Method and system for adaptively adjusting reaction temperature of organosilicon additive synthesis

Through the adaptive temperature adjustment method of silicone additive synthesis reaction, the deep neural network model and cloning selection algorithm are used to realize the dynamic temperature control during the silicone additive synthesis process, solving the problem of inaccurate temperature control and improving product quality and productivity.

CN118672325BActive Publication Date: 2024-11-15SUZHOU BONA CHEM TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411158648.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-15
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The temperature control in the existing silicone additive synthesis reaction is inaccurate, resulting in insufficient reaction activity, reduced product quality, high energy consumption and high production costs.

Method used

The temperature adaptive adjustment method of silicone additive synthesis reaction is adopted to collect and analyze raw material data, predict the optimal temperature curve, and adjust the temperature in real time to achieve dynamic temperature control. The method includes raw material data acquisition, analysis, real-time data acquisition and prediction, and automatically learning and adaptively adjusting the temperature curve using deep neural network models and cloning selection algorithms.

Benefits of technology

It improves the accuracy and flexibility of temperature control, realizes intelligent temperature monitoring throughout the process from raw materials entering the warehouse to the outflow of finished products, reduces production costs and ensures product quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118672325B_ABST
    Figure CN118672325B_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of temperature control, and discloses a method and system for adaptively adjusting the reaction temperature of an organosilicon additive synthesis reaction; the method comprises: collecting raw material data; analyzing the raw material data, and retaining m temperature curves from a preset curve set; analyzing the raw material data again, obtaining an optimal temperature curve in the synthesis process of the organosilicon additive, and synthesizing the organosilicon additive according to the optimal temperature curve; collecting real-time data in the synthesis process of the organosilicon additive; analyzing the real-time data, and predicting future temperature data; and adjusting the optimal temperature curve according to the future temperature data; the invention integrates a closed-loop system of data collection, parameter analysis and prediction, and temperature control, and can optimize and correct the temperature value in the synthesis process of the organosilicon additive in real time according to different reaction conditions, thereby greatly improving the accuracy of temperature control, and having strong flexibility and a high degree of personalization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of temperature control, and more specifically to a method and system for self-adapting temperature regulation of a synthesis reaction of an organosilicon additive. Background Art

[0002] Silicone additives are one of the important functional additives in the coating industry. They are mainly used to provide coating protection and enhance coating performance. Usually, silicone additives are synthesized by solid or liquid addition reactions of silicone raw materials with organic groups such as olefins and enols. Such chemical reactions usually use heating to generate activation energy, but the optimal reaction temperature may vary under different reaction systems and conditions.

[0003] In the prior art, the synthesis reaction of the organosilicon additive is carried out by constant temperature heating, and the temperature is kept constant in each reaction stage. For example, the Chinese patent with publication number CN117816518A discloses a coating process using PMA propylene glycol methyl ether acetate; in this invention, under the condition of different raw material ratios, the temperature required for the same reaction stage is different, and the temperature of different reaction stages needs to be manually regulated, resulting in inaccurate temperature control and insufficient flexibility, which easily leads to the following problems:

[0004] 1. The temperature may be too low in the early stage, resulting in insufficient reaction activity and low yield;

[0005] 2. The temperature in the middle stage may be too high, causing the product to decompose and resulting in a decrease in product quality;

[0006] 3. The performance of the obtained product cannot reach the ideal level;

[0007] 4. Poor energy consumption control leads to higher production costs;

[0008] In view of this, the present invention proposes a method and system for adaptively adjusting the reaction temperature of a silicone additive synthesis reaction to solve the above-mentioned problem. Summary of the invention

[0009] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for adaptively adjusting the temperature of a synthesis reaction of an organosilicon additive, comprising:

[0010] Collect raw material data;

[0011] Analyze the raw material data and retain m temperature curves from a preset curve set, where m is an integer greater than 1;

[0012] Analyze the raw material data again to obtain the optimal temperature curve in the synthesis process of the organosilicon additive, and synthesize the organosilicon additive according to the optimal temperature curve;

[0013] Collect real-time data during the synthesis of silicone additives;

[0014] Analyze real-time data and predict future temperature data;

[0015] The optimal temperature curve is adjusted based on future temperature data.

[0016] Furthermore, the raw material data includes raw material type and raw material ratio; the raw material type is the type of various raw materials required in the synthesis process of the organosilicon additive; the raw material ratio is the mass ratio of various raw materials in the synthesis process of the organosilicon additive.

[0017] Furthermore, the method of retaining m temperature curves from a preset curve set includes:

[0018] The curve set includes M temperature curves. ;

[0019] Obtain the maximum and minimum values ​​of the M temperature curves in sequence, subtract the corresponding minimum values ​​from the maximum values ​​of the M temperature curves, and obtain the extreme difference values ​​corresponding to the M temperature curves;

[0020] Get the number of raw material types, which is the number of raw material types;

[0021] Calculate the temperature range based on raw material data;

[0022] The expression of temperature extreme difference is: ;

[0023] In the formula, For extreme temperature differences, is the raw material type value of the i-th raw material, is the raw material proportion of the i-th raw material, is the weight coefficient of the i-th raw material, sec is the secant function, , n is the number of raw material types;

[0024] Subtract the temperature extreme difference from the extreme difference values ​​corresponding to the M temperature curves and take the absolute value to obtain the difference; sort the difference values ​​from small to large, and obtain the values ​​in the front in positive order. The difference value is retained from the curve set There are m temperature curves corresponding to the difference of .

