An online quantitative mixing system for wet chemicals suitable for the new materials industry

Abstract

The present invention discloses an online quantitative mixing system for wet chemicals suitable for the new materials industry, which relates to the field of chemical mixing technology, including a formula calculation and raw material preparation module, an initial addition and mixing control module, a feature extraction and data analysis module, a change classification module, a normal change processing module, an obvious change adjustment module and an insignificant change adjustment module; the formula calculation and raw material preparation module prepares various chemicals according to a pre-set formula and stores them in different raw material tanks. The present invention, by combining a machine learning model, predicts and adjusts the changes of chemical reagents in real time, and flexibly responds to different changes in the mixing process. When there is a significant change, the addition amount is reduced, and when the change is not obvious, the addition amount is increased. Under normal circumstances, the initial trace amount is maintained, thereby improving the mixing accuracy and efficiency, avoiding violent reactions and delays, reducing energy and time costs, and optimizing production efficiency and product quality control.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical mixing, and in particular to an online quantitative mixing system for wet chemicals applicable to the new materials industry. Background Art

[0002] The online wet chemical quantitative mixing system, suitable for the new materials industry, monitors and controls the automated mixing of multiple chemicals in real time, achieving precise proportioning through the integration of a concentration meter, electronic scale, and PLC controller. The system collects data such as chemical concentration and weight in real time and transmits this data to the PLC for calculation and control, ensuring that various media are accurately injected into the mixing tank according to the predetermined formula ratio. In the online wet chemical quantitative mixing system, "quantity" refers to the precise control of the injection amount of each chemical, ensuring that it is mixed according to the predetermined formula ratio and specific value.

[0003] When mixing chemical reagents, different raw materials need to be added to the mixing tank for mixing. The existing wet chemical online quantitative mixing system usually uses a single quantitative addition of chemicals for mixing, and is unable to intelligently adjust the amount of chemicals added according to the actual mixing situation. If the chemical reagents change significantly after a single addition of chemicals, continuing to use this method may trigger a violent chemical reaction, resulting in a runaway reaction. Such drastic changes will cause the product ratio to be unbalanced, seriously deviate from the predetermined standards, resulting in uneven mixing, and thus making the final product quality fail to meet the standards or even be completely unqualified. On the contrary, if the chemical reagents change very little after a single addition of chemicals, continuing to use the quantitative addition method will result in insignificant changes in concentration, and multiple additions will be required to achieve the desired effect, thereby prolonging the mixing process. This delay will not only significantly reduce production efficiency, but also increase production time and energy consumption, ultimately leading to a substantial increase in system operating costs.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide an online quantitative mixing system for wet chemicals suitable for the new materials industry. By combining a machine learning model, the system can predict and adjust the changes in chemical reagents in real time, so that the system can flexibly respond to different changes in the mixing process. When obvious changes occur, the system reduces the amount of chemicals added to avoid excessive reactions; when the changes are not obvious, the amount added is appropriately increased to speed up the reaction process; in the case of normal changes, the initial trace amount added is maintained to ensure stable production. This not only effectively improves the flexibility and accuracy of the mixing process, reduces overreactions and production delays, but also reduces energy and time costs, optimizes system efficiency and product quality control, and solves the problems in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: an online quantitative mixing system for wet chemicals suitable for the new materials industry, comprising a formula calculation and raw material preparation module, an initial addition and mixing control module, a feature extraction and data analysis module, a change classification module, a normal change processing module, an obvious change adjustment module, and an insignificant change adjustment module;

[0007] The formula calculation and raw material preparation module prepares various chemicals according to the pre-set formula and stores them in different raw material tanks. It calculates the required amount of each chemical based on production needs and the formula and determines the amount of each raw material to be added.

[0008] The initial addition and mixing control module uploads the set initial addition amount information to the PLC, opens and closes the valves in each raw material tank according to the control instructions of the PLC, and accurately injects each chemical into the mixing tank according to the initial addition amount to achieve initial mixing;

[0009] The feature extraction and data analysis module, after the initial mixing is completed, accurately injects each chemical into the mixing tank according to the initial trace amount. After the chemical injection, the chemical reagent change information after two adjacent chemical injections is obtained to establish an analysis set. Feature extraction is performed on each analysis set. After analyzing the extracted features, predictions are made using a pre-trained machine learning model to obtain the chemical reagent change after the second chemical injection;

[0010] The change classification module classifies the changes of current chemical reagents into three types based on the prediction results of the machine learning model: obvious changes, normal changes, and non-obvious changes;

[0011] Normal change processing module, for normal change situations, continues to accurately inject each chemical into the mixing tank according to the initial trace amount;

[0012] The significant change adjustment module dynamically adjusts the amount of chemicals added in the next round based on the change information of the chemical reagents after two consecutive injections in the case of significant changes. Specifically, it reduces the amount of chemicals added in the next round based on the initial trace amount.

[0013] The non-obvious change adjustment module dynamically adjusts the amount of chemicals added in a new round based on the change information of the chemical reagents after two adjacent injections in the case of non-obvious changes. Specifically, it increases the amount of chemicals added in a new round based on the initial trace amount.

