An intelligent treatment method for acetic acid production wastewater
Through multi-position monitoring and mean drift clustering technology, the problem of inaccurate pH judgment in acetic acid production wastewater is solved, the acetic acid recycling effect is improved, and the treatment effect of acetic acid production wastewater is ensured.
Patent Information
- Application Number
- CN202411288865.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-14
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-14
AI Technical Summary
The pH value in acetic acid production wastewater is inaccurate and is greatly affected by impurities, resulting in poor acetic acid recycling effect, and the difference in monitoring location and time affects the treatment effect.
By monitoring the pH value of acetic acid wastewater for a long time in multiple locations, dividing time periods, quantifying pH stability, and performing mean drift clustering to obtain a class cluster. Based on the time difference weighting of data points in the class cluster, the overall pH value of the wastewater is calculated, and Improve monitoring accuracy and recycling effect.
It improves the accuracy of pH monitoring in acetic acid production wastewater treatment, reduces the impact of time differences on the results, and ensures the adequacy of acetic acid recovery and the effective utilization of resources.
Smart Images

Figure CN118817989B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of wastewater treatment, and in particular to an intelligent treatment method for acetic acid production wastewater. Background Art
[0002] During the treatment of acetic acid production wastewater, since the acetic acid production wastewater may contain acetic acid residues, the acetic acid in it can be recovered, which can not only realize the reuse of water resources, but also improve the efficiency of acetic acid production. Therefore, acetic acid recovery is an indispensable and important step in the treatment of acetic acid production wastewater.
[0003] In the current treatment process of acetic acid production wastewater, the interference of other components in the wastewater may lead to inaccurate judgment of the pH value of the wastewater, and then inaccurate judgment of the adequacy of acetic acid recovery, which not only affects the compliance of the wastewater with the standards, but also is a waste of resources. In the monitoring process of acetic acid wastewater, the wastewater at different locations and at different times is affected differently by the pH value of other impurities. Therefore, through long-term continuous monitoring at multiple locations, the positions that can accurately reflect the pH value of the wastewater are extracted, and the interference of impurities and other components on the pH value of the wastewater is eliminated, thereby improving the treatment effect of acetic acid production wastewater. Summary of the invention
[0004] The present invention provides an intelligent treatment method for acetic acid production wastewater to solve the problem that the pH value of acetic acid production wastewater cannot be accurately obtained due to the influence of impurities when monitoring the pH value, thereby affecting the wastewater treatment effect. The technical scheme adopted is as follows:
[0005] The present invention proposes an intelligent treatment method for acetic acid production wastewater, which comprises the following steps:
[0006] Obtain the pH value of acetic acid production wastewater at multiple monitoring locations at each moment;
[0007] Obtaining several time periods at each monitoring location and the pH performance value of each time period at each monitoring location; obtaining the pH stability of each time period at each monitoring location based on the pH value difference at different times in each time period and the pH value difference at the same time at different monitoring locations;
[0008] Based on the pH performance values and pH stability of all monitoring locations at all times, several data points are obtained and a sample space is constructed, and then mean shift clustering is performed; in the clustering process, the initial drift vector of the window is obtained according to the distribution of data points in the window and the corresponding pH stability; combined with the number of data points in the window and the mean of pH stability, the corrected drift vector of the window is corrected, and then clustering is performed to obtain several clusters;
[0009] According to the time difference and pH performance value between the corresponding time periods of the data points in each cluster, the overall pH value of the wastewater is obtained to judge the wastewater recycling effect.
[0010] Optionally, the method of obtaining a plurality of time periods at each monitoring position and a pH value at each time period at each monitoring position includes:
[0011] Preset the time period length, and take the time of each time period length as a time period;
[0012] For any time period at any monitoring position, the average of the pH values at all times in the time period at the monitoring position is taken as the pH expression value of the time period at the monitoring position.