[0025] Furthermore, the step of obtaining the optimal temperature curve during the synthesis of the organosilicon additive includes:

[0026] Step 1: Preset the iteration threshold and set digital labels for m temperature curves in order. The digital label range is , mark the digital label corresponding to the temperature curve as the curve label; randomly generate an initial antibody population of size q ;

[0027] Step 2: Determine the fitness function;

[0028] Step 3: Calculate the initial antibody population The fitness of each antibody in , select r antibodies with the largest fitness, and construct a memory population ;

[0029] Step 4: Memory population Each antibody in the cloning is cloned according to the calculated clone size to generate a clonal antibody population ;

[0030] Step 5: Cloning of Antibody Populations Each antibody in the cloud uses a cloud adaptive mutation operator to perform cloud adaptive mutation to generate a mutant antibody group ;

[0031] Step 6: Cloning antibody populations Randomly select w antibodies as parent antibodies for antibody recombination to generate offspring antibodies , until the clonal antibody population When the number of unselected antibodies is less than w, antibody recombination is stopped and a recombinant antibody group is generated. ;

[0032] Step 7: Merge the variant antibody groups and recombinant antibody groups , generating a pooled antibody population ;

[0033] Step 8: Randomly generate d new antibodies and merge the antibody groups The d antibodies with the smallest fitness are replaced by d new antibodies generated randomly;

[0034] Step 9: If the number of cycles t is less than the iteration threshold, let , and select the combined antibody group The r antibodies with the largest fitness form a new memory population , loop steps 3 to 8; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group is obtained The temperature curve corresponding to the antibody with the largest fitness is taken as the optimal temperature curve.

[0035] Furthermore, in step 1, the initial antibody group The antibodies in the curve label correspond one to one, and the range of the antibody is the range of the curve label, that is, the range of the antibody is ;

[0036] In step 2, the expression of the fitness function is: ;

[0037] In the formula, is the fitness of the jth antibody, is the product mass of the jth antibody, ;

[0038] The mass of the product of the jth antibody is the mass of the organosilicon additive synthesized under the temperature curve conditions corresponding to the jth antibody;

[0039] Methods for obtaining product quality include:

[0040] The raw material type value and raw material ratio are used as updated raw material data, and the curve label corresponding to the antibody and the updated raw material data are used as analysis data; the analysis data is input into the trained quality prediction model to predict the corresponding product quality;

[0041] The training process of the quality prediction model includes:

[0042] Pre-collecting N groups of analytical data corresponding to the product masses, and converting the analytical data and the corresponding product masses into a corresponding set of feature vectors;

[0043] Each group of feature vectors is used as the input of the quality prediction model. The quality prediction model takes a group of product masses corresponding to each group of analysis data as output, and takes the actual product mass corresponding to each group of analysis data as the prediction target, where the actual product mass is the pre-collected product mass corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training target; the quality prediction model is trained until the sum of prediction errors reaches convergence and the training is stopped; the quality prediction model is a deep neural network model.

[0044] Furthermore, in step 4, the method for calculating the clone scale includes:

[0045] ;

[0046] ;

[0047] In the formula, is the cloning scale of the jth antibody, Int is the upward rounding function, q is the size of the antibody group, is the affinity of the j-th antibody, min is the minimum function, exp is the natural exponential function, is the Euclidean distance between the j-th antibody and the e-th antibody, , ;

[0048] In step 6, the progeny antibody The calculation methods include:

[0049] ;

[0050] In the formula, is the w-th parent antibody, is the w-th scaling factor, which is a randomly generated real number and not all w scaling factors are 0.

[0051] Furthermore, the real-time data includes temperature value, pressure value, pH value and synthesis time; the temperature value is the temperature value inside the reactor during the synthesis of the organosilicon additive, and the reactor is a device for synthesizing the organosilicon additive; the pressure value is the pressure value inside the reactor during the synthesis of the organosilicon additive; the pH value is the pH value of the reaction liquid during the synthesis of the organosilicon additive; the synthesis time is the length of time for the synthesis of the organosilicon additive.

[0052] Furthermore, the method for predicting future temperature data includes:

[0053] Real-time data and updated raw material data are used as test data, and the test data are input into the trained temperature prediction model to predict future temperature data; the future temperature data is the temperature value corresponding to the next time point of the synthesis time; the training process of the temperature prediction model is consistent with the training process of the quality prediction model, and it is also a deep neural network model.

[0054] Furthermore, a time point corresponding to the future temperature data is obtained and marked as an adjustment time point, and the temperature value corresponding to the adjustment time point in the optimal temperature curve is adjusted to the future temperature data.

[0055] Further, the future temperature data is analyzed to obtain multi-point temperature data, and the optimal temperature curve is adjusted according to the multi-point temperature data;

[0056] Methods for obtaining multi-point temperature data include:

[0057] Subtract the temperature value corresponding to the adjustment time point from the future temperature data to obtain the temperature error; preset a demarcation value, divide the temperature error by the demarcation value to obtain the temperature ratio, and use the integer part of the temperature ratio as the number of adjustment points, which is the number of time points that need to be adjusted in addition to the adjustment time point;

[0058] Marking additional time points that need adjustment as additional adjustment points; the multi-point temperature data includes temperature data of each additional adjustment point;

[0059] The calculation method of the temperature data corresponding to each additional adjustment point includes:

[0060] ;

[0061] In the formula, is the temperature data of the pth additional adjustment point, is the proportional coefficient of the pth additional adjustment point, is the temperature error, is the temperature value corresponding to the pth additional adjustment point in the optimal temperature curve, To adjust the points, ;

[0062] The optimum temperature curve The temperature values ​​corresponding to the additional adjustment points are adjusted in turn to the multi-point temperature data Temperature data corresponding to the additional adjustment points.

[0063] The temperature self-adapting regulating system for the synthesis reaction of an organosilicon additive implements the temperature self-adapting regulating method for the synthesis reaction of an organosilicon additive, comprising:

[0064] Raw material data collection module, used for collecting raw material data;

[0065] A raw material data analysis module is used to analyze the raw material data and retain m temperature curves from a preset curve set, where m is an integer greater than 1;

[0066] The secondary analysis module is used to analyze the raw material data again, obtain the optimal temperature curve in the synthesis process of the organosilicon additive, and synthesize the organosilicon additive according to the optimal temperature curve;

[0067] Real-time data acquisition module, used to collect real-time data during the synthesis process of organosilicon additives;

[0068] Real-time data analysis module, used to analyze real-time data and predict future temperature data;

[0069] The temperature adjustment module is used to adjust the optimal temperature curve according to future temperature data.