[0014] Preferably, the initial addition amount refers to the amount of each chemical that needs to be accurately injected into the mixing tank for the first time during the chemical mixing process, based on the pre-set production formula and actual production needs. The initial addition amount meets the proportional relationship and total amount requirements of the chemicals in the formula to ensure that the mixed chemicals can achieve the expected concentration and performance indicators.

[0015] Preferably, the initial trace amount refers to adding a small amount of chemicals again accurately according to a set small dose after the initial mixing is completed, the purpose of which is to observe the changes of the chemical reagents after the trace amount of chemicals is added.

[0016] Preferably, feature extraction is performed on each analysis set, and the extracted features include features reflecting the changes in heat release during the chemical reaction and the changes in the electrical conductivity of the chemical reagents. After analyzing the changes in heat release during the chemical reaction and the changes in the electrical conductivity of the chemical reagents, a heat release reference value and a conductivity change reference value are generated respectively. The heat release reference value is used to quantify the intensity of heat release during the chemical reaction, and the conductivity change reference value is used to quantify the changes in the electrical conductivity of the reagents during the chemical reaction, reflecting the changes in the concentration of ions in the chemical reagents or the dissociation of electrolytes in the solution.

[0017] Preferably, after obtaining the heat release reference value and the conductivity change reference value generated by analyzing the extracted features, the heat release reference value and the conductivity change reference value are input into a pre-trained machine learning model to generate a reagent change coefficient, and the reagent change coefficient is used to predict the change of the chemical reagent after the chemical is injected again.

[0018] Preferably, the characteristics extracted from each analysis set are analyzed and the reagent variation coefficient generated by the machine learning model is predicted. Reference threshold range with pre-set reagent variation coefficient A comparative analysis was performed, in which and They represent the minimum and maximum values ​​of the reference threshold range of the reagent variation coefficient, respectively. The changes of chemical reagents are divided into three types. The specific division process is as follows:

[0019] If the reagent variation coefficient is greater than the maximum value of the reagent variation coefficient reference threshold range, that is, , then the change of the current chemical reagent is classified as an obvious change;

[0020] If the reagent variation coefficient is within the reagent variation coefficient reference threshold range, that is, , then the current change of chemical reagents is classified as normal change;

[0021] If the reagent variation coefficient is less than the minimum value of the reagent variation coefficient reference threshold range, that is, , the change of the current chemical reagent is classified as a non-significant change.

[0022] Preferably, the specific steps of generating a heat release reference value after analyzing the change in heat release during a chemical reaction are as follows:

[0023] The temperature change data after two consecutive chemical injections during the chemical reaction are collected by sensors. Based on the time change of temperature, the heat release rate at each time point is preliminarily calculated. The calculation expression is: , where represents the instantaneous heat released in the chemical reaction at time t, is the thermal conductivity coefficient, is the specific heat capacity of the reactants, m is the mass of the reactants, It indicates the rate of change of temperature relative to time, reflecting the rate of temperature change;

[0024] After calculating the heat release rate, the nonlinear behavior of the heat release change is calculated. Specifically, the acceleration of the heat release rate, that is, the second-order derivative of the heat release change, is analyzed to quantify the heat release acceleration. The specific calculation expression is: , where represents the acceleration of heat release in the chemical reaction at time t, is a correction coefficient used to adjust the amplitude of heat release acceleration under different reaction conditions. It is the second derivative of the heat release rate, which indicates the acceleration of heat release, that is, the intensity of heat change;

[0025] According to the nonlinear change of heat release acceleration, a heat release reference value is generated to quantify the intensity of heat release during the entire chemical reaction process. The expression for generating the heat release reference value is: , where is the adjustment coefficient, which is used to normalize the heat release intensity under different reaction conditions. Indicates from time arrive The cube of the heat release acceleration is integrated to capture the nonlinear characteristics of the heat release change. is the initial time point after the chemical is injected, indicating the time when the heat release change begins. is the end time point of the chemical reaction, indicating the time when the heat release change ends. Indicates the heat release reference value, which is used to quantify the heat release intensity.

[0026] Preferably, the specific steps of generating a conductivity change reference value after analyzing the conductivity change of the chemical reagent are as follows:

[0027] Calculate the initial difference in conductivity of the chemical reagent after two adjacent chemical injections. The conductivity difference is used to preliminarily capture the conductivity change trend. The calculation expression is: , where is the initial difference in conductivity, reflecting the change in conductivity after two adjacent chemical injections. and are the current and last measured conductivity of the chemical reagent, is the fine-tuning factor of the resistance change rate;

[0028] According to the change of conductivity, the ion concentration change factor is further introduced to quantify the ion generation, dissociation or dissolution process in the chemical reagent. It is accurately described by the nonlinear relationship between ion migration time and molar conductivity coefficient. The calculation expression of the ion concentration change factor is: , where is the ion concentration variation factor, reflecting the effect of ion generation or dissociation, is the correction coefficient for the total polarization effect of the solution, and are the migration time of positive and negative ions, respectively, reflecting the migration rate of ions in the solution. and are the molar conductivity coefficients of positive and negative ions, respectively, and V is the total volume of the mixing tank, which is used to standardize the effect of conductivity changes on the total amount of solution. is the nonlinear amplification factor, which is used to strengthen the relationship between conductivity change and ion concentration;

[0029] Combining the initial conductivity difference and the ion concentration change factor, a conductivity change reference value is generated to fully quantify the dramatic change in conductivity during the chemical reaction. The expression for generating the conductivity change reference value is: , where is the reference value of conductivity change, The effective resistance is the comprehensive resistance after taking into account the influence of solution resistance.

