identify which component features have the greatest impact on the reaction and prioritize these features. Raw material perturbation gating is used to calculate deviations in historical data, generating a perturbation coefficient. This coefficient represents the difference between the actual raw material condition and historical data. For example, in past experiments, fluctuations in moisture content might have affected reaction stability; therefore, the perturbation coefficient tells us how this fluctuation affects reaction efficiency. For instance, suppose a batch of raw materials has significant fluctuations in impurity and moisture content. Component feature gating will identify the potential impact of these fluctuations on reaction efficiency. Historical data shows that when moisture content fluctuates significantly, reaction efficiency decreases by 5%. Therefore, raw material perturbation gating will calculate a perturbation coefficient (e.g., 0.95) representing the decrease in reaction efficiency caused by this fluctuation. The core task of component feature gating is to analyze and extract raw material components (impurities, moisture, etc.). The influence of raw material components (such as moisture content and stability of active ingredients) on the reaction process; through model learning, component feature gating can determine which raw material components have the greatest impact on reaction efficiency and extract these impact features; for example, if moisture content has a significant impact on the reaction rate, component feature gating will identify this and input it as an important feature into subsequent correction steps; perturbation gating calculates the deviation of historical data and generates a perturbation coefficient; this coefficient is used to measure the difference between the actual raw material characteristics and historical data, helping to understand the impact of deviation on reaction efficiency; for example, assuming historical data shows that the reaction efficiency is optimal at a moisture content of 10%, but the current moisture content is 15%; in this case, perturbation gating will calculate a perturbation coefficient, such as 0.9 indicates that the reaction efficiency decreased by 10% due to the change in moisture content. Finally, the influence characteristics of the perturbation coefficient and the component feature-gated extraction are fused and applied to correct the reaction efficiency. Through this process, the initial reaction efficiency (based on the preliminary characteristics of the raw materials) is corrected, resulting in a corrected reaction efficiency that more accurately reflects the actual situation. For example, assuming the initial reaction efficiency is 90%, the moisture perturbation coefficient is 0.9, and the influence characteristics of moisture content indicate that higher moisture content will decrease the efficiency by 5%, then the corrected reaction efficiency can be calculated as follows: Corrected efficiency = Initial efficiency × Perturbation coefficient × Influence characteristics; Corrected efficiency = 90% × 0.9 × (1 - 0.05) = 90% × 0.9 × 0.95 = 76.65%.

[0035] In an optional embodiment, the method for training the reaction simulation model includes: Several process parameters related to the reaction efficiency mapping of organic descaling agents were determined; whereby the reaction efficiency is the conversion rate of raw materials into the target product per unit time under standard raw material conditions; Based on all process parameters, an initial simulation model is established to characterize the mapping relationship between process parameters and reaction efficiency; Parameter sensitivity analysis was performed on the initial simulation model to obtain the weighting factors between process parameters and reaction efficiency; Based on the weighting factor, a pre-defined simulation model is used to determine the mapping relationship between the characterization process parameters, the weighting factor and the reaction efficiency. The target data for the weighting factors were determined using a data fitting method, and the reaction simulation model under standard raw material conditions was determined based on the target data. The actual data of process parameters under standard raw material conditions were input into the reaction simulation model to verify the deviation between the predicted and actual values ​​of the reaction efficiency output by the model.