[0013] Optionally, the specific method of obtaining the pH value stability of each monitoring position in each time period includes:
[0014]
[0015] in, Indicates Monitoring location pH stability over time, Indicates The monitoring location The variance of pH values at all times in a period, represents the number of monitoring locations, represents the number of moments in each period, Indicates The monitoring location In the period The pH value at that moment, Indicates The monitoring location In the period The pH value at that moment, represents the absolute value function, Represents an exponential function with a natural constant as its base.
[0016] Optionally, the obtaining of a number of data points and constructing a sample space includes the following specific methods:
[0017] According to the pH stability of all monitoring positions in all time periods, a linear transformation is performed based on the maximum pH value to obtain the pH stability factor of each monitoring position in each time period;
[0018] The sample space is constructed with pH value as the horizontal axis and pH stability factor as the vertical axis. Based on the pH performance value and pH stability factor of each monitoring position in each time period, each monitoring position in each time period is converted into a data point in the sample space.
[0019] Optionally, the obtaining of the initial drift vector of the window includes the following specific methods:
[0020] Take any window as the target window, perform softmax normalization on the pH stability factors of all data points in the target window, and use the result as the stability weight of each data point in the target window. Point the vector from the center point of the target window to each data point in the target window, and record it as the offset vector of each data point in the target window.
[0021] According to the stable weights of the data points in the target window, the offset vectors of all data points in the target window are weighted and summed, and the obtained vector sum is used as the initial drift vector of the target window.
[0022] Optionally, the corrected drift vector of the window is obtained by correction, and the specific method includes:
[0023]
[0024] in, represents the corrected drift vector of the target window, represents the initial drift vector of the target window, represents the mean of the pH stability factors of all data points in the target window, represents the maximum value of the pH stability factor of all data points in the sample space, represents the number of data points in the target window, Represents the total number of data points in the sample space.
[0025] Optionally, the clustering obtains a number of clusters, including a specific method of:
[0026] In the process of each window moving based on the corrected drift vector, if the modulus of the corrected drift vector is less than or equal to the stop threshold, the corresponding window stops moving, and so on, to complete the mean shift clustering. In the final stopped windows, the data points in each window correspond to a cluster.
[0027] Optionally, the overall pH value of the wastewater is obtained by:
[0028] According to the time in the period corresponding to the data points in each cluster, the time weight of each cluster is obtained;
[0029] For any cluster, obtain the mean of the pH expression values of all data points in the cluster, and record it as the pH mean of the cluster;
[0030] According to the momentary weight of each cluster, the pH means of all clusters are weighted and summed, and the result is taken as the overall pH value of the wastewater.
[0031] Optionally, the step of obtaining the moment weight of each cluster includes the following specific methods:
[0032] For any data point in any cluster, the central moment of the time period corresponding to the data point is taken as the corresponding moment of the data point, the absolute value of the difference between the corresponding moments of any two data points in the cluster is calculated, and the mean of all the absolute values of the differences is recorded as the moment difference of the cluster;
[0033] The moment differences of all clusters are normalized by softmax, and the result is used as the moment difference factor of each cluster; the difference obtained by subtracting the moment difference factor of each cluster from 1 is taken as the moment factor of each cluster, and the moment factors of all clusters are normalized by softmax, and the result is used as the moment weight of each cluster.
[0034] Optionally, the determination of wastewater recovery effect includes the following specific methods:
[0035] The neutral value of the pH value is used as the judgment threshold. If the overall pH value of the wastewater is greater than or equal to the judgment threshold, the acetic acid in the wastewater is fully recovered and subsequent final treatment is carried out; if the overall pH value of the wastewater is less than the judgment threshold, acetic acid recovery treatment is carried out again.