[0030] Preferably, the specific steps for dynamically adjusting the amount of chemicals added in a new round according to the chemical reagent change information after two adjacent injections are as follows:

[0031] Determine the adjustment factor for the next round of addition based on the state of reagent changes , adjustment factor Adjustment is made based on the difference between the reagent variation coefficient and the reagent variation coefficient reference threshold range. The specific expression is: , where is the adjustment factor, used to adjust the amount of addition in the new round, and They are the reduction factor and the increase factor, which are used to control the reduction range in obvious changes and the increase range in non-obvious changes respectively;

[0032] Based on the adjustment factor , dynamically adjust the amount of the new round of additions. For obvious changes, the amount of addition will be reduced; for non-obvious changes, the amount of addition will be increased; for normal changes, it will remain unchanged. The adjustment expression is: , where is the amount of chemicals added in the next round, For normal changes, continue to accurately inject various chemicals according to the set initial trace amount;

[0033] Based on the calculated amount of chemicals to be added for the next round , accurately inject a new round of chemicals into the mixing tank. After the injection is completed, continue to monitor the change information of the chemical reagents after two adjacent injections, and perform the next round of feature extraction and adjustment based on the system feedback to maintain the stability and control accuracy of the chemical reaction process.

[0034] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0035] The present invention precisely injects chemicals into the mixing tank according to a preset formula, and combines machine learning models to predict and adjust changes in chemical reagents in real time, enabling the system to flexibly respond to different changes in the mixing process. When significant changes occur, the system automatically reduces the amount of chemicals added in the new round to avoid overly violent reactions; when changes are not significant, the system appropriately increases the amount added to speed up the reaction process; and in the case of normal changes, the system maintains the initial trace amount added to ensure stable production. This not only improves the flexibility and accuracy of the mixing process, avoiding violent reactions and production delays, but also effectively reduces energy and time costs, optimizes system operating efficiency and product quality control, and provides a more efficient and intelligent solution for the production process of the new materials industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0037] Figure 1 The module diagram of an online quantitative mixing system for wet chemicals applicable to the new materials industry is shown in the figure. DETAILED DESCRIPTION

[0038] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0039] The present invention provides Figure 1 The system is an online quantitative mixing system for wet chemicals suitable for the new materials industry, comprising a formula calculation and raw material preparation module, an initial addition and mixing control module, a feature extraction and data analysis module, a change classification module, a normal change processing module, a significant change adjustment module, and a non-significant change adjustment module.

[0040] The formula calculation and raw material preparation module prepares various chemicals according to the pre-set formula and stores them in different raw material tanks. It calculates the required amount of each chemical based on production needs and the formula and determines the amount of each raw material to be added.

[0041] The initial addition and mixing control module uploads the set initial addition amount information to the PLC, opens and closes the valves in each raw material tank according to the control instructions of the PLC, and accurately injects each chemical into the mixing tank according to the initial addition amount to achieve initial mixing;

[0042] The PLC is the core control unit of the entire system, responsible for managing and coordinating the valve operations in each raw material tank to precisely control the addition of chemicals. Through the PLC's logical control, each chemical is injected into the mixing tank at the exact initial amount.

[0043] The initial addition amount refers to the amount of each chemical that needs to be accurately injected into the mixing tank for the first time during the chemical mixing process, based on the pre-set production formula and actual production needs. The initial addition amount must meet the proportional relationship and total amount requirements of the chemicals in the formula to ensure that the mixed chemicals can achieve the expected concentration and performance indicators.

[0044] The initial addition amount complies with the required proportions and total amount of each chemical in the formula, ensuring that the mixed chemicals initially achieve the desired concentration and performance indicators. Therefore, by accurately controlling the initial addition amount, a preliminary test of the mixing effect can be performed. If the concentration, proportion, and other parameters of the mixture meet expectations after the initial addition, the mixing process will proceed smoothly. If there are any deviations, subsequent adjustment steps can be used to correct them and further optimize the mixing effect.

[0045] The feature extraction and data analysis module, after the initial mixing is completed, accurately injects each chemical into the mixing tank according to the initial trace amount. After the chemical injection, the chemical reagent change information after two adjacent chemical injections is obtained to establish an analysis set. Feature extraction is performed on each analysis set. After analyzing the extracted features, predictions are made using a pre-trained machine learning model to obtain the chemical reagent change after the second chemical injection;

[0046] Initial trace addition refers to the precise addition of a small amount of chemicals at a smaller, pre-set dosage after initial mixing. The goal is to observe how the chemical reacts to the trace addition. This small addition helps the operator or system monitor the reaction trends of the mixture and ensure that the chemical reaction proceeds within the expected range. By observing the chemical reactions, it is possible to determine whether further adjustments to the chemical dosage are needed to optimize the mixing effect and ensure the quality and performance of the final product. This process provides a basis for subsequent precise adjustments, avoiding runaway reactions or mixing failures caused by excessive or insufficient chemical addition.