[0070] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for adaptively adjusting the temperature of a synthesis reaction of an organosilicon additive is implemented.

[0071] A computer-readable storage medium stores a computer program, and when the computer program is executed, the method for adaptively adjusting the temperature of a synthesis reaction of an organosilicon additive is implemented.

[0072] The technical effects and advantages of the method and system for self-adapting the temperature of the organosilicon additive synthesis reaction of the present invention are as follows:

[0073] 1. The use of multi-factor analysis of raw materials and real-time multi-parameter dynamic monitoring not only takes into account the combined effects of different raw materials, but also learns and processes the effects of multiple factors such as pressure and pH value in real time, greatly improving the accuracy of temperature control; the automatic learning and adaptive adjustment of the temperature curve are realized through the fast-converging clone selection algorithm and machine learning model, which is highly flexible and personalized; the closed-loop system integrates data acquisition, parameter analysis and prediction, and temperature control, which can optimize and correct the temperature value in the synthesis process of silicone additives in real time according to different reaction conditions, and realize intelligent temperature monitoring of the entire process from raw material storage to finished product delivery.

[0074] 2. By calculating the temperature error, the number of time points that need to be adjusted is dynamically determined according to the size of the temperature error, rather than being limited to single-point adjustment. By adjusting the temperature at multiple time points, the prediction information can be transmitted more widely, the time range of prediction optimization can be extended, the error of a single prediction point can be compensated, and a smooth temperature curve can be formed, thereby more fully alleviating the lag of temperature control. By utilizing multi-point joint optimization, the control efficiency can be improved while ensuring the control quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a schematic diagram of a temperature adaptive adjustment system for a synthesis reaction of an organosilicon additive according to Example 1 of the present invention;

[0076] Figure 2 This is a schematic diagram of a temperature adaptive adjustment system for a synthesis reaction of an organosilicon additive according to Example 2 of the present invention;

[0077] Figure 3 This is a flow chart of a method for self-adapting the temperature of a synthesis reaction of an organosilicon additive according to Example 3 of the present invention;

[0078] Figure 4 This is a schematic diagram of an electronic device according to Embodiment 4 of the present invention;

[0079] Figure 5 This is a schematic diagram of a storage medium according to Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0080] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0081] Embodiment 1:

[0082] See also Figure 1As shown, the temperature adaptive regulation system for the synthesis reaction of the organosilicon additive described in this embodiment includes a raw material data acquisition module, a raw material data analysis module, a secondary analysis module, a real-time data acquisition module, a real-time data analysis module and a temperature regulation module; each module is connected by wire and / or wireless means to realize data transmission between modules;

[0083] Raw material data collection module, used for collecting raw material data;

[0084] Raw material data includes raw material type and raw material proportion;

[0085] The raw material type refers to the type of various raw materials required in the synthesis process of the organosilicon additive; the raw material type is detected and obtained by a Raman spectrometer installed above a sampling platform, and the sampling platform is a platform for placing raw materials; wherein the sampling platform is divided into m areas, m is an integer greater than 1, and different types of raw materials are placed in different areas, and the Raman spectrometer is used to sequentially align different areas to obtain the corresponding types of raw materials in different areas; different types of raw materials have different corresponding thermal stability, dissociation temperature and activation energy, so the temperature in the synthesis process of the organosilicon additive is also different; and in each synthesis process of the organosilicon additive, due to different raw material types, different interactions between raw materials will occur, such as catalysis, inhibition, etc., which will affect the conversion rate of different raw material molecules or intermediates in the reaction process, and then affect the temperature;

[0086] The raw material ratio is the mass ratio of various raw materials in the synthesis process of the organosilicon additive; the raw material ratio is obtained by the mass sensor installed under the sampling platform, wherein a command sensor is installed under each area to obtain the mass of each raw material; the mass of each area is added in turn to obtain the total mass, and then the mass of each raw material is divided by the total mass in turn to obtain the mass ratio of each raw material;

[0087] A raw material data analysis module is used to analyze the raw material data and retain m temperature curves from a preset curve set, where m is an integer greater than 1;

[0088] The method of retaining m temperature curves from a preset curve set includes:

[0089] The curve set includes M temperature curves. ; The M temperature curves are all temperature curves when the historical organosilicon additive is successfully synthesized and the performance of the organosilicon additive product is better. When the historical organosilicon additive is synthesized, the technical personnel in this field collect the temperature of the entire synthesis process and draw the corresponding temperature curve, retain the temperature curve corresponding to the successful synthesis of the organosilicon additive, and test the performance of the organosilicon additive product, and retain the M temperature curves corresponding to the better performance of the organosilicon additive product;

[0090] Obtain the maximum and minimum values ​​of the M temperature curves in sequence, subtract the corresponding minimum values ​​from the maximum values ​​of the M temperature curves, and obtain the extreme difference values ​​corresponding to the M temperature curves;

[0091] Get the number of raw material types, which is the number of raw material types;

[0092] Calculate the temperature range based on raw material data;

[0093] The expression of temperature extreme difference is: ;

[0094] In the formula, For extreme temperature differences, is the raw material type value of the i-th raw material, is the raw material proportion of the i-th raw material, is the weight coefficient of the i-th raw material, sec is the secant function, , n is the number of raw material types;

[0095] The specific value of the weight coefficient in the formula can be set according to the actual situation. The weight coefficient reflects the influence of different raw material type values ​​and raw material proportions on the temperature range. Those skilled in the art can preset the corresponding weight coefficient according to the actual influence of different raw material type values ​​and raw material proportions on the temperature range, so as to accurately evaluate the temperature range when synthesizing silicone additives with different raw materials;