[0036] The beneficial effects of the present invention are as follows: the present invention analyzes the pH values at different monitoring positions at various times in the treatment of acetic acid wastewater, divides time periods and quantifies the stability of the pH values, and then performs clustering to obtain a number of clusters. The time differences of data points in the clusters are used as weights to weight the pH means of the clusters to obtain the overall pH value of the wastewater, thereby ensuring the timeliness of wastewater pH monitoring, avoiding large time differences from affecting the accuracy of the monitoring results, and finally completing the treatment of acetic acid production wastewater. The pH values at all times at the monitoring locations are divided into time periods for analysis. Based on the changes in pH values within the time period and the differences in pH value changes at different monitoring locations during the same time period, the pH value stability at each monitoring location is quantified, thereby providing a basis for subsequent clustering of each time period. The greater the pH stability of a monitoring location, the less likely it is to be affected by impurities. Mean shift clustering is performed based on the pH value stability and pH performance value at each monitoring location during the time period. During the clustering process, the initial drift vector of the window is obtained based on the offset vector of the data point and the pH value stability. The stability of the data points in the window is further analyzed, and the corrected drift vector is obtained by correction. Finally, clustering is completed, so that the pH value changes in the time periods of the monitoring locations corresponding to the data points in the same cluster are more similar, thereby improving the accuracy of the overall pH value acquisition. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, 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.
[0038] Figure 1 A schematic flow chart of an intelligent treatment method for acetic acid production wastewater provided by one embodiment of the present invention;
[0039] Figure 2 The figure is a schematic diagram of the process of treating acetic acid production wastewater according to the present invention. DETAILED DESCRIPTION
[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0041] See also Figure 1 , which shows a flow chart of an intelligent treatment method for acetic acid production wastewater provided by an embodiment of the present invention, the method comprising the following steps:
[0042] Step S001, obtaining the pH value of acetic acid production wastewater at each moment at multiple monitoring locations.
[0043] The purpose of this embodiment is to determine whether acetic acid is sufficiently recovered by monitoring the pH value of acetic acid production wastewater, thereby realizing intelligent treatment of acetic acid production wastewater. The chemical reactions between other pollutants such as organic matter and inorganic salts in acetic acid production wastewater affect the pH performance of acetic acid production wastewater. Therefore, it is necessary to monitor the pH value of acetic acid production wastewater at multiple locations for a long period of time, and then obtain monitoring locations that can accurately reflect the pH value of acetic acid production wastewater at each moment.
[0044] Specifically, in this embodiment, 10 monitoring positions are evenly arranged in the wastewater pool of acetic acid production, and the pH value of each monitoring position is monitored in real time by a pH meter. The sampling frequency is set to once per second, and the pH value of each monitoring position at several moments in the last five minutes is collected. The entire treatment process of acetic acid production wastewater is as follows: Figure 2 As shown, after the acetic acid production wastewater is adjusted and precipitated, biochemically treated and membrane separated, the pH value is monitored to determine whether the acetic acid is fully recovered, and then the final treatment is completed and the wastewater is discharged.
[0045] Step S002, obtaining several time periods at each monitoring position and the pH performance value of each time period at each monitoring position; according to the pH value difference at different times in each time period and the pH value difference at the same time at different monitoring positions, the pH value stability of each time period at each monitoring position is obtained.
[0046] Preferably, in one embodiment of the present invention, the specific method for obtaining several time periods at each monitoring position and the pH performance value at each time period at each monitoring position is as follows:
[0047] The preset time period length, in this embodiment, is described as 10 seconds, and every 10 seconds is taken as a time period. For any time period at any monitoring position, the average pH value of all moments in the time period at the monitoring position is taken as the pH performance value of the monitoring position in the time period.
[0048] Preferably, in one embodiment of the present invention, the pH stability of each monitoring position in each time period is obtained according to the pH value difference at different times in each time period and the pH value difference at different monitoring positions at the same time, and the specific method includes:
[0049] First The monitoring location For example, the pH stability of The calculation method is:
[0050]
[0051] in, Indicates The monitoring location The variance of pH values at all times in a period, represents the number of monitoring locations, represents the number of moments in each period, Indicates The monitoring location In the period The pH value at that moment, Indicates The monitoring location In the period The pH value at that moment, represents the absolute value function, represents an exponential function with a natural constant as the base. In this embodiment, The model shows an inverse proportional relationship. As the input of the model, the implementer can set the inverse proportional function according to the actual situation; according to the above method, the pH stability of each monitoring location in each time period is obtained.