[0047] Feature extraction is performed on each analysis set. The extracted features include features reflecting the changes in heat release during the chemical reaction and the changes in the electrical conductivity of the chemical reagents. After analyzing the changes in heat release during the chemical reaction and the changes in the electrical conductivity of the chemical reagents, a heat release reference value and a conductivity change reference value are generated respectively. The heat release reference value is used to quantify the intensity of heat release during the chemical reaction, and the conductivity change reference value is used to quantify the changes in the electrical conductivity of the reagents during the chemical reaction, reflecting the changes in the concentration of ions in the chemical reagents or the dissociation of electrolytes in the solution.

[0048] After obtaining the heat release reference value and conductivity change reference value generated by analyzing the extracted features, the heat release reference value and the conductivity change reference value are input into a pre-trained machine learning model to generate a reagent change coefficient. The reagent change coefficient is used to predict the change of chemical reagents after the chemicals are injected again.

[0049] After chemical injection, if the heat release of the current chemical reagent increases significantly compared to the previous injection, this generally indicates a significant change in the chemical reagent. A significant increase in heat release indicates a stronger chemical reaction, possibly due to increased reactant concentration, faster reaction rate, or the introduction of a new reaction mechanism. This significant heat change often indicates a significant shift in key reaction conditions, such as material conversion and energy release, during the chemical reaction, and is therefore a key indicator of a significant change in the chemical reagent.

[0050] The specific steps for generating a heat release reference value after analyzing the changes in heat release during a chemical reaction are as follows:

[0051] The temperature change data after two consecutive chemical injections during the chemical reaction are collected by sensors. Based on the time change of temperature, the heat release rate at each time point is preliminarily calculated. The calculation expression is: , where represents the instantaneous heat released in the chemical reaction at time t, is the thermal conductivity coefficient, which reflects the heat transfer efficiency in the system. is the specific heat capacity of the reactants, which indicates the ability of a unit mass of a substance to change temperature when absorbing or releasing heat. m is the mass of the reactants, which affects the overall amount of heat released. It indicates the rate of change of temperature relative to time, reflecting the rate of temperature change;

[0052] After calculating the heat release rate, the nonlinear behavior of the heat release change is calculated. Specifically, the acceleration of the heat release rate, that is, the second-order derivative of the heat release change, is analyzed to quantify the heat release acceleration. The specific calculation expression is: , where It represents the acceleration of heat release in the chemical reaction at time t, reflecting the intensity of the change in the heat release rate. is a correction coefficient used to adjust the amplitude of heat release acceleration under different reaction conditions. It is the second derivative of the heat release rate, which indicates the acceleration of heat release, that is, the intensity of heat change;

[0053] According to the nonlinear change of heat release acceleration, a heat release reference value is generated to quantify the intensity of heat release during the entire chemical reaction process. The expression for generating the heat release reference value is: , where is the adjustment coefficient, which is used to normalize the heat release intensity under different reaction conditions. Indicates from time arrive The cube of the heat release acceleration is integrated to capture the nonlinear characteristics of heat release changes and amplify the impact of drastic heat changes. The initial time point after the chemical is injected, indicating the time when the heat release change begins. It is usually the moment when the chemical reaction starts or a significant change occurs. The end time point of the chemical reaction, indicating the time when the heat release change ends, that is, the moment when the reaction reaches a stable or end state. Indicates the heat release reference value, which is used to quantify the heat release intensity.

[0054] The heat release reference value, generated by analyzing the changes in heat release during a chemical reaction, shows that a larger heat release reference value indicates a more dramatic change in heat release during the reaction, indicating a significant reaction or conversion of the chemical reagents, with both the intensity and degree of change being noticeable. Conversely, a smaller heat release reference value indicates a more gradual change in heat release, with no noticeable changes in the reagents during the reaction, and a relatively mild or stable reaction. Therefore, the heat release reference value can be used as an important indicator for evaluating chemical reaction intensity and reagent changes.

[0055] If the conductivity of a chemical reactant changes significantly after chemical injection compared to the previous one, this typically indicates a significant change in the chemical's composition or state. Changes in conductivity are often related to changes in ion concentrations in the solution, the products produced by chemical reactions, and ion dissociation. A significant change in conductivity may indicate that a chemical reaction is generating new ions or that the concentration of existing ions has significantly changed, indicating a significant chemical or physical change in the chemical reactant compared to the previous one.