[0036] It is important to note that it is necessary to identify the process parameters related to reaction efficiency. For organic descaling agent reactions, for example, suppose temperature, pH, reaction time, and concentration are selected as the main process parameters in the experiment. Data on these process parameters is collected and used to build an initial simulation model. This initial model describes the relationship between each process parameter and reaction efficiency. For example, regression analysis, neural networks, and other methods can be used to build the initial simulation model. This initial simulation model provides a basic framework for predicting reaction efficiency under different process conditions. Parameter sensitivity analysis is performed on the initial simulation model to determine the degree of influence of each process parameter on reaction efficiency. Sensitivity analysis helps to understand which parameters have the greatest impact on reaction efficiency, thereby optimizing the process design. For example, by making small-scale changes to different process parameters, the changes in reaction efficiency are observed. By calculating the weight of each parameter's influence on reaction efficiency, weighting factors can be obtained. These weighting factors represent the relative importance of each parameter. If the change in temperature has the greatest impact on reaction efficiency, the weighting factor will be higher; if the change in reaction time has a smaller impact on efficiency, the weighting factor will be lower. Based on the weighting factors and the initial simulation model, a pre-defined simulation model can be constructed. This model considers the influence weight of each process parameter and... The reaction efficiency prediction is optimized by adjusting parameter weights; a data fitting method is used to determine the target data, i.e., the expected reaction efficiency under the combination of process parameters; these target data may come from experiments or historical data; by fitting these data, the model parameters are adjusted so that the simulation model can accurately reflect the actual reaction situation; for example, a set of process conditions (temperature=60℃, pH=4, reaction time=2 hours, concentration=1M) is obtained through experiments, and the target reaction efficiency is 85%; these data are fitted to the model, and the weight factors and functional relationships are adjusted until the reaction efficiency output by the model is as close as possible to 85%; finally, the actual data of the process parameters under standard raw material conditions are used. Based on the input reaction simulation model, verify the deviation between the model's predicted reaction efficiency and the actual value. If the prediction error is small, the model is effective; if the error is large, the model needs to be adjusted or the data fitting needs to be repeated. For example, input the process conditions used in actual operation, such as temperature = 65℃, pH = 4.5, reaction time = 2.5 hours, and concentration = 1.1M. The model outputs a reaction efficiency of 83%. If the actual measured reaction efficiency is also 83%, the model prediction is accurate. If the deviation is large, it may be necessary to: adjust the parameters in the model (e.g., change the weighting factors of certain process parameters); optimize the data fitting process to ensure the accuracy of the target data used.

[0037] In an optional embodiment, a multi-objective optimization problem is constructed by incorporating a global parameter optimization matrix, including: The objective function is set with the optimization objectives of minimizing the first preparation time and minimizing the first raw material consumption; Based on the parameter coordination coefficients in the global parameter optimization matrix, the value constraints and interaction constraints of each process parameter are determined. The reaction temperature range, raw material ratio limits, and product purity standards are treated as hard constraints and incorporated into the constraints of the multi-objective optimization problem.

[0038] It should be noted that two objective functions need to be set for the optimization problem: one to minimize the preparation time and the other to minimize raw material consumption. In multi-objective optimization problems, the objective functions are usually independent and need to be optimized simultaneously. The objective of minimizing the preparation time is to shorten the total time of the reaction process as much as possible. This time may be affected by multiple factors, such as reaction temperature, catalyst concentration, and stirring rate. The objective of minimizing raw material consumption is to reduce the amount of raw materials consumed in the reaction process to reduce production costs or improve resource utilization efficiency. These two objective functions usually need to be balanced, that is, one objective cannot be focused on only, because there may be trade-offs. For example, shortening the reaction time may require increasing the temperature or concentration, which may lead to increased raw material consumption. In multi-objective optimization, the relationships between different process parameters are very complex. It is necessary to understand the synergistic effects or interactions between the parameters in order to reasonably set value constraints and interaction constraints. During the reaction process, the process parameters may affect each other, and some parameters may affect each other. Changes in parameters can amplify or diminish their effects on other parameters. For example, temperature and reaction time may be positively correlated (higher temperatures allow for shorter reaction times), while reaction time and feed concentration may be negatively correlated (longer reaction times and higher feed concentrations may lead to unnecessary consumption). To handle these interrelationships, a synergy coefficient matrix can be used. For example, the elements of the matrix can represent the synergistic effect between process parameters i and j (positive values ​​indicate a positive correlation, and negative values ​​indicate a negative correlation). Based on the synergy coefficient matrix, the range of values ​​for each process parameter can be determined. For example, some process parameters may not be allowed to take values ​​within a certain range, or their interactions may require specific ratios. For example, there may be an interaction constraint between temperature and feed concentration, meaning that when the temperature exceeds a certain value, the feed concentration must be within a certain range. For example, assuming a positive correlation between temperature and reaction time, higher temperatures allow for shorter reaction times. A constraint can be set: temperature T∈[50,90]. ℃; reaction time t∈[10,30]min; interaction constraint between temperature and reaction time: T−t≤60 (i.e., when the temperature is high, the reaction time must be short); hard constraints refer to constraints that must be strictly satisfied in optimization problems; reaction temperature range, raw material ratio limit, and product purity standard are hard constraints, which must be strictly followed in the optimization process; reaction temperature directly affects the reaction rate and product selectivity, so a certain temperature range must be set; the ratio of different reactants must be controlled within a certain range, otherwise it may lead to the generation of by-products or a decrease in reaction efficiency; product purity is one of the key indicators of whether the reaction is successful; usually, purity requirements are used as one of the hard constraints; product purity ≥90%; these hard constraints must be strictly satisfied, otherwise the result will be considered unqualified; in multi-objective optimization, it is usually necessary to use some optimization algorithms (such as genetic algorithms, particle swarm optimization, etc.) to simultaneously optimize the objective function and constraints;A common approach is to find the optimal solution by balancing two objectives (minimizing preparation time and minimizing raw material consumption) while satisfying hard constraints.