[0052] It should be noted that the variance is used to reflect the degree of change of the pH value at each moment in the time period of the monitoring position. The larger the variance, the greater the degree of change, and the smaller the stability of the pH value in the corresponding time period; the greater the difference between the pH value at each moment in the time period of the monitoring position and the pH value of other monitoring positions at the same moment, the greater the overall change of the pH value of the monitoring position in the time period compared with other monitoring positions, and the smaller the corresponding pH value stability.
[0053] At this point, by dividing the pH values at all times at the monitoring locations into time periods, the pH stability of each monitoring location in each time period is quantified based on the changes in pH values within the time period and the differences in pH value changes at different monitoring locations in the same time period, thereby providing a basis for subsequent clustering of each time period. The greater the pH stability of a monitoring location, the less likely it is to be affected by impurities.
[0054] Step S003, based on the pH performance values and pH stability of all monitoring positions in all time periods, a number of data points are obtained and a sample space is constructed, and then mean shift clustering is performed; in the clustering process, the initial drift vector of the window is obtained according to the distribution of data points in the window and the corresponding pH stability; combined with the number of data points in the window and the mean of pH stability, the corrected drift vector of the window is corrected to obtain a number of clusters.
[0055] Preferably, in one embodiment of the present invention, based on the pH performance values and pH stability of all monitoring positions at all time periods, a number of data points are obtained and a sample space is constructed, and then mean shift clustering is performed, including the specific method of:
[0056] First The monitoring location For example, the pH stability factor The calculation method is:
[0057]
[0058] in, Indicates Monitoring location pH stability over time, It indicates the maximum pH stability value of all monitoring locations at all time periods. Indicates the maximum value of pH value. ; Obtain the pH stability factor for each monitoring location and each time period according to the above method.
[0059] It should be noted that, since there is a difference in the dimension of pH stability and pH value, a linear transformation of pH stability is performed to obtain a pH stability factor.
[0060] Furthermore, the sample space is constructed with pH value as the horizontal axis and pH value stability factor as the vertical axis; based on the pH performance value and pH value stability factor of each monitoring position in each time period, each monitoring position and each time period are converted into a data point in the sample space, where the pH performance value is obtained by averaging the pH value, and can be directly converted to obtain the data point of each monitoring position and each time period in the sample space.
[0061] Furthermore, the range of the horizontal and vertical coordinates in the sample space is , the number of windows is preset. In this embodiment, 10 windows are evenly set in the sample space. The side length of each window is described as 1. The number of windows, window distribution and window side length are all existing means in mean shift clustering, and will not be repeated in this embodiment.
[0062] Preferably, in one embodiment of the present invention, in the clustering process, the initial drift vector of the window is obtained according to the distribution of data points in the window and the corresponding pH value stability, including the specific method of:
[0063] Take any window as the target window, perform softmax normalization on the pH stability factors of all data points in the target window, and use the result as the stability weight of each data point in the target window. Point the vector from the center of the target window to each data point in the target window, and record it as the offset vector of each data point in the target window. According to the stability weight of the data points in the target window, weighted sum is taken for the offset vectors of all data points in the target window, and the sum of the vectors is used as the initial drift vector of the target window.
[0064] It should be noted that the drift vector of the window is obtained by weighted summing of the offset vectors of the data points in the window. However, if the pH stability factors of the data points in the window are small and the number of data points is small, the credibility will be low, which will affect the accuracy of the drift vector of the window. Therefore, the initial drift vector needs to be corrected.