[0056] The specific steps for generating a conductivity change reference value after analyzing the conductivity change of a chemical reagent are as follows:

[0057] Calculate the initial difference in conductivity of the chemical reagent after two adjacent chemical injections. The conductivity difference is used to preliminarily capture the conductivity change trend. The calculation expression is: , where is the initial difference in conductivity, reflecting the change in conductivity after two adjacent chemical injections. and The conductivity of the chemical reagent measured at the current (nth time) and the last (n-1th time) time, It is a fine-tuning factor for the resistance change rate, used to correct conductivity fluctuations caused by external factors (such as temperature and pressure) and enhance calculation accuracy;

[0058] According to the change of conductivity, the ion concentration change factor is further introduced to quantify the ion generation, dissociation or dissolution process in the chemical reagent. It is accurately described by the nonlinear relationship between ion migration time and molar conductivity coefficient. The calculation expression of the ion concentration change factor is: , where is the ion concentration variation factor, reflecting the effect of ion generation or dissociation, is the correction coefficient for the total polarization effect of the solution (since during the electrochemical reaction, the electrodes in the solution may be polarized due to charge accumulation, resulting in a deviation between the actual and theoretical values ​​of conductivity. This correction coefficient corrects the measured conductivity data by considering the accumulation and discharge process of charge on the electrode surface to ensure a more accurate calculation of the conductivity change, thereby better reflecting the actual impact of ion generation, dissociation and migration in chemical reagents on conductivity). It takes into account the electrode effect and interface effect of the solution during the reaction process. and are the migration time of positive and negative ions, respectively, reflecting the migration rate of ions in the solution. and are the molar conductivity coefficients of positive and negative ions, respectively (the molar conductivity coefficients are usually obtained through experimental measurements or reference data), which describe the conductivity of ions. V is the total volume of the mixing tank, which is used to standardize the effect of conductivity changes on the total amount of solution. is the nonlinear amplification factor, which is used to strengthen the relationship between conductivity change and ion concentration;

[0059] Combining the initial conductivity difference and the ion concentration change factor, a conductivity change reference value is generated to fully quantify the dramatic change in conductivity during the chemical reaction. The expression for generating the conductivity change reference value is: , where is the reference value of conductivity change, Effective resistance refers to the comprehensive resistance after accounting for the influence of solution resistance (or electrolyte resistance). This includes not only the resistance between the electrode and the electrolyte, but also the effects of factors such as the solution's conductivity, ion concentration, and ion migration rate. Effective resistance reflects the actual resistance value in conductivity measurements, eliminating errors caused by polarization effects or other non-ideal factors. This more accurately characterizes the relationship between conductivity changes and resistance during chemical reactions.

[0060] As can be seen from the conductivity change reference value, the larger the value of the conductivity change reference value generated after analyzing the conductivity change of the chemical reagent, the more obvious the change in the current chemical reagent. This is because the conductivity change reference value reflects the changes in ion concentration, migration rate, and other electrochemical parameters of the chemical reagent during the reaction process. When the conductivity change reference value is large, it means that the conductivity has fluctuated significantly, indicating that the state or composition of the chemical reagent has undergone significant changes, such as ion formation, dissociation, or migration. Conversely, if the conductivity change reference value is small, it means that the conductivity of the chemical reagent changes more slowly, indicating that the composition or reaction state of the reagent has not changed significantly and the system is in a relatively stable state.

[0061] A pre-trained machine learning model is one that's trained using extensive historical data from chemical reagent analysis. This data includes characteristics such as changes in conductivity and heat release during chemical reactions. By feeding this historical data into a machine learning algorithm, the model learns the chemical reagent's behavior under different reaction conditions, the complex relationships between these characteristics, and how these characteristics influence the reaction's outcome. Through multiple iterations and optimization, this model can predict chemical reaction behavior under specific conditions, such as ion dissociation, heat release, and concentration changes. Once trained, the model can be applied to new chemical reaction prediction tasks.

[0062] In practical applications, pre-trained models can be used to predict potential changes in chemical reagents during future reactions by inputting new feature data (such as reference values ​​for conductivity change and heat release). Because the model has been optimized based on extensive historical data, it can accurately predict chemical reaction trends, helping researchers or production systems detect potential problems or abnormal reactions in advance.

[0063] The machine learning model is not specifically limited here, and can achieve the reference value of heat release and conductivity change reference value Perform comprehensive analysis to generate reagent variation coefficients In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; the reagent variation coefficient The resulting calculation formula is: , where 、 Heat release reference values and conductivity change reference value The preset scaling factor of 、 Both are greater than 0.

[0064] It can be seen from the reagent variation coefficient that the greater the performance value of the heat release reference value generated after analyzing the change in heat release during the chemical reaction, the greater the performance value of the conductivity change reference value generated after analyzing the conductivity change of the chemical reagent. That is, the greater the performance value of the reagent variation coefficient predicted by the machine learning model after analyzing the features extracted from each analysis set, the more obvious the change in the current chemical reagent is, and vice versa.

[0065] The change classification module classifies the changes of current chemical reagents into three types based on the prediction results of the machine learning model: obvious changes, normal changes, and non-obvious changes;

[0066] The features extracted from each analysis set will be analyzed and the reagent variation coefficient generated will be predicted by the machine learning model Reference threshold range with pre-set reagent variation coefficient A comparative analysis was performed, in which and They represent the minimum and maximum values ​​of the reference threshold range of the reagent variation coefficient, respectively. The changes of chemical reagents are divided into three types. The specific division process is as follows:

[0067] If the reagent variation coefficient is greater than the maximum value of the reagent variation coefficient reference threshold range, that is, If the chemical reagent changes significantly, the chemical reagent's state is classified as a significant change. This indicates a dramatic change in the chemical reagent's state. Such changes typically indicate a significant increase in reaction rate, significant heat release, or ion generation during the reaction. This may indicate that the chemical reaction has entered a critical or unstable phase, requiring close monitoring by operators to ensure reaction safety and product quality.

