A methyl methacrylate process production data processing system

Abstract

The present invention relates to the technical field of data processing, and in particular to a methyl methacrylate process production data processing system, the system comprising a data acquisition module for constructing a concentration curve and a temperature curve of methyl methacrylate; a data analysis module for obtaining an abnormal concentration sub-curve according to data fluctuations in the concentration curve, obtaining a temperature sub-curve corresponding to a time period where the abnormal concentration sub-curve is located in the temperature curve, obtaining an abnormal degree of inhibitor placement according to data in the abnormal concentration sub-curve and the temperature sub-curve, and obtaining an adjustment category of inhibitor placement dosage according to the abnormal degree of inhibitor placement; a model construction module for using concentration data and temperature data, and an adjustment category of inhibitor placement dosage as sample data, using multiple sample data to train an adjustment category recognition model, and adjusting the inhibitor placement dosage in real time according to the trained adjustment category recognition model to reduce losses and reduce costs.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a methyl methacrylate process production data processing system. Background Art

[0002] Methyl methacrylate (MMA) is an important organic chemical raw material, mainly used in the production of polymethyl methacrylate (PMMA) - organic glass. It is an important thermoplastic that was developed earlier. It has excellent transparency, chemical stability, weather resistance and other excellent properties, and has good market prospects. In the production process of methyl methacrylate, by monitoring and analyzing the numerical indicators of the MTBE cracking unit, oxidation absorption unit, refining unit, alcoholization unit and waste treatment unit in the MMA process flow, the production efficiency can be effectively improved, the cost can be reduced, the product quality can be guaranteed and the production safety can be increased. In recent years, the importance of the waste treatment unit in the MMA production process has become increasingly high, and it has become a top priority in the chemical industry.

[0003] In actual production, the MMA waste treatment unit collects unreacted MMA through a tail gas recovery system. However, in the process of MMA tail gas recovery, since the tail gas contains easily polymerizable components such as methyl methyl ester, a polymerization reaction will occur, and the polymerization reaction will be accelerated under the conditions of light, heat, ionizing radiation and catalysts. Therefore, the equipment and pipelines in the MMA tail gas recovery system will be blocked by polymers due to the easy polymerization characteristics of MMA, affecting the stable operation of the tail gas recovery system.

[0004] In order to prevent polymer blockage, fixed doses of inhibitors such as hydroquinone (TBH) and catechol (OPH) are usually used in traditional methods to inhibit the rate of polymerization reactions. However, the traditional method cannot effectively adjust the dosage of inhibitors dynamically according to environmental conditions (such as temperature, MMA concentration, etc.), resulting in relatively high or insufficient dosage of inhibitors, which wastes raw materials and increases production costs. At the same time, the risk of thermal runaway increases due to temperature changes, leading to equipment damage.

[0005] Therefore, how to determine abnormalities in the dosage of inhibitors and adjust the dosage of inhibitors in a timely manner to reduce losses and reduce costs has become an urgent problem to be solved. Summary of the invention

[0006] In view of this, an embodiment of the present invention provides a methyl methacrylate process production data processing system to solve the problem of how to determine abnormalities in the dosage of inhibitors and adjust the dosage of inhibitors in time to reduce losses and reduce costs.

[0007] The embodiment of the present invention provides a methyl methacrylate process production data processing system, which includes the following steps:

[0008] A data acquisition module is used for collecting concentration data and temperature data of methyl methacrylate within a preset historical period at any component in a tail gas recovery system in the process production of methyl methacrylate, and constructing a concentration curve and a temperature curve respectively;

[0009] A data analysis module, configured to obtain an abnormal concentration sub-curve in the concentration curve according to data fluctuations in the concentration curve, obtain a temperature sub-curve corresponding to the time period in which the abnormal concentration sub-curve is located in the temperature curve, obtain an abnormal degree of inhibitor administration in the historical period according to data in the abnormal concentration sub-curve and the temperature sub-curve, and obtain an adjustment category of inhibitor administration dosage according to the abnormal degree of inhibitor administration;

[0010] A model building module is used to use the concentration data and temperature data, and the adjustment category of the inhibitor dosage as sample data in the historical period, use the sample data in multiple historical periods to train the adjustment category recognition model, obtain the trained adjustment category recognition model, and adjust the inhibitor dosage at the component in real time according to the trained adjustment category recognition model.

[0011] Furthermore, the data analysis module obtains an abnormal concentration sub-curve in the concentration curve according to the data fluctuation in the concentration curve, including:

[0012] Obtaining all extreme points on the concentration curve, taking each extreme point as a cutoff point of the concentration curve, obtaining at least two concentration sub-curves, respectively obtaining the slope between the first data point and the last data point on each concentration sub-curve, obtaining the curve slope of each concentration sub-curve, and calculating the average of the absolute values ​​of the curve slopes of all concentration sub-curves to obtain the average slope absolute value;

[0013] For any concentration sub-curve, calculate the absolute value of the difference between the curve slope of the concentration sub-curve and the absolute value of the average slope to obtain a first variable;

[0014] Obtaining a concentration mean value of all data points on the concentration curve, and taking an absolute value of a difference between the last data point in the concentration sub-curve and the concentration mean value as a second variable;

[0015] Using the sum of the first variable and the second variable as an independent variable of a hyperbolic tangent function to obtain an abnormal curve determination coefficient of the concentration sub-curve;

[0016] Obtain abnormal curve determination coefficients of all concentration sub-curves. For any abnormal curve determination coefficient, if the abnormal curve determination coefficient is within a preset abnormal curve range, take the last data point in the concentration sub-curve corresponding to the abnormal curve determination coefficient as the abnormal starting point, obtain the abnormal ending point based on the data in each concentration sub-curve after the abnormal starting point, and take the curve between the abnormal starting point and the abnormal ending point as the abnormal concentration reference sub-curve;

[0017] All abnormal concentration reference sub-curves in the concentration curve are obtained, and according to the position of each abnormal concentration reference sub-curve in the concentration curve, the last abnormal concentration reference sub-curve is selected as the abnormal concentration sub-curve.