[0065] Preferably, in one embodiment of the present invention, the number of data points in the window and the mean value of pH stability are combined to correct the corrected drift vector of the window, and then cluster to obtain several clusters, including the specific method of:
[0066]
[0067] in, represents the corrected drift vector of the target window, represents the initial drift vector of the target window, represents the mean of the pH stability factors of all data points in the target window, represents the maximum value of the pH stability factor of all data points in the sample space, represents the number of data points in the target window, Represents the total number of data points in the sample space.
[0068] It should be noted that the smaller the mean value of the pH stability factor of the data points in the window and the smaller the number of data points, the smaller the modulus of the drift vector should be, so as to obtain the corrected drift vector.
[0069] Furthermore, a stopping threshold is preset. In this embodiment, the stopping threshold is described as 0.2. In the process of each window moving based on the corrected drift vector, if the modulus of the corrected drift vector obtained is less than or equal to the stopping threshold, the corresponding window stops moving, and so on, to complete the mean shift clustering. Among the several windows that are finally stopped, the data points in each window correspond to a cluster respectively, that is, each window obtained in the end is a cluster.
[0070] At this point, mean shift clustering is performed based on the pH value stability and pH performance value of each monitoring position in each time period. During the clustering process, the initial drift vector of the window is obtained based on the offset vector of the data point and the pH value stability, and the stability of the data points in the window is further analyzed. The corrected drift vector is corrected and finally clustering is completed, so that the pH value changes in the time period of the monitoring position corresponding to the data points in the same cluster are more similar, thereby improving the accuracy of subsequent overall pH value acquisition.
[0071] Step S004: According to the time difference and pH expression value between the corresponding time periods of the data points in each cluster, the overall pH value of the wastewater is obtained to determine the wastewater recovery effect.
[0072] Preferably, in one embodiment of the present invention, the overall pH value of the wastewater is obtained according to the time difference and pH expression value between the corresponding time periods of the data points in each cluster, and the wastewater recovery effect is judged, including the specific method of:
[0073] For any data point in any cluster, take the central moment of the time period corresponding to the data point as the corresponding moment of the data point, calculate the absolute value of the difference between the corresponding moments of any two data points in the cluster, and record the mean of all the absolute values of the difference as the moment difference of the cluster; perform softmax normalization on the moment differences of all clusters, and use the result as the moment difference factor of each cluster; obtain the difference obtained by subtracting the moment difference factor of each cluster from 1, and use it as the moment factor of each cluster; perform softmax normalization on the moment factors of all clusters, and use the result as the moment weight of each cluster.
[0074] Furthermore, for any cluster, the mean of the pH expression values of all data points in the cluster is obtained, which is recorded as the pH mean of the cluster; according to the moment weight of each cluster, the pH means of all clusters are weighted and summed, and the result is taken as the overall pH value of the wastewater; the neutral value of pH value pH=7 is used as the judgment threshold, if the overall pH value of the wastewater is greater than or equal to the judgment threshold, the acetic acid in the wastewater is fully recovered and subsequent final treatment is carried out; if the overall pH value of the wastewater is less than the judgment threshold, acetic acid recovery treatment is required again.
[0075] At this point, this embodiment is completed.