[0096] It should be noted that the raw material type value is pre-set by a person skilled in the art according to the temperature difference when each raw material type reacts separately, and the temperature difference is the highest temperature minus the lowest temperature when the raw material reacts separately; a plurality of raw material types are reacted separately in advance to obtain the corresponding temperature difference, and a raw material type with a larger temperature difference when reacting separately is assigned a larger raw material type value, and a raw material type with a smaller temperature difference when reacting separately is assigned a smaller raw material type value; for example, the raw material type value of carbon tetrachloride is 20, the raw material type value of polycarbonate polyol is 15, and the raw material type value of methyl silicone oil is 5;

[0097] Subtract the temperature extreme difference from the extreme difference values ​​corresponding to the M temperature curves and take the absolute value to obtain the difference; sort the difference values ​​from small to large, and obtain the values ​​in the front in positive order. The difference value is retained from the curve set There are m temperature curves corresponding to the difference of ;

[0098] The secondary analysis module is used to analyze the raw material data again, obtain the optimal temperature curve in the synthesis process of the organosilicon additive, and synthesize the organosilicon additive according to the optimal temperature curve;

[0099] The steps to obtain the optimal temperature profile during the synthesis of silicone additives include:

[0100] Step 1: Preset the iteration threshold and set digital labels for m temperature curves in order. The digital label range is , mark the digital label corresponding to the temperature curve as the curve label; randomly generate an initial antibody population of size q ;The iteration threshold is pre-set according to the algorithm accuracy;

[0101] Step 2: Determine the fitness function;

[0102] Step 3: Calculate the initial antibody population The fitness of each antibody in , select r antibodies with the largest fitness, and construct a memory population ;

[0103] Step 4: Memory population Each antibody in the cloning is cloned according to the calculated clone size to generate a clonal antibody population ;

[0104] Step 5: Cloning of Antibody Populations Each antibody in the cloud uses a cloud adaptive mutation operator to perform cloud adaptive mutation to generate a mutant antibody group ; The cloud adaptive mutation operator is an existing technology and will not be described in detail here;

[0105] Step 6: Cloning antibody populations Randomly select w antibodies as parent antibodies for antibody recombination to generate offspring antibodies , until the clonal antibody population When the number of unselected antibodies is less than w, antibody recombination is stopped and a recombinant antibody group is generated. ; In this embodiment, w is preferably 3; recombinant antibody group Multiple progeny antibodies ;

[0106] Step 7: Merge the variant antibody groups and recombinant antibody groups , generating a pooled antibody population ;

[0107] Step 8: Randomly generate d new antibodies and merge the antibody groups The d antibodies with the smallest fitness are replaced by d new antibodies generated randomly;

[0108] Step 9: If the number of cycles t is less than the iteration threshold, let , and select the combined antibody group The r antibodies with the largest fitness form a new memory population , loop steps 3 to 8; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group is obtained The temperature curve corresponding to the antibody with the largest fitness is taken as the optimal temperature curve; among them, For computer languages, the value of the number of iterations t is added by 1 and then assigned to the number of iterations t;

[0109] In step 1 above, the initial antibody population The antibodies in the curve label correspond one to one, and the range of the antibody is the range of the curve label, that is, the range of the antibody is ;

[0110] In the above step 2, the expression of the fitness function is: ;

[0111] In the formula, is the fitness of the jth antibody, is the product mass of the jth antibody, ;

[0112] The mass of the product of the jth antibody is the mass of the organosilicon additive synthesized under the temperature curve conditions corresponding to the jth antibody;

[0113] Methods for obtaining product quality include:

[0114] The raw material type value and raw material ratio are used as updated raw material data, and the curve label corresponding to the antibody and the updated raw material data are used as analysis data; the analysis data is input into the trained quality prediction model to predict the corresponding product quality;

[0115] The specific training process of the quality prediction model includes:

[0116] Pre-collecting N groups of analytical data corresponding to the product masses, and converting the analytical data and the corresponding product masses into a corresponding set of feature vectors;

[0117] Each set of feature vectors is used as the input of the quality prediction model. The quality prediction model uses a set of product masses corresponding to each set of analytical data as output, and uses the actual product mass corresponding to each set of analytical data as the prediction target. The actual product mass is the pre-collected product mass corresponding to the analytical data. The training target is to minimize the sum of the prediction errors of all analytical data. The calculation formula of the prediction error is: ,in is the prediction error, K is the group number of the eigenvector corresponding to the analysis data, is the product mass corresponding to the Kth group of analysis data, is the actual product quality corresponding to the Kth group of analysis data; the quality prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped;

[0118] The above-mentioned quality prediction model is specifically a deep neural network model;

[0119] It should be noted that the product quality corresponding to the analysis data is determined by a technician in the field who collects multiple sets of different raw material data during the synthesis of the organosilicon additives in the past. Under the conditions of each set of raw material data, the organosilicon additives are synthesized according to m temperature curves. The quality of the synthesized multiple organosilicon additives is evaluated based on actual experience, and corresponding evaluation scores are assigned. The evaluation score of each organosilicon additive is used as the product quality corresponding to the corresponding analysis data; the quality evaluation, for example, evaluates the concentration, adhesion, wear resistance, etc. of the organosilicon additive;

[0120] In the above step 4, the calculation method of the clone size includes:

[0121] ;

[0122] ;

[0123] In the formula, is the cloning scale of the jth antibody, Int is the upward rounding function, q is the size of the antibody group, is the affinity of the j-th antibody, min is the minimum function, exp is the natural exponential function, is the Euclidean distance between the j-th antibody and the e-th antibody, , ;

[0124] It should be noted that when calculating the Euclidean distance between the j-th antibody and the e-th antibody, normalization is performed so that ;

[0125] In step 6 above, the progeny antibody The calculation methods include:

[0126] ;

[0127] In the formula, is the w-th parent antibody, is the wth scale factor, the scale factor is a randomly generated real number and the w scale factors are not all 0;