[0018] Furthermore, the data analysis module obtains the abnormal end point according to the data in each concentration sub-curve after the abnormal start point, including:

[0019] Taking the first section of the concentration sub-curve after the abnormal starting point as the sub-curve to be analyzed, calculating the absolute value of the difference between the curve slope of the sub-curve to be analyzed and the absolute value of the average slope, to obtain a third variable;

[0020] On the concentration curve, a preset number of data points after the last data point on the sub-curve to be analyzed are obtained, the last data point on the sub-curve to be analyzed and the preset number of data points are used as target data points, the absolute value of the difference between each target data point and the concentration mean is calculated respectively, and the average value of all the absolute values ​​of the differences is used as the fourth variable;

[0021] Using the sum of the third variable and the fourth variable as an independent variable of a hyperbolic tangent function to obtain an abnormal end parameter of the sub-curve to be analyzed;

[0022] If the abnormal end parameter of the sub-curve to be analyzed is within the preset abnormal end parameter range, the last data point on the sub-curve to be analyzed is taken as the abnormal end point;

[0023] If the abnormal end parameter of the sub-curve to be analyzed is not within the preset abnormal end parameter range, the first concentration sub-curve after the sub-curve to be analyzed is used as the sub-curve to be analyzed, and the abnormal end parameter acquisition step is repeated until the abnormal end parameter of the sub-curve to be analyzed is within the preset abnormal end parameter range, thereby obtaining the abnormal end point.

[0024] Furthermore, the data analysis module obtains the abnormal degree of inhibitor administration in the historical period according to the data in the abnormal concentration sub-curve and the temperature sub-curve, including:

[0025] In the concentration curve, the slope between the abnormal starting point corresponding to the abnormal concentration sub-curve and the extreme value point adjacent to its left is obtained to obtain the concentration abnormal starting slope;

[0026] Calculating a first ratio between an average value of all data in the abnormal concentration sub-curve and the concentration mean, taking the product of the first ratio and the concentration anomaly start slope as an independent variable of a hyperbolic tangent function, and obtaining a concentration fluctuation characteristic coefficient of the abnormal concentration sub-curve;

[0027] According to the data fluctuation characteristics of the temperature sub-curve, a temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve is obtained;

[0028] The absolute value of the concentration fluctuation characteristic coefficient and the absolute value of the temperature fluctuation characteristic coefficient are weightedly summed to obtain the abnormal degree of inhibitor administration in the historical period.

[0029] Furthermore, the data analysis module obtains the temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve according to the data fluctuation characteristics of the temperature sub-curve, including:

[0030] Obtaining the slope between the first data point on the temperature sub-curve and the extreme value point adjacent to its left, to obtain the temperature anomaly start slope;

[0031] Obtaining the mean of all data in the temperature curve to obtain a temperature mean, and calculating the ratio between the mean of all data on the temperature sub-curve and the temperature mean to obtain a second ratio;

[0032] Obtaining the number of temperature extreme value points included in the temperature sub-curve to obtain a target number, obtaining an average number according to the number of temperature extreme value points included in the temperature sub-curves corresponding to all abnormal concentration reference sub-curves, and calculating the ratio between the target number and the average number to obtain a third ratio;

[0033] The sum of the second ratio and the third ratio is calculated, and the product of the sum and the temperature anomaly start slope is used as the independent variable of the hyperbolic tangent function to obtain the temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve.

[0034] Furthermore, the adjustment category of the inhibitor dosage is obtained according to the abnormal degree of the inhibitor dosage, including:

[0035] If the inhibitor delivery abnormality degree is not within the preset inhibitor delivery abnormality degree range, determining that the adjustment category of the inhibitor delivery dosage at the component is the first category, wherein the first category is to keep the inhibitor delivery dosage at the component unchanged;

[0036] If the inhibitor dosage abnormality is within a preset inhibitor dosage abnormality range, then the average values ​​of the concentration fluctuation characteristic coefficient and the temperature fluctuation characteristic coefficient corresponding to the inhibitor dosage abnormality are calculated to obtain the polymerization reaction degree;

[0037] If the polymerization reaction degree is less than a preset polymerization reaction degree threshold, determining that the adjustment category of the inhibitor dosage at the component is a second category, wherein the second category is to reduce the inhibitor dosage at the component;

[0038] If the polymerization reaction degree is greater than or equal to the preset polymerization reaction degree threshold, the adjustment category of the inhibitor dosage at the component is determined to be the third category, and the third category is to increase the inhibitor dosage at the component.

