CN115546720A - Image type analytic regulation and control method and device for flocculation working condition - Google Patents
Image type analytic regulation and control method and device for flocculation working condition Download PDFInfo
- Publication number
- CN115546720A CN115546720A CN202211222711.4A CN202211222711A CN115546720A CN 115546720 A CN115546720 A CN 115546720A CN 202211222711 A CN202211222711 A CN 202211222711A CN 115546720 A CN115546720 A CN 115546720A
- Authority
- CN
- China
- Prior art keywords
- image
- flocculation
- effluent
- gray level
- working condition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The invention discloses an image type analysis regulation and control method and device for flocculation working conditions, which are used for obtaining water outlet images of a flocculation tank of a water treatment unit under different flocculant dosing quantities, and manually measuring the water outlet turbidity of a flocculation tank outlet under each flocculant dosing quantity; graying the floc image set, calculating the average fractal dimension and equivalent particle size of the flocs in each image, and calculating texture characteristic entropy and correlation of a gray level co-occurrence matrix set and each gray level co-occurrence matrix set; constructing and training a random forest classifier; obtaining a regression classifier; and generating a flocculating working condition lookup table of the flocculant dosing amount, the effluent image characteristics and the effluent turbidity, and judging whether the effluent turbidity of the current flocculating tank and the working condition are qualified. The overall texture features of the flocculation image improve the stability of the image features on the characterization of the flocculation effect, and a random forest classifier is constructed to automatically study and judge whether the effluent reaches the standard, so that the automatic study and judgment of the water treatment working condition and the automatic regulation and control of the flocculant dosage are realized.
Description
Technical Field
The invention belongs to the technical field of water purification treatment, and relates to an image type analytic regulation and control method and device for flocculation working conditions.
Background
Among the usual water treatment process, need the engineer regularly to get the water sample and use turbidity appearance to detect out the water turbidity to the flocculation operating mode after the raw water medicine is added in the visual observation, complex operation, the actual effect is poor, can appear the error that the too big or analysis time overlength of sampling interval leads to, and then influences the effect of water purification flocculation.
With the rapid development of machine vision technology, academic circles turn to the characterization of flocculation conditions by using flocculation process images, and at present, the academic circles generally use fractal dimension and particle size of flocs in the flocculation images to characterize the flocculation conditions and can obtain better results, so that the defects of tedious manual treatment and poor timeliness are overcome. However, the fractal dimension and the particle size characteristics of the flocs only stay at the aspect of individual morphological characteristics of the flocs, and are easily influenced by the distribution of the flocs and the adhesion of the flocs on an image. The main reason is that flocs can be randomly distributed under the influence of water flow disturbance, and the influence of floc adhesion overlapping and uneven distribution cannot be eliminated when images are sampled, so that the error condition is judged according to the images. Aiming at the problems of floc adhesion overlapping and uneven distribution, the overall texture characteristics of a flocculated image can well represent the overall distribution condition of flocs, so that from the perspective of machine vision, the prior art still has a great improvement space.
Disclosure of Invention
In order to solve the above defects in the prior art, the present invention aims to provide an image type analysis regulation method and device for flocculation working conditions, which introduces integral texture features of a flocculation image on the basis of the prior art, improves the problem that the prior art is prone to errors caused by uneven floc distribution, and improves the stability of the image features on the characterization of flocculation effect. According to the invention, the effluent turbidity of the flocculation tank is analyzed by using an image analysis technology through the effluent image and the historical data of the effluent turbidity under different flocculant dosing amounts, a random forest classifier is constructed, the independent study and judgment of the effluent turbidity and whether the flocculation working condition reaches the standard or not are supported by using the flocculant dosing amount and the effluent image characteristics, and the independent study and judgment of the water treatment working condition and the automatic regulation and control of the flocculant dosing amount are realized.
The invention is realized by the following technical scheme.
In one aspect of the invention, an image type analytic regulation and control method for flocculation conditions is provided, which comprises the following steps:
acquiring water outlet images of the flocculation tanks of the water treatment units under different flocculant dosing amounts to obtain a floc image set; meanwhile, the effluent turbidity of the water outlet of the flocculation tank under the dosage of each flocculating agent is manually measured;
carrying out graying processing on the floc image set to obtain a gray image set;
carrying out binarization processing on the obtained gray level image set after removing the illumination influence to obtain a binary image set, and calculating the average fractal dimension and the equivalent particle size of flocs in each image in the binary image set to obtain a morphological feature set of a water outlet image;
respectively calculating texture feature entropies and correlations of the gray level co-occurrence matrix set and each gray level co-occurrence matrix set to the obtained gray level image set to obtain a texture feature set of the effluent image;
constructing a forest classifier model by using the flocculant dosing amount sequence, the morphological feature set and the textural feature set of the effluent image and the effluent turbidity of the water outlet of the flocculation tank, and training a random forest classifier; obtaining a regression classifier among the dosage of the flocculating agent, the effluent image characteristics and the effluent turbidity;
generating a flocculant dosing amount, effluent image characteristics and effluent turbidity flocculation working condition lookup table, judging whether the effluent turbidity and the working condition of the current flocculation tank are qualified or not, and if the effluent turbidity and the working condition are not qualified, sending a signal to a flocculant dosing control part by a main control center to regulate the flocculant dosing amount; if the flocculant is qualified, keeping the current flocculant dosing amount unchanged.
Furthermore, a working condition monitoring component is used for obtaining an outlet water image of the flocculation tank of the water treatment unit, and the sampling area is arranged in the center of the monitoring picture.