[0039] In an optional embodiment, the optimal process parameter scheme is obtained by solving a multi-objective optimization problem using a multi-objective optimization algorithm, including: A non-dominated sorting genetic algorithm is used to solve the multi-objective optimization problem, and the population includes various combinations of process parameters. The objective function value of each individual in the population is calculated based on the global parameter optimization matrix, and non-dominated sorting and crowding calculation are performed. The next generation population is generated through crossover and mutation operations, while retaining the Pareto optimal solution; After iterating to a preset number of times, the parameter combination with the best overall performance is selected from the final Pareto optimal solution set as the optimal process parameter scheme.

[0040] It's important to note that initializing the population is the first step in multi-objective optimization algorithms. The population consists of multiple individuals, each representing a combination of process parameters. For example, suppose the objective is to optimize a chemical reaction process with three process parameters to optimize: temperature, reaction time, and feed concentration. These parameters are randomly generated within a certain range and form an initial population. Based on the global parameter optimization matrix, the objective function value for each individual in the population is calculated. The objective function is usually multi-objective, such as minimizing reaction time or minimizing feed consumption. Each individual calculates its corresponding objective function value based on its process parameters. These objective functions often conflict: for example, shortening the reaction time may lead to higher feed consumption. Therefore, the optimization objective is to find an equilibrium point. In multi-objective optimization, the "dominance relationship" of individuals is a crucial concept. An individual dominates another individual if and only if that individual performs better than the other individual on all objectives, or is better on some objectives and not worse on others. A non-dominated individual is one that no other individual outperforms it on all objectives. The set of all non-dominated individuals is called the Pareto optimal set. Crowding is used to measure... The "sparseness" of an individual in the Pareto optimal solution set; the difference in crowding among individuals determines the priority of individual selection; in Pareto optimal solutions, individuals with lower crowding are selected first to ensure the diversity of the solution set; for example, individual A is better than individual B in both objective 1 and objective 2, therefore individual A dominates individual B; if neither individual A nor individual B is dominated by other individuals, they belong to the same Pareto optimal solution set; crossover and mutation are the core operations of genetic algorithms, used to generate the next generation population. Two parent individuals are selected from the current population, and one or more offspring individuals are generated by exchanging their genes (process parameter combinations); for example: parent 1 (T=70∘C, t=15min, C_raw material=1.0M) and parent 2 (T=80∘C, t=20min, C_raw material=1.5M), and a offspring individual (T=75∘C, t=17min, C_raw material=1.25M) is generated by crossover; some process parameters of the individuals are slightly and randomly varied to increase the diversity of the population; for example: individual (T=70∘C, t=15min, C_raw material=1.0M).The genetic algorithm (GAL) may use a mutation operation to change the temperature to T=72°C to explore new solution spaces. After crossover and mutation, the next generation of the population is obtained. Individuals in the population are evaluated using non-dominated sorting and crowding calculation, and the Pareto optimal solution is selected. The genetic algorithm iteratively updates the population by repeating the above process (crossover, mutation, selection) until a preset maximum number of iterations is reached. For example, assuming a maximum number of iterations of 50, after 50 generations of evolution, a solution set containing multiple Pareto optimal solutions is obtained, each solution representing... A balance between reaction time and raw material consumption is achieved. After multiple iterations, a Pareto optimal solution set is obtained. At this point, an optimal combination of process parameters needs to be selected based on actual requirements. This combination can be selected through comprehensive performance evaluation, considering factors such as reaction time, raw material consumption, and cost. For example, among the multiple solutions in the Pareto optimal solution set, there might be a solution with T=75°C, t=18min, and Craw material=1.2M. This solution achieves a good balance between reaction time and raw material consumption while meeting cost requirements.