[0039] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0040] The present invention provides a methyl methacrylate process production data processing system, comprising a data acquisition module, which is used for, in a tail gas recovery system in the process production of methyl methacrylate, collecting concentration data and temperature data of methyl methacrylate in a preset historical period at any component of the component, and respectively constructing a concentration curve and a temperature curve; a data analysis module, which is used for obtaining an abnormal concentration sub-curve in the concentration curve according to data fluctuations in the concentration curve, obtaining a temperature sub-curve corresponding to the time period where the abnormal concentration sub-curve is located in the temperature curve, obtaining an abnormal degree of inhibitor placement in the historical period according to data in the abnormal concentration sub-curve and the temperature sub-curve, and obtaining an adjustment category of inhibitor placement dosage according to the abnormal degree of inhibitor placement; and a model construction module, which is used for using the concentration data and the temperature data, and the adjustment category of the inhibitor placement dosage as sample data in the historical period, using the sample data in multiple historical periods to train an adjustment category recognition model, obtaining a trained adjustment category recognition model, and adjusting the inhibitor placement dosage at the component in real time according to the trained adjustment category recognition model. Among them, according to the fluctuation characteristics of concentration data in the historical time period, an abnormal concentration sub-curve is obtained; the inhibitor dosage is abnormally judged by combining the data in the abnormal concentration sub-curve and the temperature data corresponding to the abnormal concentration sub-curve, and the abnormal degree of inhibitor dosage is obtained, so that the abnormal judgment of inhibitor dosage is more accurate; according to the abnormal degree of inhibitor dosage, the adjustment category of inhibitor dosage is obtained; the concentration data and temperature data in multiple historical time periods, as well as the adjustment category of inhibitor dosage are used as training sets to train an adjustment category recognition model to obtain a trained adjustment category recognition model, and then according to the trained adjustment category recognition model, the adjustment category of inhibitor dosage is obtained to dynamically adjust the inhibitor dosage, reduce losses, and reduce costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 This is a structural block diagram of a methyl methacrylate process production data processing system provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0043] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0044] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0045] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0046] See also Figure 1 , is a structural block diagram of a methyl methacrylate process production data processing system provided in Example 1 of the present invention, such as Figure 1 As shown, the system may include:

[0047] The data acquisition module 11 is used for collecting the concentration data and temperature data of methyl methacrylate within a preset historical period at any component in the tail gas recovery system in the process production of methyl methacrylate, and constructing a concentration curve and a temperature curve respectively.

[0048] In the recycling system of methyl methacrylate process production, the main components are reactors, condensers, separators and washing towers. In the embodiment of the present invention, taking the reactor as an example, the preset length is set to 10 minutes, which is not limited here. The implementer can set the concentration data and temperature of methyl methacrylate (MMA) within 10 minutes at the reactor according to the implementation scenario. In order to ensure the accuracy and consistency of each data, each data is preprocessed to achieve the purpose of data cleaning, wherein the data preprocessing includes but is not limited to: processing duplicate values, checking whether there are duplicate records in the data set, and deleting them; processing missing values, filling or deleting missing data points; processing outliers, identifying and deleting outliers in the data. Among them, the data preprocessing technology belongs to the prior art and will not be repeated here.

[0049] Since the curve graph can visually display the changing trend of MMA concentration and temperature over time, help identify trends and anomalies, and can provide great support for subsequent analysis, the pre-processed concentration data and temperature data are respectively graphed using Excel, MATLAB and other drawing software to obtain concentration curves and temperature curves, respectively. The horizontal axis of the curve graph represents time, and the vertical axis represents the value. Among them, using Excel, MATLAB and other drawing software to draw curves is a prior art and will not be repeated here.

[0050] The data analysis module 12 is used to obtain an abnormal concentration sub-curve in the concentration curve according to the data fluctuations in the concentration curve, obtain a temperature sub-curve corresponding to the time period of the abnormal concentration sub-curve in the temperature curve, obtain the abnormal degree of inhibitor administration in the historical period according to the data in the abnormal concentration sub-curve and the temperature sub-curve, and obtain the adjustment category of the inhibitor administration dosage according to the abnormal degree of inhibitor administration.

[0051] Due to interference from production processes, production equipment, etc., the MMA concentration in the tail gas collected at the reactor of the MMA tail gas recovery system is not stable, but fluctuates within a certain range. Therefore, when a fixed dose of inhibitor is added, the dosage of the inhibitor will be relatively too high or insufficient. Ideally, when the dosage of the inhibitor is appropriate, the MMA concentration shows regular periodic and stable fluctuations, and when the dosage of the inhibitor is relatively poor, the MMA concentration curve will show a significant increase or decrease. Therefore, it is necessary to find the abnormal concentration sub-curve that shows a significant increase or decrease in the concentration curve, and analyze the fluctuation characteristics of the abnormal concentration sub-curve to determine the abnormal situation of the inhibitor dosage at this stage. The specific method of obtaining the abnormal concentration sub-curve is:

[0052] Obtaining all extreme points on the concentration curve, taking each extreme point as a cutoff point of the concentration curve, obtaining at least two concentration sub-curves, respectively obtaining the slope between the first data point and the last data point on each concentration sub-curve, obtaining the curve slope of each concentration sub-curve, and calculating the average of the absolute values ​​of the curve slopes of all concentration sub-curves to obtain the average slope absolute value;

[0053] For any concentration sub-curve, calculate the absolute value of the difference between the curve slope of the concentration sub-curve and the absolute value of the average slope to obtain a first variable;

[0054] Obtaining a concentration mean value of all data points on the concentration curve, and taking an absolute value of a difference between the last data point in the concentration sub-curve and the concentration mean value as a second variable;

[0055] Using the sum of the first variable and the second variable as an independent variable of a hyperbolic tangent function to obtain an abnormal curve determination coefficient of the concentration sub-curve;

[0056] Obtain abnormal curve determination coefficients of all concentration sub-curves. For any abnormal curve determination coefficient, if the abnormal curve determination coefficient is within a preset abnormal curve range, take the last data point in the concentration sub-curve corresponding to the abnormal curve determination coefficient as the abnormal starting point, obtain the abnormal ending point based on the data in each concentration sub-curve after the abnormal starting point, and take the curve between the abnormal starting point and the abnormal ending point as the abnormal concentration reference sub-curve;

[0057] All abnormal concentration reference sub-curves in the concentration curve are obtained, and according to the position of each abnormal concentration reference sub-curve in the concentration curve, the last abnormal concentration reference sub-curve is selected as the abnormal concentration sub-curve.