[0128] It should be noted that the purpose of analyzing the raw material data twice is to reduce the temperature curve range when obtaining the optimal temperature curve, that is, to reduce the antibody range; it helps to improve the algorithm's exploration ability and better cover the problem solution space; it can also reduce the size of the search space, making it easier for the algorithm to find the optimal solution or a solution close to the optimal solution, and speed up the convergence of the algorithm; at the same time, it helps to avoid overfitting the algorithm to the training data and improve the generalization ability of the algorithm; in addition, it can also improve the stability and reliability of the algorithm;

[0129] Real-time data acquisition module, used to collect real-time data during the synthesis process of organosilicon additives;

[0130] Real-time data includes temperature, pressure, pH and synthesis time;

[0131] The temperature value is the temperature value in the reactor during the synthesis of the organosilicon additive. The reactor is a device for synthesizing the organosilicon additive. The temperature value is obtained by a temperature sensor installed in the reactor.

[0132] The pressure value is the pressure value in the reactor during the synthesis of the organosilicon additive; the pressure value is obtained by a pressure sensor installed in the reactor;

[0133] pH value: The pH value of the reaction solution during the synthesis of silicone additives; the pH value is obtained by the embedded pH electrode installed in the reactor;

[0134] The synthesis time is the length of time for the synthesis of the organosilicon additive; the synthesis time is obtained by the timer built into the reactor;

[0135] Real-time data analysis module, used to analyze real-time data and predict future temperature data;

[0136] Methods for predicting future temperature data include:

[0137] The real-time data and updated raw material data are used as test data, and the test data are input into the trained temperature prediction model to predict future temperature data; the future temperature data is the temperature value corresponding to the next time point of the synthesis time; the specific training process of the temperature prediction model is consistent with the training process of the quality prediction model, and it is also a deep neural network model;

[0138] For example, the synthesis time in the real-time data is N seconds, that is, the real-time data is the data collected when the synthesis of the silicone additive is carried out for N seconds, and the predicted future temperature data at this time is the temperature value at N+1 seconds;

[0139] It should be noted that real-time data will have an impact on future temperature data; temperature values ​​are the basis for predicting future temperature data; pressure values ​​and pH values ​​will affect the reaction rate and heat production, thereby affecting future temperature data; and synthesis time is used to reflect the time corresponding to future temperature data. Different times in the synthesis process of silicone additives correspond to different temperature values;

[0140] The temperature adjustment module is used to adjust the optimal temperature curve according to future temperature data;

[0141] Obtain the time point corresponding to the future temperature data and mark it as the adjustment time point, and adjust the temperature value corresponding to the adjustment time point in the optimal temperature curve to the future temperature data;

[0142] This embodiment adopts multi-factor analysis of raw materials and real-time multi-parameter dynamic monitoring, which not only takes into account the combined influence of different raw materials, but also learns and processes the influence of multiple factors such as pressure and pH value in real time, greatly improving the accuracy of temperature control; through the fast-converging clone selection algorithm and machine learning model, automatic learning and adaptive adjustment of the temperature curve are realized, which has strong flexibility and high degree of personalization; the closed-loop system integrating data acquisition, parameter analysis and prediction and temperature control can optimize and correct the temperature value in the synthesis process of silicone additives in real time according to different reaction conditions, and realize intelligent temperature monitoring of the whole process from raw material storage to finished product delivery.

[0143] Embodiment 2:

[0144] See also Figure 2 As shown, this embodiment further improves the design on the basis of embodiment 1. In embodiment 1, when adjusting the optimal temperature curve, only the temperature value at the time point corresponding to the future temperature data is adjusted. If it is predicted that the future temperature data has a large error with the optimal temperature curve, a single point replacement is difficult to alleviate the trend of the error increase at the subsequent time point, so that the temperature value at the subsequent time point cannot be adjusted in time, resulting in a lag in temperature control; therefore, this embodiment provides a temperature adaptive adjustment system for the synthesis reaction of an organosilicon additive, which also includes adjusting the temperature values ​​at multiple time points;

[0145] Multi-point control module, used to analyze future temperature data, obtain multi-point temperature data, and adjust the optimal temperature curve according to the multi-point temperature data;

[0146] Methods for obtaining multi-point temperature data include:

[0147] The temperature value corresponding to the adjustment time point is subtracted from the future temperature data to obtain the temperature error; a demarcation value is preset, and the temperature error is divided by the demarcation value to obtain the temperature ratio, and the integer part of the temperature ratio is used as the adjustment point number, and the adjustment point number is the number of time points that need to be adjusted in addition to the adjustment time point; the demarcation value is pre-set by a person skilled in the art according to the temperature error range, and the temperature error range is obtained by a person skilled in the art in the historical synthesis process of organic silicone additives by collecting real-time data and calculating future temperature data for multiple times to obtain the corresponding temperature error, and the largest temperature error and the smallest temperature error among the multiple temperature errors are used as the temperature error range;

[0148] Marking additional time points that need adjustment as additional adjustment points; the multi-point temperature data includes temperature data of each additional adjustment point;

[0149] The calculation method of the temperature data corresponding to each additional adjustment point includes:

[0150] ;

[0151] In the formula, is the temperature data of the pth additional adjustment point, is the proportional coefficient of the pth additional adjustment point, is the temperature error, is the temperature value corresponding to the pth additional adjustment point in the optimal temperature curve, To adjust the points, ;

[0152] It should be noted that the proportionality coefficient It is preset by technicians in this field based on their practical experience. , this embodiment preferably is 0.9;

[0153] The optimal temperature curve The temperature values ​​corresponding to the additional adjustment points are adjusted in turn to the multi-point temperature data Temperature data corresponding to additional adjustment points;

[0154] This embodiment calculates the temperature error and dynamically determines the number of time points that need to be adjusted according to the size of the temperature error, rather than being limited to single-point adjustment. By adjusting the temperature at multiple time points, prediction information can be transmitted more widely, the time range of prediction optimization can be extended, the error of a single prediction point can be compensated, and a smooth temperature curve can be formed, thereby more fully alleviating the lag of temperature control. By utilizing multi-point joint optimization, control efficiency can be improved while ensuring control quality.