[0068] If the reagent variation coefficient is within the reagent variation coefficient reference threshold range, that is, , the current chemical reagent changes are classified as normal changes. At this point, the chemical reaction proceeds as planned, and parameters such as reagent concentration and reaction rate are stable, indicating that the reaction system is in equilibrium. This indicates that the reaction progress meets the design requirements and no special adjustments to the reaction conditions are required, and the system can maintain normal operation.

[0069] If the reagent variation coefficient is less than the minimum value of the reagent variation coefficient reference threshold range, that is, , the change of the current chemical reagent is classified as a non-significant change. This situation usually reflects that the chemical reaction progresses slowly and the concentration change of the reagent is insufficient, which may mean that the reaction conditions are not ideal and the reaction rate is lower than expected.

[0070] Normal change processing module, for normal change situations, continues to accurately inject each chemical into the mixing tank according to the initial trace amount;

[0071] The significant change adjustment module dynamically adjusts the amount of chemicals added in the next round based on the change information of the chemical reagents after two consecutive injections in the case of significant changes. Specifically, it reduces the amount of chemicals added in the next round based on the initial trace amount.

[0072] The non-obvious change adjustment module dynamically adjusts the amount of chemicals added in the next round based on the change information of the chemical reagents after two consecutive injections. Specifically, it increases the amount of chemicals added in the next round based on the initial trace amount.

[0073] This ensures that the amount of chemicals added can be flexibly adapted to different needs under changing mixing conditions;

[0074] Normal variation: When chemical reagent changes remain within the expected normal range, the reaction process is stable, and the reagent concentration and reaction rate meet design requirements. In this case, no adjustments to the chemical dosage are required; continue to precisely inject each chemical according to the set initial micro-amounts. This indicates that the system is operating smoothly, and maintaining the current formulation ratios and injection volumes ensures smooth chemical reactions and the desired mixing results.

[0075] Significant changes: When chemical reagents undergo significant changes, exceeding the preset reagent variation coefficient threshold, this indicates that the chemical reaction has become too intense, potentially leading to problems such as rapid ion formation and excessive heat release. In this case, to avoid runaway reactions or product quality issues, the amount of added chemicals must be dynamically adjusted. Specifically, based on the changes between two consecutive injections, the amount of chemicals added in the next round is reduced, essentially reducing the initial amount to moderate the reaction intensity and maintain system stability.

[0076] In the case of non-significant changes: When the change in chemical reagents is very slight, below the preset reagent variation coefficient threshold, it indicates that the reaction is progressing slowly, which may lead to insufficient chemical reaction or substandard product quality. In this case, the amount of chemical added needs to be dynamically increased. Specifically, the specific operation is to increase the amount of chemical added in the next round based on the change information between two consecutive injections, and appropriately increase the initial trace amount to promote the reaction, accelerate the change in reagent concentration, and ensure the chemical reaction is completed smoothly as expected.

[0077] The specific steps for dynamically adjusting the amount of chemicals added in a new round based on the chemical reagent change information after two consecutive injections are as follows:

[0078] Determine the adjustment factor for the next round of addition based on the state of reagent changes , adjustment factor Adjustment is made based on the difference between the reagent variation coefficient and the reagent variation coefficient reference threshold range. The specific expression is: , where is the adjustment factor, used to adjust the amount of addition in the new round, and They are the reduction factor and the increase factor, which are used to control the reduction range in obvious changes and the increase range in non-obvious changes respectively;

[0079] The reduction factor is a regulatory parameter used in situations of significant change, controlling the magnitude of the reduction in the amount of chemicals added in a new round. The reduction factor should be adjusted based on the severity of the specific chemical reaction and its impact on product quality. Generally, a larger reduction factor indicates that when the chemical reagent changes more significantly, the system needs to significantly reduce the amount of chemical added. This reduction factor can be set based on experimental or historical data and may be related to the sensitivity of the reaction system, reagent reaction rate, temperature changes, and other factors to avoid overreaction or reaction runaway. Therefore, the reduction factor is not specifically defined here.

[0080] The ramp factor applies to situations with less pronounced changes. This means that when the chemical reagent changes slowly, the system needs to increase the amount of the new chemical addition appropriately. The ramp factor setting can be adjusted based on the needs of the reaction process to ensure the reaction achieves the desired speed or effect. Generally, a larger ramp factor value increases the system's adjustment range, thereby accelerating the reaction process. The specific value of this ramp factor can be optimized experimentally, taking into account factors such as reagent concentration, temperature, and reactant consumption rate to ensure that the system accelerates the chemical reaction while maintaining stability. Therefore, the ramp factor is not specifically limited here.