[0058] In one embodiment, taking the u-th segment concentration sub-curve as an example, the slope between the first data point and the last data point on the u-th segment concentration sub-curve is used as the curve slope of the u-th segment concentration sub-curve, and the fluctuation characteristics of the concentration data are characterized by the curve slope of the u-th segment concentration sub-curve. According to the fluctuation characteristics of the concentration data on the u-th segment concentration sub-curve, the abnormal curve determination coefficient of the u-th segment concentration sub-curve is obtained, which is used to determine whether the fluctuation of the u-th segment concentration sub-curve is caused by the poor dosage of the inhibitor. The calculation formula of the abnormal curve determination coefficient of the u-th segment concentration sub-curve is:

[0059]

[0060] in, represents the abnormal curve determination coefficient of the u-th concentration sub-curve, represents the slope of the u-th concentration sub-curve, represents the slope of the concentration sub-curve of the i-th segment, represents the tth concentration data on the concentration curve (the last data point in the uth concentration sub-curve), represents the jth concentration data on the concentration curve, represents the number of all concentration sub-curves in the concentration curve, n represents the number of all concentration data on the concentration curve, represents the hyperbolic tangent function, Represents the absolute value symbol.

[0061] It should be noted that is the absolute value of the average slope, That is, the first variable. The greater the difference between the slope of the u-th concentration sub-curve and the absolute value of the average slope, the greater the value of the first variable, indicating that the u-th concentration sub-curve has experienced a significant rise or fall, and thus The larger the value is, the more likely it is that the inhibitor dosage is not good in the time period of the u-th concentration sub-curve; is the mean concentration, The second variable is the larger the value of the second variable, the greater the difference between the last data point in the u-th concentration sub-curve and the concentration mean, indicating that the u-th concentration sub-curve has a significant rise or fall, and thus The larger the value of , the more likely it is that the inhibitor dosage is not good during the time period of the u-th concentration sub-curve.

[0062] Similarly, obtain the abnormal curve determination coefficients of all concentration sub-curves on the concentration curve. Set the preset abnormal curve range to , there is no restriction here, and the implementer can set it according to the implementation scenario. If the abnormal curve determination coefficient of the u-th concentration sub-curve is not , it means that the fluctuation of the concentration sub-curve in the uth segment is not caused by the poor dosage of the inhibitor; if the abnormal curve determination coefficient of the concentration sub-curve in the uth segment is , it is believed that the fluctuation of the concentration sub-curve in the uth segment is caused by the poor dosage of the inhibitor.

[0063] Since the MMA concentration fluctuates within a certain range, the inhibitor dosage will change from poor to appropriate as the MMA concentration fluctuates, or, as the inhibitor dosage is adjusted, the inhibitor dosage will change from poor to appropriate. At this time, the fluctuation of the concentration sub-curve decreases, that is, the curve change of the concentration sub-curve tends to be stable. Therefore, if the abnormal curve determination coefficient of the u-th segment concentration sub-curve is , the last data point on the u-th concentration sub-curve is taken as the abnormal starting point of the abnormal concentration reference sub-curve, and the abnormal ending point of the abnormal concentration reference sub-curve is obtained according to the data changes in each concentration sub-curve after the abnormal starting point, and the curve between the abnormal starting point and the abnormal ending point is taken as the abnormal concentration reference sub-curve.

[0064] Similarly, all abnormal concentration reference sub-curves on the concentration curve are obtained. Since each abnormal concentration reference sub-curve corresponds to a different time period, in order to follow up the effect of the adjusted inhibitor dosage, the abnormal concentration reference sub-curve that is far away from the current moment has no reference value. Therefore, the abnormal concentration reference sub-curve closest to the current moment is selected as the abnormal concentration sub-curve, that is, according to the position of each abnormal concentration sub-reference sub-curve in the concentration curve, the last abnormal concentration reference sub-curve is selected as the abnormal concentration sub-curve. By analyzing the fluctuation characteristics of the abnormal concentration sub-curve, it is determined whether the inhibitor dosage in the historical period corresponding to the concentration curve is appropriate.

[0065] The specific method for obtaining the abnormal end point is as follows:

[0066] Taking the first section of the concentration sub-curve after the abnormal starting point as the sub-curve to be analyzed, calculating the absolute value of the difference between the curve slope of the sub-curve to be analyzed and the absolute value of the average slope, to obtain a third variable;

[0067] On the concentration curve, a preset number of data points after the last data point on the sub-curve to be analyzed are obtained, the last data point on the sub-curve to be analyzed and the preset number of data points are used as target data points, the absolute value of the difference between each target data point and the concentration mean is calculated respectively, and the average value of all the absolute values ​​of the differences is used as the fourth variable;

[0068] Using the sum of the third variable and the fourth variable as an independent variable of a hyperbolic tangent function to obtain an abnormal end parameter of the sub-curve to be analyzed;

[0069] If the abnormal end parameter of the sub-curve to be analyzed is within the preset abnormal end parameter range, the last data point on the sub-curve to be analyzed is taken as the abnormal end point;

[0070] If the abnormal end parameter of the sub-curve to be analyzed is not within the preset abnormal end parameter range, the first concentration sub-curve after the sub-curve to be analyzed is used as the sub-curve to be analyzed, and the abnormal end parameter acquisition step is repeated until the abnormal end parameter of the sub-curve to be analyzed is within the preset abnormal end parameter range, thereby obtaining the abnormal end point.

[0071] In one embodiment, it is assumed that the first concentration sub-curve after the abnormal starting point corresponding to the u-th concentration sub-curve is the first concentration sub-curve in the concentration curve. The concentration sub-curve is The concentration sub-curve of the first segment is taken as the sub-curve to be analyzed. The calculation formula for the abnormal end parameter of the segment concentration sub-curve (sub-curve to be analyzed) is:

[0072]

[0073] in, Indicates Abnormal end parameters of the segment concentration sub-curve (sub-curve to be analyzed), Indicates The slope of the segment concentration sub-curve (the sub-curve to be analyzed), represents the slope of the concentration sub-curve of the i-th segment, represents the number of all concentration sub-curves in the concentration curve, Indicates the preset quantity, Indicates the concentration curve concentration data (the last data point in the sub-curve to be analyzed), represents the jth concentration data on the concentration curve, n represents the number of all concentration data on the concentration curve, represents the hyperbolic tangent function, Indicates absolute value.