[0041] Example 2, please refer to Figure 2 This invention provides a technical solution: a system for optimizing the preparation process of an organic descaling agent, applicable to the aforementioned method for optimizing the preparation process of an organic descaling agent, comprising: The parameter acquisition unit 1 is used to acquire the raw material characteristic data, formulation parameters and reaction condition parameters of the preparation process to be optimized; wherein, the formulation parameters include several reaction stages, the optimal process path of each reaction stage and the preset control nodes of the corresponding optimal process path, and the preset control nodes are set as reaction inflection points. Parameter coupling unit 2 is used to calculate the parameter sensitivity distribution of the preparation process based on raw material characteristic data and reaction condition parameters through reaction kinetic simulation, and to determine the influence range of process parameters according to the parameter coupling model; Range division unit 3 is used to divide the parameter adjustment space of the preparation process into multiple connected and independently optimizable adjustment sub-regions according to the influence range of the process parameters. Each adjustment sub-region includes the process parameters. Efficiency analysis unit 4 is used to take the execution order of each reaction stage as the first optimization parameter and the core process parameters of each regulation sub-region as the second optimization parameter; based on the first optimization parameter, the second optimization parameter and the pre-trained reaction simulation model, the initial reaction efficiency of each preset control node is determined, wherein the reaction simulation model is the mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions; the initial reaction efficiency is corrected based on the pre-trained influence correction model using raw material characteristic data. The parameter optimization unit 5 is used to perform sensitivity analysis and optimization calculations on the process parameters of multiple adjustment sub-regions according to a preset parameter adjustment order, so as to obtain the sub-region optimization parameter set; and to perform coupling verification on the sub-region optimization parameter set to generate a global parameter optimization matrix. The process optimization unit 6 is used to determine the first preparation time and the first raw material consumption as optimization objectives based on the corrected reaction efficiency, the first optimization parameter and the second optimization parameter, and to construct a multi-objective optimization problem by combining the global parameter optimization matrix; and to solve the multi-objective optimization problem by using a multi-objective optimization algorithm to obtain the optimal process parameter scheme.

[0042] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. An optimized preparation process method for an organic descaling agent, characterized in that, include: Obtain raw material characteristic data, formulation parameters, and reaction condition parameters for the preparation process to be optimized; wherein, the formulation parameters include several reaction stages, the optimal process path for each reaction stage, and the preset control nodes for the corresponding optimal process paths, and the preset control nodes are set as reaction inflection points. Based on the raw material characteristic data and reaction condition parameters, the parameter sensitivity distribution of the preparation process is calculated through reaction kinetics simulation, and the influence range of the process parameters is determined according to the parameter coupling model. Based on the influence range of the process parameters, the parameter adjustment space of the preparation process is divided into multiple interconnected and independently optimizable adjustment sub-regions, each of which includes process parameters. The execution order of each reaction stage is used as the first optimization parameter, and the core process parameters of each regulation sub-region are used as the second optimization parameter. Based on the first optimization parameter, the second optimization parameter, and the pre-trained reaction simulation model, the initial reaction efficiency of each preset control node is determined. The reaction simulation model is a mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions. The initial reaction efficiency is corrected based on the raw material characteristic data using the pre-trained influence correction model. According to the preset parameter adjustment order, sensitivity analysis and optimization calculations are performed on the process parameters of multiple adjustment sub-regions to obtain sub-region optimization parameter sets; the sub-region optimization parameter sets are coupled and verified to generate a global parameter optimization matrix; Based on the corrected reaction efficiency, the first optimized parameter, and the second optimized parameter, the first preparation time and the first raw material consumption are determined as optimization objectives. A multi-objective optimization problem is constructed by combining the global parameter optimization matrix. The multi-objective optimization problem is solved by a multi-objective optimization algorithm to obtain the optimal process parameter scheme.