[0155] Embodiment 3:

[0156] See also Figure 3As shown, the part not described in detail in this embodiment is described in Examples 1 and 2, and a method for adaptively adjusting the reaction temperature of a silicone additive synthesis reaction is provided, the method comprising:

[0157] Collect raw material data;

[0158] Analyze the raw material data and retain m temperature curves from a preset curve set, where m is an integer greater than 1;

[0159] Analyze the raw material data again to obtain the optimal temperature curve in the synthesis process of the organosilicon additive, and synthesize the organosilicon additive according to the optimal temperature curve;

[0160] Collect real-time data during the synthesis of silicone additives;

[0161] Analyze real-time data and predict future temperature data;

[0162] The optimal temperature curve is adjusted based on future temperature data.

[0163] Furthermore, the raw material data includes raw material type and raw material ratio; the raw material type is the type of various raw materials required in the synthesis process of the organosilicon additive; the raw material ratio is the mass ratio of various raw materials in the synthesis process of the organosilicon additive.

[0164] Furthermore, the method of retaining m temperature curves from a preset curve set includes:

[0165] The curve set includes M temperature curves. ;

[0166] Obtain the maximum and minimum values ​​of the M temperature curves in sequence, subtract the corresponding minimum values ​​from the maximum values ​​of the M temperature curves, and obtain the extreme difference values ​​corresponding to the M temperature curves;

[0167] Get the number of raw material types, which is the number of raw material types;

[0168] Calculate the temperature range based on raw material data;

[0169] The expression of temperature extreme difference is: ;

[0170] In the formula, For extreme temperature differences, is the raw material type value of the i-th raw material, is the raw material proportion of the i-th raw material, is the weight coefficient of the i-th raw material, sec is the secant function, , n is the number of raw material types;

[0171] Subtract the temperature extreme difference from the extreme difference values ​​corresponding to the M temperature curves and take the absolute value to obtain the difference; sort the difference values ​​from small to large, and obtain the values ​​in the front in positive order. The difference value is retained from the curve set There are m temperature curves corresponding to the difference of .

[0172] Furthermore, the step of obtaining the optimal temperature curve during the synthesis of the organosilicon additive includes:

[0173] Step 1: Preset the iteration threshold and set digital labels for m temperature curves in order. The digital label range is , mark the digital label corresponding to the temperature curve as the curve label; randomly generate an initial antibody population of size q ;

[0174] Step 2: Determine the fitness function;

[0175] Step 3: Calculate the initial antibody population The fitness of each antibody in , select r antibodies with the largest fitness, and construct a memory population ;

[0176] Step 4: Memory population Each antibody in the cloning is cloned according to the calculated clone size to generate a clonal antibody population ;

[0177] Step 5: Cloning of Antibody Populations Each antibody in the cloud uses a cloud adaptive mutation operator to perform cloud adaptive mutation to generate a mutant antibody group ;

[0178] Step 6: Cloning antibody populations Randomly select w antibodies as parent antibodies for antibody recombination to generate offspring antibodies , until the clonal antibody population When the number of unselected antibodies is less than w, antibody recombination is stopped and a recombinant antibody group is generated. ;

[0179] Step 7: Merge the variant antibody groups and recombinant antibody groups , generating a pooled antibody population ;

[0180] Step 8: Randomly generate d new antibodies and merge the antibody groups The d antibodies with the smallest fitness are replaced by d new antibodies generated randomly;

[0181] Step 9: If the number of cycles t is less than the iteration threshold, let , and select the combined antibody group The r antibodies with the largest fitness form a new memory population , loop steps 3 to 8; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group is obtained The temperature curve corresponding to the antibody with the largest fitness is taken as the optimal temperature curve.

[0182] Furthermore, in step 1, the initial antibody group The antibodies in the curve label correspond one to one, and the range of the antibody is the range of the curve label, that is, the range of the antibody is ;

[0183] In step 2, the expression of the fitness function is: ;

[0184] In the formula, is the fitness of the jth antibody, is the product mass of the jth antibody, ;

[0185] The mass of the product of the jth antibody is the mass of the organosilicon additive synthesized under the temperature curve conditions corresponding to the jth antibody;

[0186] Methods for obtaining product quality include:

[0187] The raw material type value and raw material ratio are used as updated raw material data, and the curve label corresponding to the antibody and the updated raw material data are used as analysis data; the analysis data is input into the trained quality prediction model to predict the corresponding product quality;

[0188] The training process of the quality prediction model includes:

[0189] Pre-collecting N groups of analytical data corresponding to the product masses, and converting the analytical data and the corresponding product masses into a corresponding set of feature vectors;

[0190] Each group of feature vectors is used as the input of the quality prediction model. The quality prediction model takes a group of product masses corresponding to each group of analysis data as output, and takes the actual product mass corresponding to each group of analysis data as the prediction target, where the actual product mass is the pre-collected product mass corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training target; the quality prediction model is trained until the sum of prediction errors reaches convergence and the training is stopped; the quality prediction model is a deep neural network model.

[0191] Furthermore, in step 4, the method for calculating the clone scale includes:

[0192] ;

[0193] ;

[0194] In the formula, is the cloning scale of the jth antibody, Int is the upward rounding function, q is the size of the antibody group, is the affinity of the j-th antibody, min is the minimum function, exp is the natural exponential function, is the Euclidean distance between the j-th antibody and the e-th antibody, , ;

[0195] In step 6, the progeny antibody The calculation methods include:

[0196] ;

[0197] In the formula, is the w-th parent antibody, is the w-th scaling factor, which is a randomly generated real number and not all w scaling factors are 0.