[0074] It should be noted that setting There is no restriction here, and the implementer can set it according to the implementation scenario. is the absolute value of the average slope, is the third variable, The smaller the difference between the absolute value of the slope of the segment concentration sub-curve (the sub-curve to be analyzed) and the average slope, the smaller the value of the third variable, indicating that the The concentration sub-curve (sub-curve to be analyzed) tends to be stable, and then The smaller the value, the The amount of inhibitor added in the time period of the segment concentration sub-curve (the sub-curve to be analyzed) is relatively appropriate; is the mean concentration, This is the fourth variable, The last data point on the concentration sub-curve (the sub-curve to be analyzed) of the segment concentration data), and the The smaller the difference between the individual concentration data and the mean concentration, the smaller the value of the fourth variable, indicating that on the concentration curve, the The concentration data after the concentration sub-curve (the sub-curve to be analyzed) also tends to be stable, and then The smaller the value, the The amount of inhibitor added in the time period of the segment concentration sub-curve (the sub-curve to be analyzed) is relatively appropriate.

[0075] Set the preset abnormal end parameter range to There is no restriction here, and the implementer can set it according to the implementation scenario. The abnormal end parameter of the segment concentration sub-curve (sub-curve to be analyzed) is , then the The last data point on the concentration sub-curve (the sub-curve to be analyzed) of the segment is taken as the abnormal end point; if the The abnormal end parameter of the segment concentration sub-curve (sub-curve to be analyzed) is not , then the The first concentration sub-curve after the first concentration sub-curve (the sub-curve to be analyzed) is taken as the sub-curve to be analyzed. The concentration sub-curve of the segment is taken as the sub-curve to be analyzed, according to the The abnormal end parameter acquisition method of the segment concentration sub-curve is used to obtain the The abnormal end parameter of the segment concentration sub-curve is set until the abnormal end parameter is , and obtain the abnormal end point corresponding to the u-th concentration sub-curve.

[0076] It should be noted that, since there may not be a relatively suitable inhibitor dosage in the concentration data collected within 10 minutes, there will be a situation where there is no abnormal end point. At this time, the abnormal concentration sub-curve is the curve from the abnormal start point to the last data on the concentration curve.

[0077] In the recovery system of MMA process production, there are many factors that affect the MMA concentration, such as process differences, equipment differences, operation differences, etc., which may cause changes in the MMA concentration in the tail gas recovery system. Therefore, the concentration data on the abnormal concentration sub-curve cannot be fully used as a basis for judging whether the inhibitor dosage is poor.

[0078] Considering that polymerization is an exothermic reaction, and the dosage of inhibitors will have a great impact on the rate of polymerization, and this impact will not change due to process differences, equipment differences, and operation differences, the temperature data corresponding to the time period of the abnormal concentration sub-curve is used as an auxiliary judgment condition, and the fluctuation characteristics of the temperature sub-curve corresponding to the time period of the abnormal concentration sub-curve are analyzed. Combined with the fluctuation characteristics of the abnormal concentration sub-curve, the abnormal degree of the inhibitor dosage is obtained to enhance the rationality and accuracy of the abnormal judgment of the inhibitor dosage. Then the abnormal degree of the inhibitor dosage is obtained as follows:

[0079] (1) According to the data fluctuation characteristics of the abnormal concentration sub-curve, the concentration fluctuation characteristic coefficient of the abnormal concentration sub-curve is obtained.

[0080] Specifically, in the concentration curve, the slope between the abnormal starting point corresponding to the abnormal concentration sub-curve and the extreme value point adjacent to its left is obtained to obtain the concentration abnormal starting slope;

[0081] A first ratio between the average value of all data in the abnormal concentration sub-curve and the concentration mean is calculated, and the product of the first ratio and the concentration anomaly start slope is used as an independent variable of the hyperbolic tangent function to obtain the concentration fluctuation characteristic coefficient of the abnormal concentration sub-curve.

[0082] In one embodiment, the calculation formula of the concentration fluctuation characteristic coefficient of the abnormal concentration sub-curve is:

[0083]

[0084] in, represents the concentration fluctuation characteristic coefficient of the abnormal concentration sub-curve, It represents the slope between the abnormal starting point corresponding to the abnormal concentration sub-curve and its left adjacent extreme point, that is, the concentration anomaly starting slope. Indicates the abnormal concentration sub-curve Concentration data, Represents the number of all concentration data on the abnormal concentration sub-curve, Indicates the concentration curve Concentration data, Indicates the number of all concentration data on the concentration curve, represents the hyperbolic tangent function.

[0085] It should be noted that is the mean concentration, is the average value of all data in the abnormal concentration sub-curve, is the first ratio, Used to control the output results The larger the average value of all data in the abnormal concentration sub-curve, the larger the value of the first ratio, indicating that the overall value on the abnormal concentration sub-curve is much higher than the average level, and thus The closer the value is to the constant 1, the higher the MMA concentration on the abnormal concentration sub-curve suddenly increases, and the inhibitor dosage is relatively poor; the smaller the average value of all data in the abnormal concentration sub-curve, the smaller the value of the first ratio, indicating that the overall value on the abnormal concentration sub-curve is far below the average level, and thus The closer the value is to the constant -1, it means that the MMA concentration on the abnormal concentration sub-curve suddenly decreases and the inhibitor dosage is relatively poor.

[0086] (2) According to the data fluctuation characteristics of the temperature sub-curve, the temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve is obtained.