2. The optimized preparation process of the organic descaling agent according to claim 1, characterized in that, Based on the raw material characteristic data and reaction condition parameters, the parameter sensitivity distribution of the preparation process is calculated through reaction kinetic simulation. The influence range of the process parameters is determined according to the parameter coupling model, including: Input the content and purity of active ingredients from the raw material characteristic data, and the initial temperature and pressure from the reaction condition parameters into the reaction kinetics simulation model; The effect of a single parameter change on the reaction rate was simulated using the controlled variable method, and the sensitivity coefficients of each process parameter were calculated. The interaction between parameters is analyzed based on the parameter coupling model. The influence range of each process parameter is determined according to the parameter influence boundary where the sensitivity coefficient is greater than the preset threshold.

3. The optimized preparation process of the organic descaling agent according to claim 2, characterized in that, Based on the influence range of the process parameters, the parameter adjustment space of the preparation process is divided into multiple interconnected and independently optimizable adjustment sub-regions, including: The initial region is divided according to the reaction process sequence, with the parameter baseline value of the reaction initiation stage as the starting point. Adjust the initial region boundary according to the influence range of each process parameter to ensure that the parameter influence of adjacent regions does not overlap; Correlation analysis is performed on the process parameters in each region, and strongly correlated parameters are grouped into the same adjustment sub-region, resulting in multiple adjustment sub-regions.

4. The optimized preparation process of the organic descaling agent according to claim 3, characterized in that, Sensitivity analysis and optimization calculations were performed on the process parameters of multiple regulation sub-regions to obtain a set of optimized parameters for each sub-region, including: The preset parameter adjustment order is determined based on the importance of the parameters, and process parameters with high sensitivity coefficients are optimized first. For each regulation sub-region, the first priority parameter for this round of optimization is determined, and multi-level testing is conducted using response surface methodology to obtain the parameter effect curve; The optimal range of values ​​for the first priority parameter is determined based on the parameter effect curve, forming an optimized subset of the first parameter; Determine the second priority parameters sequentially according to the adjustment order, repeat the sensitivity test and optimization process, and obtain the optimized subset of the second parameters; Integrate the optimization subsets of each parameter to generate a sub-region optimization parameter set.

5. The optimized preparation process of an organic descaling agent according to claim 4, characterized in that, The optimization parameter set of the sub-region is coupled and verified to generate a global parameter optimization matrix, including: Establish parameter correlation matrices between sub-regions to identify key parameters with cross-regional impact; Substitute the optimized parameter sets of each sub-region into the full-process reaction simulation to detect efficiency fluctuations caused by parameter coupling; The parameter combinations that exceed the fluctuation limit are corrected, and a global parameter optimization matrix including parameter coordination coefficients is constructed based on the correction results.

6. The optimized preparation process of an organic descaling agent according to claim 5, characterized in that, The reaction efficiency is corrected based on a pre-trained impact correction model combined with raw material characteristic data, including: The impurity content, moisture content, and active ingredient stability parameters of each reaction stage are extracted from the raw material characteristic data as correction features. The initial reaction efficiency and the correction features are input into the influence correction model, which includes component feature gating and raw material perturbation gating; The influence of raw material components on the reaction is extracted by gating based on component characteristics, and the perturbation coefficient of historical data deviation is calculated by gating based on raw material perturbation. The initial reaction efficiency is corrected based on the perturbation coefficient and feature fusion results, and the corrected reaction efficiency is obtained.