[0081] Based on the adjustment factor , dynamically adjust the amount of the new round of additions. For obvious changes, the amount of addition will be reduced; for non-obvious changes, the amount of addition will be increased; for normal changes, it will remain unchanged. The adjustment expression is: , where is the amount of chemicals added in the next round, is the initial trace amount; for normal changes, it can be seen from the expression that the system will continue to accurately inject various chemicals according to the set initial trace amount;

[0082] Based on the calculated amount of chemicals to be added for the next round , accurately inject a new round of chemicals into the mixing tank. After the injection is completed, continue to monitor the change information of the chemical reagents after two adjacent injections, and perform the next round of feature extraction and adjustment based on the system feedback to maintain the stability and control accuracy of the chemical reaction process.

[0083] The present invention precisely injects chemicals into the mixing tank according to a preset formula, and combines machine learning models to predict and adjust changes in chemical reagents in real time, enabling the system to flexibly respond to different changes in the mixing process. When significant changes occur, the system automatically reduces the amount of chemicals added in the new round to avoid overly violent reactions; when changes are not significant, the system appropriately increases the amount added to speed up the reaction process; and in the case of normal changes, the system maintains the initial trace amount added to ensure stable production. This not only improves the flexibility and accuracy of the mixing process, avoiding violent reactions and production delays, but also effectively reduces energy and time costs, optimizes system operating efficiency and product quality control, and provides a more efficient and intelligent solution for the production process of the new materials industry.

[0084] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0085] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0086] It should be noted that, in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements that are inherent to such process, method, article or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0087] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0088] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0089] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0091] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0093] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An online quantitative mixing system for wet chemicals suitable for the new materials industry, characterized in that: It includes formula calculation and raw material preparation module, initial addition and mixing control module, feature extraction and data analysis module, change classification module, normal change processing module, obvious change adjustment module and non-obvious change adjustment module; The formula calculation and raw material preparation module prepares various chemicals according to the pre-set formula and stores them in different raw material tanks. It calculates the required amount of each chemical based on production needs and the formula and determines the amount of each raw material to be added. The initial addition and mixing control module uploads the set initial addition amount information to the PLC, opens and closes the valves in each raw material tank according to the control instructions of the PLC, and accurately injects each chemical into the mixing tank according to the initial addition amount to achieve initial mixing; The feature extraction and data analysis module, after the initial mixing is completed, accurately injects each chemical into the mixing tank according to the initial trace amount. After the chemical injection, the chemical reagent change information after two adjacent chemical injections is obtained to establish an analysis set. Feature extraction is performed on each analysis set. After analyzing the extracted features, predictions are made using a pre-trained machine learning model to obtain the chemical reagent change after the second chemical injection; The change classification module classifies the changes of current chemical reagents into three types based on the prediction results of the machine learning model: obvious changes, normal changes, and non-obvious changes; Normal change processing module, for normal change situations, continues to accurately inject each chemical into the mixing tank according to the initial trace amount; The significant change adjustment module dynamically adjusts the amount of chemicals added in the next round based on the change information of the chemical reagents after two consecutive injections in the case of significant changes. Specifically, it reduces the amount of chemicals added in the next round based on the initial trace amount. The non-obvious change adjustment module dynamically adjusts the amount of chemicals added in the next round based on the change information of the chemical reagents after two consecutive injections. Specifically, it increases the amount of chemicals added in the next round based on the initial trace amount. Feature extraction is performed on each analysis set. The extracted features include features reflecting changes in heat release and conductivity of chemical reagents during the chemical reaction. After analyzing the changes in heat release and conductivity of chemical reagents during the chemical reaction, a heat release reference value and a conductivity change reference value are generated respectively. The heat release reference value is used to quantify the intensity of heat release during the chemical reaction, and the conductivity change reference value is used to quantify the change in reagent conductivity during the chemical reaction, reflecting changes in ion concentration in the chemical reagent or the dissociation of electrolytes in the solution. After obtaining the heat release reference value and conductivity change reference value generated by analyzing the extracted features, the heat release reference value and conductivity change reference value are input into a pre-trained machine learning model to generate a reagent change coefficient. The reagent change coefficient is used to predict the change of chemical reagents after the chemical is injected again; Heat release reference value and conductivity change reference value Perform comprehensive analysis to generate reagent variation coefficients The calculation formula is: , where 、 Heat release reference values and conductivity change reference value The preset scaling factor of 、 All greater than 0; The specific steps for generating a conductivity change reference value after analyzing the conductivity change of a chemical reagent are as follows: Calculate the initial difference in conductivity of the chemical reagent after two adjacent chemical injections. The conductivity difference is used to preliminarily capture the conductivity change trend. The calculation expression is: , where is the initial conductivity difference, and are the current and last measured conductivity of the chemical reagent, is the fine-tuning factor of the resistance change rate; According to the change of conductivity, the ion concentration change factor is further introduced to quantify the ion generation, dissociation or dissolution process in the chemical reagent. The calculation expression of the ion concentration change factor is: , where is the ion concentration variation factor, is the correction coefficient for the total polarization effect of the solution, and are the migration times of positive and negative ions, respectively. and are the molar conductivity coefficients of positive and negative ions respectively, V is the total volume of the mixing tank, is the nonlinear amplification factor; Combining the initial conductivity difference and the ion concentration change factor, a conductivity change reference value is generated to fully quantify the dramatic change in conductivity during the chemical reaction. The expression for generating the conductivity change reference value is: , where is the reference value of conductivity change, is the effective resistance.