[0087] Specifically, the slope between the first data point on the temperature sub-curve and the extreme value point adjacent to its left is obtained to obtain the temperature anomaly start slope;

[0088] Obtaining the mean of all data in the temperature curve to obtain a temperature mean, and calculating the ratio between the mean of all data on the temperature sub-curve and the temperature mean to obtain a second ratio;

[0089] Obtaining the number of temperature extreme value points included in the temperature sub-curve to obtain a target number, obtaining an average number according to the number of temperature extreme value points included in the temperature sub-curves corresponding to all abnormal concentration reference sub-curves, and calculating the ratio between the target number and the average number to obtain a third ratio;

[0090] The sum of the second ratio and the third ratio is calculated, and the product of the sum and the temperature anomaly start slope is used as the independent variable of the hyperbolic tangent function to obtain the temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve.

[0091] In one embodiment, the temperature fluctuation characteristic coefficient is calculated as follows:

[0092]

[0093] in, Indicates the temperature fluctuation characteristic coefficient of the temperature sub-curve corresponding to the time period of the abnormal concentration sub-curve, Indicates the first data point on the temperature sub-curve corresponding to the time period of the abnormal concentration sub-curve (the first data point on the temperature curve). The slope between a data point) and its left adjacent extreme point, that is, the starting slope of the temperature anomaly, Indicates the temperature sub-curve Temperature data, Indicates the number of all temperature data on the temperature sub-curve, Indicates the temperature curve Temperature data, Indicates the number of all temperature data on the temperature curve. Represents the number of all temperature extreme points on the temperature sub-curve, It represents the average value of the number of temperature extreme points contained in the temperature sub-curves corresponding to all abnormal concentration reference sub-curves. represents the hyperbolic tangent function.

[0094] It should be noted that is the mean temperature, That is, the mean of all data on the temperature sub-curve, This is the second ratio, This is the third ratio, Used to control the output results The larger the mean of all data on the temperature sub-curve, the larger the second ratio, indicating that the temperature has risen significantly during the time period of the temperature sub-curve and is much higher than the average level, and thus The closer the value is to the constant 1, the poorer the inhibitor dosage is. The larger the third ratio is, the more the temperature data in the temperature sub-curve fluctuates violently, resulting in an obvious increase in extreme points. The closer the value is to the constant 1, the poorer the inhibitor dosage is. The smaller the second ratio is, the lower the temperature is during the time period of the temperature sub-curve, and the lower the temperature is. The closer the value is to the constant -1, the poorer the inhibitor dosage is. The smaller the third ratio is, the smaller the temperature data fluctuation in the temperature sub-curve is, and the data changes tend to be stable, resulting in a significant reduction in extreme points. The closer the value is to the constant -1, the poorer the inhibitor dosage is.

[0095] (3) The abnormal degree of inhibitor administration is obtained by combining the concentration fluctuation characteristic coefficient and the temperature fluctuation characteristic coefficient of the abnormal concentration sub-curve.

[0096] Specifically, a weighted sum is taken for the absolute value of the concentration fluctuation characteristic coefficient and the absolute value of the temperature fluctuation characteristic coefficient to obtain the abnormal degree of inhibitor administration in the historical period.

[0097] In one embodiment, the calculation formula for the abnormal degree of inhibitor administration is:

[0098]

[0099] in, Indicates the abnormal degree of inhibitor release, represents the concentration fluctuation characteristic coefficient, represents the temperature fluctuation characteristic coefficient, represents the first weight coefficient, represents the second weight coefficient, Represents the absolute value symbol.

[0100] It should be noted that setting , There is no restriction here. The implementer can set it according to the implementation scenario. Since the MMA concentration directly affects the polymerization reaction rate and the efficiency of the recovery system, it is necessary to give A higher priority should always be maintained . The larger the value of is, the greater the difference between the concentration data in the abnormal concentration sub-curve and other concentration data. The larger the value of , the greater the possibility that the inhibitor dosage is poor; The larger the value of is, the greater the difference between the temperature data in the temperature sub-curve corresponding to the time period of the abnormal concentration sub-curve and other temperature data. The larger the value of , the more likely it is that the inhibitor dosage is not good.

[0101] Due to the influence of production process and environmental changes, the MMA concentration collected in the MMA tail gas recovery system is not stable, but fluctuates within a certain range. If a specific inhibitor dosage is set according to the abnormal degree of inhibitor dosage, poor inhibitor dosage will still occur, and the specific inhibitor dosage cannot be accurately set. Therefore, it is necessary to increase or decrease the inhibitor dosage according to the abnormal degree of inhibitor dosage on the basis of the existing inhibitor dosage, so that the inhibitor dosage gradually reaches the appropriate amount, minimizes losses and reduces costs.

[0102] Set the preset inhibitor delivery abnormality range to There is no restriction here, and the implementer can set it according to the implementation scenario. , indicating that the inhibitor dosage is more appropriate, then the adjustment category of the inhibitor dosage at the reactor is determined to be the first category: keeping the inhibitor dosage at the reactor unchanged.

[0103] If the inhibitor administration is abnormal , indicating that the inhibitor dosage is relatively poor, it is necessary to obtain the degree of polymerization at the reactor based on the concentration fluctuation characteristic coefficient and the temperature fluctuation characteristic coefficient, and then determine the adjustment category of the inhibitor dosage at the reactor to adjust the inhibitor dosage. The calculation formula for the polymerization degree is:

[0104]

[0105] in, Indicates the degree of polymerization reaction. represents the concentration fluctuation characteristic coefficient, Represents the temperature fluctuation characteristic coefficient.

[0106] The preset polymerization reaction degree threshold is set to 0.5. There is no limit here. The implementer can set it according to the implementation scenario. When the inhibitor dosage is too high, although the polymerization reaction is inhibited, that is, the polymerization reaction degree is small, it leads to the occurrence of side reactions, that is, MMA self-polymerizes, resulting in a decrease in MMA concentration. At the same time, due to the small degree of polymerization, the heat released by the polymerization reaction is reduced, that is, the temperature drops and remains stable at a lower temperature. Therefore, if , indicating that the inhibitor dosage is excessive and the inhibitor dosage should be reduced, that is, the adjustment category of the inhibitor dosage at the reactor is determined to be the second category: reducing the inhibitor dosage at the reactor.