7. The optimized preparation process of an organic descaling agent according to claim 6, characterized in that, The training method for the reaction simulation model includes: Several process parameters related to the reaction efficiency mapping of organic descaling agents are determined; wherein, the reaction efficiency is the conversion rate of raw materials into target products per unit time under standard raw material conditions; Based on all the process parameters, an initial simulation model is established to characterize the mapping relationship between process parameters and reaction efficiency; A parameter sensitivity analysis was performed on the initial simulation model to obtain the weighting factors between process parameters and reaction efficiency; Based on the weighting factor, a preset simulation model is determined to characterize the mapping relationship between process parameters, weighting factor and reaction efficiency. The target data for the weighting factors are determined using a data fitting method, and a reaction simulation model under standard raw material conditions is determined based on the target data. The actual data of process parameters under standard raw material conditions are input into the reaction simulation model to verify the deviation between the predicted and actual values ​​of the reaction efficiency output by the model.

8. The optimized preparation process of an organic descaling agent according to claim 7, characterized in that, Constructing multi-objective optimization problems by combining global parameter optimization matrices, including: The objective function is set with the optimization objectives of minimizing the first preparation time and minimizing the first raw material consumption; Based on the parameter coordination coefficients in the global parameter optimization matrix, the value constraints and interaction constraints of each process parameter are determined. The reaction temperature range, raw material ratio limits, and product purity standards are treated as hard constraints and incorporated into the constraints of the multi-objective optimization problem.

9. The optimized preparation process of an organic descaling agent according to claim 8, characterized in that, The multi-objective optimization problem is solved using a multi-objective optimization algorithm to obtain the optimal process parameter scheme, including: A non-dominated sorting genetic algorithm is used to solve the multi-objective optimization problem, and the population includes various combinations of process parameters. The objective function value of each individual in the population is calculated based on the global parameter optimization matrix, and non-dominated sorting and crowding calculation are performed. The next generation population is generated through crossover and mutation operations, while retaining the Pareto optimal solution; After iterating to a preset number of times, the parameter combination with the best overall performance is selected from the final Pareto optimal solution set as the optimal process parameter scheme.

10. A system for optimizing the preparation process of an organic descaling agent, applicable to the method for optimizing the preparation process of an organic descaling agent as described in any one of claims 1-9, characterized in that, include: The parameter acquisition unit is used to acquire raw material characteristic data, formulation parameters, and reaction condition parameters of the preparation process to be optimized; wherein, the formulation parameters include several reaction stages, the optimal process path for each reaction stage, and the preset control nodes corresponding to the optimal process path, and the preset control nodes are set as reaction inflection points. The parameter coupling unit is used to calculate the parameter sensitivity distribution of the preparation process through reaction kinetics simulation based on the raw material characteristic data and reaction condition parameters, and to determine the influence range of process parameters according to the parameter coupling model. The range division unit is used to divide the parameter adjustment space of the preparation process into multiple interconnected and independently optimizable adjustment sub-regions according to the influence range of the process parameters. Each adjustment sub-region includes the process parameters. The efficiency analysis unit is used to take the execution order of each reaction stage as the first optimization parameter and the core process parameters of each regulation sub-region as the second optimization parameter; based on the first optimization parameter, the second optimization parameter and the pre-trained reaction simulation model, the initial reaction efficiency of each preset control node is determined, wherein the reaction simulation model is a mapping relationship between parameter values ​​and reaction efficiency under standard raw material conditions; the initial reaction efficiency is corrected based on the raw material characteristic data using the pre-trained influence correction model. The parameter optimization unit is used to perform sensitivity analysis and optimization calculations on the process parameters of multiple adjustment sub-regions according to a preset parameter adjustment order to obtain a sub-region optimized parameter set; and to perform coupling verification on the sub-region optimized parameter set to generate a global parameter optimization matrix. The process optimization unit is used to determine the first preparation time and the first raw material consumption as optimization objectives based on the corrected reaction efficiency, the first optimization parameter, and the second optimization parameter, and to construct a multi-objective optimization problem by combining the global parameter optimization matrix; and to solve the multi-objective optimization problem by using a multi-objective optimization algorithm to obtain the optimal process parameter scheme.

Images