2. The online quantitative mixing system for wet chemicals suitable for the new materials industry according to claim 1, characterized in that: The initial addition amount refers to the amount of each chemical that needs to be accurately injected into the mixing tank for the first time during the chemical mixing process, based on the pre-set production formula and actual production needs. The initial addition amount meets the proportional relationship and total amount requirements of each chemical in the formula to ensure that the mixed chemicals can achieve the expected concentration and performance indicators.

3. The online quantitative mixing system for wet chemicals suitable for the new materials industry according to claim 1, characterized in that: Initial trace refers to the precise addition of a small amount of chemicals according to a set small dose after the initial mixing is completed. The purpose is to observe the changes in the chemical reagents after the trace amount of chemicals is added.

4. The online quantitative mixing system for wet chemicals suitable for the new materials industry according to claim 1, characterized in that: After obtaining the heat release reference value and conductivity change reference value generated by analyzing the extracted features, the heat release reference value and the conductivity change reference value are input into a pre-trained machine learning model to generate a reagent change coefficient. The reagent change coefficient is used to predict the change of chemical reagents after the chemicals are injected again.

5. The online quantitative mixing system for wet chemicals suitable for the new materials industry according to claim 1, characterized in that: The features extracted from each analysis set will be analyzed and the reagent variation coefficient generated will be predicted by the machine learning model Reference threshold range with pre-set reagent variation coefficient A comparative analysis was performed, in which and They represent the minimum and maximum values ​​of the reference threshold range of the reagent variation coefficient, respectively. The changes of chemical reagents are divided into three types. The specific division process is as follows: If the reagent variation coefficient is greater than the maximum value of the reagent variation coefficient reference threshold range, that is, , then the change of the current chemical reagent is classified as an obvious change; If the reagent variation coefficient is within the reagent variation coefficient reference threshold range, that is, , then the current change of chemical reagents is classified as normal change; If the reagent variation coefficient is less than the minimum value of the reagent variation coefficient reference threshold range, that is, , the change of the current chemical reagent is classified as a non-significant change.

6. The online quantitative mixing system for wet chemicals suitable for the new materials industry according to claim 1, characterized in that: The specific steps for generating a heat release reference value after analyzing the changes in heat release during a chemical reaction are as follows: The temperature change data after two consecutive chemical injections during the chemical reaction are collected by sensors. Based on the time change of temperature, the heat release rate at each time point is preliminarily calculated. The calculation expression is: , where represents the instantaneous heat released in the chemical reaction at time t, is the thermal conductivity coefficient, is the specific heat capacity of the reactants, m is the mass of the reactants, It indicates the rate of change of temperature relative to time, reflecting the rate of temperature change; After calculating the heat release rate, the nonlinear behavior of the heat release change is calculated. Specifically, the acceleration of the heat release rate, that is, the second-order derivative of the heat release change, is analyzed to quantify the heat release acceleration. The specific calculation expression is: , where represents the acceleration of heat release in the chemical reaction at time t, is a correction coefficient used to adjust the amplitude of heat release acceleration under different reaction conditions. It is the second derivative of the heat release rate, which indicates the acceleration of heat release, that is, the intensity of heat change; According to the nonlinear change of heat release acceleration, a heat release reference value is generated to quantify the intensity of heat release during the entire chemical reaction process. The expression for generating the heat release reference value is: , where is the adjustment coefficient, which is used to normalize the heat release intensity under different reaction conditions. Indicates from time arrive The cube of the heat release acceleration is integrated to capture the nonlinear characteristics of the heat release change. is the initial time point after the chemical is injected, indicating the time when the heat release change begins. is the end time point of the chemical reaction, indicating the time when the heat release change ends. Indicates the heat release reference value, which is used to quantify the heat release intensity.

7. The online quantitative mixing system for wet chemicals applicable to the new materials industry according to claim 5, characterized in that: The specific steps for dynamically adjusting the amount of chemicals added in a new round based on the chemical reagent change information after two consecutive injections are as follows: Determine the adjustment factor for the next round of addition based on the state of reagent changes , adjustment factor Adjustment is made based on the difference between the reagent variation coefficient and the reagent variation coefficient reference threshold range. The specific expression is: , where is the adjustment factor, used to adjust the amount of addition in the new round, and They are the reduction factor and the increase factor, which are used to control the reduction range in obvious changes and the increase range in non-obvious changes respectively; Based on the adjustment factor , dynamically adjust the amount of the new round of additions. For obvious changes, the amount of addition will be reduced; for non-obvious changes, the amount of addition will be increased; for normal changes, it will remain unchanged. The adjustment expression is: , where is the amount of chemicals added in the next round, For normal changes, continue to accurately inject various chemicals according to the set initial trace amount; Based on the calculated amount of chemicals to be added for the next round , accurately inject a new round of chemicals into the mixing tank. After the injection is completed, continue to monitor the change information of the chemical reagents after two adjacent injections, and perform the next round of feature extraction and adjustment based on the system feedback to maintain the stability and control accuracy of the chemical reaction process.

Images