[0107] When the inhibitor dosage is insufficient, the polymerization reaction is not effectively inhibited, that is, the degree of polymerization is large, the polymerization rate is accelerated, the chemical balance of the polymerization reaction is destroyed, resulting in an increase in the MMA concentration. At the same time, the polymerization reaction releases a large amount of heat, causing the temperature to rise sharply and accompanied by violent fluctuations. Therefore, if , indicating that the inhibitor dosage is insufficient and the inhibitor dosage should be increased, that is, the adjustment category of the inhibitor dosage at the reactor is determined to be the third category: increase the inhibitor dosage at the reactor.

[0108] It should be noted that the focus of the present invention is to make abnormal judgments on the inhibitor dosage and obtain an adjustment plan for the inhibitor dosage based on the abnormal situation. The specific adjustment amount of the inhibitor dosage is not the focus of the present invention and will not be repeated here. It can be set according to the specific implementation scenario, for example: dynamically increase or decrease by setting a fixed adjustment value.

[0109] The model building module 13 is used to use the concentration data and temperature data, and the adjustment category of the inhibitor dosage as sample data in the historical period, and use the sample data in multiple historical periods to train the adjustment category recognition model to obtain a trained adjustment category recognition model, and according to the trained adjustment category recognition model, adjust the inhibitor dosage at the component in real time.

[0110] The concentration data and temperature data within 10 minutes, as well as the adjustment category of the inhibitor dosage are taken as sample data, and the sample data in multiple time periods are divided into a training set and a test set. The concentration data and temperature data are taken as input, and the adjustment category of the corresponding inhibitor dosage is taken as output. The training set is used to train the adjustment category recognition model, and the test set is used to evaluate the performance of the adjustment category recognition model to obtain the final trained adjustment category recognition model. In the process of training the adjustment category recognition model, a convolutional neural network model is used. Training the convolutional neural network model is a prior art and will not be repeated here.

[0111] After obtaining the trained adjustment category recognition model, the real-time temperature data and real-time concentration data of MMA within 10 minutes are acquired in real time, and the real-time concentration data and real-time temperature data are input into the trained adjustment category recognition model to obtain the corresponding adjustment category of the inhibitor dosage, and then the inhibitor dosage at the reactor is dynamically adjusted according to the adjustment category of the inhibitor dosage output by the trained adjustment category recognition model.

[0112] In summary, the present invention provides a methyl methacrylate process production data processing system, comprising a data acquisition module, for collecting concentration data and temperature data of methyl methacrylate within a preset length of historical period at any component in a tail gas recovery system in the process production of methyl methacrylate, and respectively constructing a concentration curve and a temperature curve; a data analysis module, for obtaining an abnormal concentration sub-curve in the concentration curve according to data fluctuations in the concentration curve, obtaining a temperature sub-curve corresponding to the time period where the abnormal concentration sub-curve is located in the temperature curve, obtaining an abnormal degree of inhibitor administration within the historical period according to data in the abnormal concentration sub-curve and the temperature sub-curve, and obtaining an adjustment category of inhibitor administration dosage according to the abnormal degree of inhibitor administration; a model construction module, for using the concentration data and temperature data, and the adjustment category of the inhibitor administration dosage as sample data within the historical period, using sample data within multiple historical periods to train an adjustment category recognition model, obtaining a trained adjustment category recognition model, and adjusting the inhibitor administration dosage at the component in real time according to the trained adjustment category recognition model. Among them, according to the fluctuation characteristics of concentration data in the historical time period, an abnormal concentration sub-curve is obtained; the inhibitor dosage is abnormally judged by combining the data in the abnormal concentration sub-curve and the temperature data corresponding to the abnormal concentration sub-curve, and the abnormal degree of inhibitor dosage is obtained, so that the abnormal judgment of inhibitor dosage is more accurate; according to the abnormal degree of inhibitor dosage, the adjustment category of inhibitor dosage is obtained; the concentration data and temperature data in multiple historical time periods, as well as the adjustment category of inhibitor dosage are used as training sets to train an adjustment category recognition model to obtain a trained adjustment category recognition model, and then according to the trained adjustment category recognition model, the adjustment category of inhibitor dosage is obtained to dynamically adjust the inhibitor dosage, reduce losses, and reduce costs.