[0198] Furthermore, the real-time data includes temperature value, pressure value, pH value and synthesis time; the temperature value is the temperature value inside the reactor during the synthesis of the organosilicon additive, and the reactor is a device for synthesizing the organosilicon additive; the pressure value is the pressure value inside the reactor during the synthesis of the organosilicon additive; the pH value is the pH value of the reaction liquid during the synthesis of the organosilicon additive; the synthesis time is the length of time for the synthesis of the organosilicon additive.

[0199] Furthermore, the method for predicting future temperature data includes:

[0200] Real-time data and updated raw material data are used as test data, and the test data are input into the trained temperature prediction model to predict future temperature data; the future temperature data is the temperature value corresponding to the next time point of the synthesis time; the training process of the temperature prediction model is consistent with the training process of the quality prediction model, and it is also a deep neural network model.

[0201] Furthermore, a time point corresponding to the future temperature data is obtained and marked as an adjustment time point, and the temperature value corresponding to the adjustment time point in the optimal temperature curve is adjusted to the future temperature data.

[0202] Further, the future temperature data is analyzed to obtain multi-point temperature data, and the optimal temperature curve is adjusted according to the multi-point temperature data;

[0203] Methods for obtaining multi-point temperature data include:

[0204] Subtract the temperature value corresponding to the adjustment time point from the future temperature data to obtain the temperature error; preset a demarcation value, divide the temperature error by the demarcation value to obtain the temperature ratio, and use the integer part of the temperature ratio as the number of adjustment points, which is the number of time points that need to be adjusted in addition to the adjustment time point;

[0205] Marking additional time points that need adjustment as additional adjustment points; the multi-point temperature data includes temperature data of each additional adjustment point;

[0206] The calculation method of the temperature data corresponding to each additional adjustment point includes:

[0207] ;

[0208] In the formula, is the temperature data of the pth additional adjustment point, is the proportional coefficient of the pth additional adjustment point, is the temperature error, is the temperature value corresponding to the pth additional adjustment point in the optimal temperature curve, To adjust the points, ;

[0209] The optimal temperature curve The temperature values ​​corresponding to the additional adjustment points are adjusted in turn to the multi-point temperature data Temperature data corresponding to the additional adjustment points.

[0210] Embodiment 4:

[0211] See also Figure 4 As shown, according to another aspect of the present application, an electronic device 500 is also provided. The electronic device 500 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are run by one or more processors, the method for adaptively adjusting the temperature of the synthesis reaction of the organosilicon additive as described above may be executed.

[0212] The method or system according to the embodiment of the present application can also be used by Figure 5 The electronic device architecture shown in FIG. Figure 4 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, ROM 503, RAM 504, a communication port 505 connected to a network, an input / output 506, a hard disk 507, etc. The storage device in the electronic device 500, such as ROM 503 or hard disk 507, may store the method for adaptively adjusting the temperature of the organic silicon auxiliary synthesis reaction provided in the present application. Furthermore, the electronic device 500 may also include a user interface 508. Of course, Figure 4The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 4 One or more components of an electronic device are shown.

[0213] Embodiment 5:

[0214] See also Figure 5 As shown, a computer-readable storage medium 600 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are executed by the processor, the method for adaptively adjusting the temperature of the organic silicon auxiliary agent synthesis reaction according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0215] In addition, according to the embodiments of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: a method for adaptively adjusting the temperature of a synthesis reaction of an organosilicon additive. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0216] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0217] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for adaptively adjusting the reaction temperature of a silicone additive synthesis reaction, characterized in that: include: Collecting raw material data; the raw material data includes raw material type and raw material proportion; Analyze the raw material data and retain m temperature curves from a preset curve set, where m is an integer greater than 1; the method for retaining m temperature curves from a preset curve set includes: The curve set includes M temperature curves. ; Obtain the maximum and minimum values ​​of the M temperature curves in sequence, subtract the corresponding minimum values ​​from the maximum values ​​of the M temperature curves, and obtain the extreme difference values ​​corresponding to the M temperature curves; Get the number of raw material types, which is the number of raw material types; Calculate the temperature range based on raw material data; The expression of temperature extreme difference is: ; In the formula, For extreme temperature differences, is the raw material type value of the i-th raw material, is the raw material proportion of the i-th raw material, is the weight coefficient of the i-th raw material, sec is the secant function, , n is the number of raw material types; Subtract the temperature extreme difference from the extreme difference values ​​corresponding to the M temperature curves and take the absolute value to obtain the difference; sort the difference values ​​from small to large, and obtain the values ​​in the front in positive order. The difference value is retained from the curve set There are m temperature curves corresponding to the difference of ; The raw material data is analyzed again to obtain the optimal temperature curve in the synthesis process of the organosilicon additive, and the organosilicon additive is synthesized according to the optimal temperature curve; the step of obtaining the optimal temperature curve in the synthesis process of the organosilicon additive includes: Step 1: Preset the iteration threshold and set digital labels for m temperature curves in order. The digital label range is , mark the digital label corresponding to the temperature curve as the curve label; randomly generate an initial antibody population of size q ; Step 2: Determine the fitness function; Step 3: Calculate the initial antibody population The fitness of each antibody in , select r antibodies with the largest fitness, and construct a memory population ; Step 4: Memory population Each antibody in the cloning is cloned according to the calculated clone size to generate a clonal antibody population ; Step 5: Cloning of Antibody Populations Each antibody in the cloud uses a cloud adaptive mutation operator to perform cloud adaptive mutation to generate a mutant antibody group ; Step 6: Cloning antibody populations Randomly select w antibodies as parent antibodies for antibody recombination to generate offspring antibodies , until the clonal antibody population When the number of unselected antibodies is less than w, antibody recombination is stopped and a recombinant antibody group is generated. ; Step 7: Merge the variant antibody groups and recombinant antibody groups , generating a pooled antibody population ; Step 8: Randomly generate d new antibodies and merge the antibody groups The d antibodies with the smallest fitness are replaced by d new antibodies generated randomly; Step 9: If the number of cycles t is less than the iteration threshold, let , and select the combined antibody group The r antibodies with the largest fitness form a new memory population , loop steps 3 to 8; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group is obtained The temperature curve corresponding to the antibody with the largest fitness is taken as the optimal temperature curve; In step 1, the initial antibody population The antibodies in the curve label correspond one to one, and the range of the antibody is the range of the curve label, that is, the range of the antibody is ; In step 2, the expression of the fitness function is: ; In the formula, is the fitness of the jth antibody, is the product mass of the jth antibody, ; The mass of the product of the jth antibody is the mass of the organosilicon additive synthesized under the temperature curve conditions corresponding to the jth antibody; Methods for obtaining product quality include: The raw material type value and raw material ratio are used as updated raw material data, and the curve label corresponding to the antibody and the updated raw material data are used as analysis data; the analysis data is input into the trained quality prediction model to predict the corresponding product quality; The training process of the quality prediction model includes: Pre-collecting N groups of analytical data corresponding to the product masses, and converting the analytical data and the corresponding product masses into a corresponding set of feature vectors; Each group of feature vectors is used as input of a quality prediction model. The quality prediction model uses a group of product masses corresponding to each group of analysis data as output, and uses the actual product mass corresponding to each group of analysis data as a prediction target, where the actual product mass is the mass of the product corresponding to the analysis data collected in advance; minimizing the sum of prediction errors of all analysis data is used as a training target; the quality prediction model is trained until the sum of prediction errors reaches convergence and the training is stopped; the quality prediction model is a deep neural network model; In step 4, the method for calculating the clone size includes: ; ; In the formula, is the cloning scale of the jth antibody, Int is the upward rounding function, q is the size of the antibody group, is the affinity of the j-th antibody, min is the minimum function, exp is the natural exponential function, is the Euclidean distance between the j-th antibody and the e-th antibody, , ; In step 6, the progeny antibody The calculation methods include: ; In the formula, is the w-th parent antibody, is the wth scale factor, the scale factor is a randomly generated real number and the w scale factors are not all 0; Collect real-time data during the synthesis of silicone additives; Analyze real-time data and predict future temperature data; Adjust the optimal temperature curve according to future temperature data; Get the time point corresponding to the future temperature data and mark it as the adjustment time point; Analyze future temperature data, obtain multi-point temperature data, and adjust the optimal temperature curve based on the multi-point temperature data; Methods for obtaining multi-point temperature data include: Subtract the temperature value corresponding to the adjustment time point from the future temperature data to obtain the temperature error; preset a demarcation value, divide the temperature error by the demarcation value to obtain the temperature ratio, and use the integer part of the temperature ratio as the number of adjustment points, which is the number of time points that need to be adjusted in addition to the adjustment time point; Marking additional time points that need adjustment as additional adjustment points; the multi-point temperature data includes temperature data of each additional adjustment point; The calculation method of the temperature data corresponding to each additional adjustment point includes: ; In the formula, is the temperature data of the pth additional adjustment point, is the proportional coefficient of the pth additional adjustment point, is the temperature error, is the temperature value corresponding to the pth additional adjustment point in the optimal temperature curve, To adjust the points, ; The optimal temperature curve The temperature values ​​corresponding to the additional adjustment points are adjusted in turn to the multi-point temperature data Temperature data corresponding to the additional adjustment points.

2. The method for adaptively adjusting the reaction temperature of a silicone additive synthesis reaction according to claim 1, characterized in that: The raw material type is the type of various raw materials required in the synthesis process of the organosilicon additive; the raw material proportion is the mass proportion of various raw materials in the synthesis process of the organosilicon additive.

3. The method for adaptively adjusting the temperature of the organic silicon auxiliary agent synthesis reaction according to claim 2, characterized in that: The real-time data includes temperature value, pressure value, pH value and synthesis time; the temperature value is the temperature value in the reactor during the synthesis of the organosilicon additive, and the reactor is a device for synthesizing the organosilicon additive; the pressure value is the pressure value in the reactor during the synthesis of the organosilicon additive; the pH value is the pH value of the reaction liquid during the synthesis of the organosilicon additive; the synthesis time is the length of time for the synthesis of the organosilicon additive.

4. The method for adaptively adjusting the reaction temperature of a silicone additive synthesis according to claim 3, characterized in that: The method for predicting future temperature data comprises: Real-time data and updated raw material data are used as test data, and the test data are input into the trained temperature prediction model to predict future temperature data; the future temperature data is the temperature value corresponding to the next time point of the synthesis time; the training process of the temperature prediction model is consistent with the training process of the quality prediction model, and it is also a deep neural network model.

5. The method for adaptively adjusting the reaction temperature of a silicone additive synthesis according to claim 4, characterized in that: The temperature value corresponding to the adjustment time point in the optimal temperature curve is adjusted to the future temperature data.

6. A temperature adaptive adjustment system for a synthesis reaction of an organosilicon additive, which implements the temperature adaptive adjustment method for a synthesis reaction of an organosilicon additive according to any one of claims 1 to 5, characterized in that: include: Raw material data collection module, used for collecting raw material data; A raw material data analysis module is used to analyze the raw material data and retain m temperature curves from a preset curve set, where m is an integer greater than 1; The secondary analysis module is used to analyze the raw material data again, obtain the optimal temperature curve in the synthesis process of the organosilicon additive, and synthesize the organosilicon additive according to the optimal temperature curve; Real-time data acquisition module, used to collect real-time data during the synthesis process of organosilicon additives; Real-time data analysis module, used to analyze real-time data and predict future temperature data; The temperature adjustment module is used to adjust the optimal temperature curve according to future temperature data.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for adaptively adjusting the temperature of the organosilicon additive synthesis reaction according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method for adaptively adjusting the temperature of the organosilicon additive synthesis reaction according to any one of claims 1 to 5 is implemented.