[0076] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. An intelligent treatment method for acetic acid production wastewater, characterized in that: The method comprises the following steps: Obtain the pH value of acetic acid production wastewater at multiple monitoring locations at each moment; Obtaining several time periods at each monitoring location and the pH performance value of each time period at each monitoring location; obtaining the pH stability of each time period at each monitoring location based on the pH value difference at different times in each time period and the pH value difference at the same time at different monitoring locations; Based on the pH performance values and pH stability of all monitoring locations at all times, several data points are obtained and a sample space is constructed, and then mean shift clustering is performed; in the clustering process, the initial drift vector of the window is obtained according to the distribution of data points in the window and the corresponding pH stability; combined with the number of data points in the window and the mean of pH stability, the corrected drift vector of the window is corrected, and then clustering is performed to obtain several clusters; According to the time difference and pH performance value between the corresponding time periods of the data points in each cluster, the overall pH value of the wastewater is obtained to judge the wastewater recycling effect; The specific method of obtaining the pH value of each monitoring position at a plurality of time periods and each time period at each monitoring position is as follows: Preset the time period length, and take the time of each time period length as a time period; For any period of time at any monitoring position, the average of the pH values at all times in the period of time at the monitoring position is taken as the pH performance value of the period of time at the monitoring position; The specific method for obtaining the pH stability of each monitoring position in each time period is as follows: in, Indicates Monitoring location pH stability over time, Indicates The monitoring location The variance of pH values at all times in a period, represents the number of monitoring locations, represents the number of moments in each period, Indicates The monitoring location In the period The pH value at that moment, Indicates The monitoring location In the period The pH value at that moment, represents the absolute value function, represents an exponential function with a natural constant as base; The specific method of obtaining a number of data points and constructing a sample space includes: According to the pH stability of all monitoring positions in all time periods, a linear transformation is performed based on the maximum pH value to obtain the pH stability factor of each monitoring position in each time period; The sample space is constructed with pH value as the horizontal axis and pH stability factor as the vertical axis; based on the pH performance value and pH stability factor of each monitoring position in each period, each monitoring position and each period are converted into a data point in the sample space; pH Stabilizing Factor The calculation method is: in, Indicates Monitoring location pH stability over time, It indicates the maximum pH stability value of all monitoring locations at all time periods. Indicates the maximum pH value; The specific method of obtaining the initial drift vector of the window includes: Take any window as the target window, perform softmax normalization on the pH stability factors of all data points in the target window, and use the result as the stability weight of each data point in the target window. Point the vector from the center point of the target window to each data point in the target window, and record it as the offset vector of each data point in the target window. According to the stable weights of the data points in the target window, the offset vectors of all the data points in the target window are weighted and summed, and the obtained vector sum is used as the initial drift vector of the target window; The modified drift vector of the window is obtained by modification, and the specific method includes: in, represents the corrected drift vector of the target window, represents the initial drift vector of the target window, represents the mean of the pH stability factors of all data points in the target window, represents the maximum value of the pH stability factor of all data points in the sample space, represents the number of data points in the target window, Represents the total number of data points in the sample space; The clustering obtains several clusters, including the following specific methods: In the process of each window moving based on the corrected drift vector, if the modulus of the corrected drift vector is less than or equal to the stop threshold, the corresponding window stops moving, and so on, the mean shift clustering is completed, and the data points in each window of the final stopped windows correspond to a cluster; The overall pH value of the wastewater is specifically obtained by: According to the time in the period corresponding to the data points in each cluster, the time weight of each cluster is obtained; For any cluster, obtain the mean of the pH expression values of all data points in the cluster, and record it as the pH mean of the cluster; According to the moment weight of each cluster, the pH mean values of all clusters are weighted and summed, and the result is taken as the overall pH value of the wastewater; The specific method of obtaining the moment weight of each cluster is as follows: For any data point in any cluster, the central moment of the time period corresponding to the data point is taken as the corresponding moment of the data point, the absolute value of the difference between the corresponding moments of any two data points in the cluster is calculated, and the mean of all the absolute values of the differences is recorded as the moment difference of the cluster; The moment differences of all clusters are normalized by softmax, and the result is used as the moment difference factor of each cluster; the difference obtained by subtracting the moment difference factor of each cluster from 1 is taken as the moment factor of each cluster, and the moment factors of all clusters are normalized by softmax, and the result is used as the moment weight of each cluster.
2. The intelligent treatment method for acetic acid production wastewater according to claim 1, characterized in that: The specific method for judging the wastewater recovery effect is as follows: The neutral value of the pH value is used as the judgment threshold. If the overall pH value of the wastewater is greater than or equal to the judgment threshold, the acetic acid in the wastewater is fully recovered and subsequent final treatment is carried out; if the overall pH value of the wastewater is less than the judgment threshold, acetic acid recovery treatment is carried out again.