[0113] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A methyl methacrylate process production data processing system, characterized in that: The methyl methacrylate process production data processing system comprises: A data acquisition module is used for collecting concentration data and temperature data of methyl methacrylate within a preset historical period at any component in a tail gas recovery system in the process production of methyl methacrylate, and constructing a concentration curve and a temperature curve respectively; A data analysis module, configured to obtain an abnormal concentration sub-curve in the concentration curve according to data fluctuations in the concentration curve, obtain a temperature sub-curve corresponding to the time period in which the abnormal concentration sub-curve is located in the temperature curve, obtain an abnormal degree of inhibitor administration in the historical period according to data in the abnormal concentration sub-curve and the temperature sub-curve, and obtain an adjustment category of inhibitor administration dosage according to the abnormal degree of inhibitor administration; a model building module, for using the concentration data and temperature data, and the adjustment category of the inhibitor dosage as sample data in the historical period, training an adjustment category recognition model using the sample data in multiple historical periods, obtaining a trained adjustment category recognition model, and adjusting the inhibitor dosage at the component in real time according to the trained adjustment category recognition model; According to the data fluctuation in the concentration curve, an abnormal concentration sub-curve in the concentration curve is obtained, including: Obtaining all extreme points on the concentration curve, taking each extreme point as a cutoff point of the concentration curve, obtaining at least two concentration sub-curves, respectively obtaining the slope between the first data point and the last data point on each concentration sub-curve, obtaining the curve slope of each concentration sub-curve, and calculating the average of the absolute values ​​of the curve slopes of all concentration sub-curves to obtain the average slope absolute value; For any concentration sub-curve, calculate the absolute value of the difference between the curve slope of the concentration sub-curve and the absolute value of the average slope to obtain a first variable; Obtaining a concentration mean value of all data points on the concentration curve, and taking an absolute value of a difference between the last data point in the concentration sub-curve and the concentration mean value as a second variable; Using the sum of the first variable and the second variable as an independent variable of a hyperbolic tangent function to obtain an abnormal curve determination coefficient of the concentration sub-curve; Obtain abnormal curve determination coefficients of all concentration sub-curves. For any abnormal curve determination coefficient, if the abnormal curve determination coefficient is within a preset abnormal curve range, take the last data point in the concentration sub-curve corresponding to the abnormal curve determination coefficient as the abnormal starting point, obtain the abnormal ending point based on the data in each concentration sub-curve after the abnormal starting point, and take the curve between the abnormal starting point and the abnormal ending point as the abnormal concentration reference sub-curve; Acquire all abnormal concentration reference sub-curves in the concentration curve, and select the last abnormal concentration reference sub-curve as the abnormal concentration sub-curve according to the position of each abnormal concentration reference sub-curve in the concentration curve; According to the data in each concentration sub-curve after the abnormal starting point, the abnormal ending point is obtained, including: Taking the first section of the concentration sub-curve after the abnormal starting point as the sub-curve to be analyzed, calculating the absolute value of the difference between the curve slope of the sub-curve to be analyzed and the absolute value of the average slope, to obtain a third variable; On the concentration curve, a preset number of data points after the last data point on the sub-curve to be analyzed are obtained, the last data point on the sub-curve to be analyzed and the preset number of data points are used as target data points, the absolute value of the difference between each target data point and the concentration mean is calculated respectively, and the average value of all the absolute values ​​of the differences is used as the fourth variable; Using the sum of the third variable and the fourth variable as an independent variable of a hyperbolic tangent function to obtain an abnormal end parameter of the sub-curve to be analyzed; If the abnormal end parameter of the sub-curve to be analyzed is within the preset abnormal end parameter range, the last data point on the sub-curve to be analyzed is taken as the abnormal end point; If the abnormal end parameter of the sub-curve to be analyzed is not within the preset abnormal end parameter range, the first concentration sub-curve after the sub-curve to be analyzed is used as the sub-curve to be analyzed, and the abnormal end parameter acquisition step is repeated until the abnormal end parameter of the sub-curve to be analyzed is within the preset abnormal end parameter range, thereby obtaining the abnormal end point.

2. A methyl methacrylate process production data processing system according to claim 1, characterized in that: The data analysis module obtains the abnormal degree of inhibitor delivery in the historical period according to the data in the abnormal concentration sub-curve and the temperature sub-curve, including: In the concentration curve, the slope between the abnormal starting point corresponding to the abnormal concentration sub-curve and the extreme value point adjacent to its left is obtained to obtain the concentration abnormal starting slope; Calculating a first ratio between an average value of all data in the abnormal concentration sub-curve and the concentration mean, taking the product of the first ratio and the concentration anomaly start slope as an independent variable of a hyperbolic tangent function, and obtaining a concentration fluctuation characteristic coefficient of the abnormal concentration sub-curve; According to the data fluctuation characteristics of the temperature sub-curve, a temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve is obtained; The absolute value of the concentration fluctuation characteristic coefficient and the absolute value of the temperature fluctuation characteristic coefficient are weightedly summed to obtain the abnormal degree of inhibitor administration in the historical period.

3. A methyl methacrylate process production data processing system according to claim 2, characterized in that: The data analysis module obtains the temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve according to the data fluctuation characteristics of the temperature sub-curve, including: Obtaining the slope between the first data point on the temperature sub-curve and the extreme value point adjacent to its left, to obtain the temperature anomaly start slope; Obtaining the mean of all data in the temperature curve to obtain a temperature mean, and calculating the ratio between the mean of all data on the temperature sub-curve and the temperature mean to obtain a second ratio; Obtaining the number of temperature extreme value points included in the temperature sub-curve to obtain a target number, obtaining an average number according to the number of temperature extreme value points included in the temperature sub-curves corresponding to all abnormal concentration reference sub-curves, and calculating the ratio between the target number and the average number to obtain a third ratio; The sum of the second ratio and the third ratio is calculated, and the product of the sum and the temperature anomaly start slope is used as the independent variable of the hyperbolic tangent function to obtain the temperature fluctuation characteristic coefficient corresponding to the abnormal concentration sub-curve.

4. A methyl methacrylate process production data processing system according to claim 1, characterized in that: The adjustment category of the inhibitor dosage is obtained according to the abnormal degree of the inhibitor dosage, including: If the inhibitor administration abnormality degree is not within the preset inhibitor administration abnormality degree range, determining that the adjustment category of the inhibitor administration dosage at the component is the first category, and the first category is to keep the inhibitor administration dosage at the component unchanged; If the inhibitor dosage abnormality is within a preset inhibitor dosage abnormality range, then the average values ​​of the concentration fluctuation characteristic coefficient and the temperature fluctuation characteristic coefficient corresponding to the inhibitor dosage abnormality are calculated to obtain the polymerization reaction degree; If the polymerization reaction degree is less than a preset polymerization reaction degree threshold, determining that the adjustment category of the inhibitor dosage at the component is a second category, wherein the second category is to reduce the inhibitor dosage at the component; If the polymerization reaction degree is greater than or equal to the preset polymerization reaction degree threshold, the adjustment category of the inhibitor dosage at the component is determined to be the third category, and the third category is to increase the inhibitor dosage at the component.

Images