Further, carrying out Gaussian blur processing on the obtained gray level image set to obtain an approximate global illumination distribution map; subtracting the global illumination distribution diagram from the original gray map to obtain an image without illumination influence, and performing binarization processing on the image; and calculating the projection area and the perimeter of the flocs in the binary flocculation image, the average fractal dimension and the average equivalent particle size of all the flocs in the image, and obtaining the morphological feature set of the effluent image.
Further, a findCounters method in OpenCV is used for obtaining a floc outline on the floc binary image, and the projection area S 'of the floc is calculated' i And circumference length of P' i Calculating fractal dimension D 'of flocs' i And equivalent particle diameter R' i (ii) a The average fractal dimension D of all flocs in the image was calculated using the same method i And equivalent particle diameter R i As morphological features of the floc image.
Further, calculating texture feature entropy and correlation of the gray level co-occurrence matrix set and each gray level co-occurrence matrix set respectively for the obtained gray level image set, including: and determining the gray level of the gray level co-occurrence matrix as L, the metering direction and the pixel step distance of single calculation, and calculating the gray level co-occurrence matrix from the original gray level graph.
And further, constructing a learnable sample set by taking a flocculant dosing amount sequence, and a morphological feature set and a texture feature set of the obtained water image as independent variables and taking the corresponding outlet water turbidity as a dependent variable, and dividing the sample set into a training set and a testing set according to the proportion of 7.
Further, the number of initial decision trees and the maximum tree depth of a random forest algorithm are set, regression training and classification testing of a random forest classifier are conducted by using the constructed sample set, and the classifier among the flocculant dosing amount, the effluent image characteristics and the effluent turbidity value is obtained.
And further, taking the turbidity value of the effluent 1 as a boundary threshold value, dividing the turbidity less than or equal to 1 into 'qualified flocculation', and otherwise 'unqualified flocculation', and regulating and controlling the dosage of the flocculating agent.
In another aspect of the present invention, an image-based analysis control apparatus for flocculation conditions according to the method is provided, which includes:
the working condition monitoring component is used for sending the collected water outlet images of the flocculation tanks of the water treatment units under different flocculant dosing quantities into the master control center;
the main control center is used for analyzing morphological characteristics and textural characteristics of flocs in the flocculation image, constructing a machine learning sample set, training and testing a random forest classifier, and obtaining classifiers among different flocculant dosing amounts, effluent image characteristics and effluent turbidity;
the flocculant dosing control part is used for executing a master control center to send a signal for regulating and controlling the dosing amount of the flocculant; and controlling the current dosage of the flocculant.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
according to the invention, after the integral texture features of the floc image are introduced to represent the distribution condition of the flocs on the basis of the individual fractal dimension and the particle size of the flocs, a random forest classifier is used to combine a flocculant dosing amount sequence, the integral texture features of the floc image, the individual characteristics of the flocs and the effluent turbidity thereof to obtain a classification model of the flocculant dosing amount, the effluent image features and the effluent turbidity, and a flocculation working condition query table is generated. And then designed corresponding regulation and control device on the basis of this technique and judged whether qualified in flocculation operating mode, and then regulated and control flocculating agent dosage automatically, realized the stable control effect to water purification process. The method and the device can effectively improve the automation control level of the feed water flocculation process and save human resources at the same time.
The functions of machine vision and image processing automatic monitoring technology are exerted, and new characteristics are introduced to overcome the defects of the prior art. The designed flocculation working condition image type analysis method and device can realize automatic monitoring and regulation of the water supply plant process, not only saves labor cost and ensures the effect of water purification flocculation, but also can maintain the stability of the water purification flocculation process, and have practical value.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention:
FIG. 1 is a block diagram of the process flow of the present invention;
FIG. 2 is a schematic view of the structure of the apparatus of the present invention.
Detailed Description
The invention will be described in detail with reference to the drawings and specific embodiments, which are provided herein for the purpose of illustrating the invention and are not to be construed as limiting the invention.
A schematic diagram in accordance with a disclosed embodiment of the invention is shown in fig. 1. In order to verify the reasonability and the effectiveness of the flocculating agent, an original coagulation water inlet experiment in a certain water treatment plant in Gansu province is selected as a case, the inlet water turbidity distribution of the experimental data is 2.1-20.2, the outlet water turbidity distribution is 0.58-9.92, and the dosage distribution of the flocculating agent is 5.5-16.5 mg/L.
As shown in figure 1, the image type analytic regulation method for flocculation working conditions comprises the following steps:
s101, acquiring different flocculant dosing quantities T by using a working condition monitoring component i (i =1,2, \8230;, k) effluent image A of flocculation tank of water treatment unit under i Obtaining floc image set A = { A = { (A) } 1 ,A 2 ,…,A k }; simultaneously measuring the water outlet turbidity F = { F ] of the water outlet of the flocculation tank under the dosage of each flocculant 1 ,F 2 ,…,F k }。
In order to ensure that the distribution of the flocs in the image sampling area can represent the current flocculation working condition, the selected sampling area is positioned in the central part of the monitoring picture as far as possible.
S102, floc image set A = { A = 1 ,A 2 ,…,A k Performing graying processing as shown in formula (1) to obtain a grayscale image set B = { B = } 1 ,B 2 ,…,B k }. In the formula r i 、g i 、b i The color values of the red, green and blue channels of the color floc image are respectively.
B i ＝r i ×0.299+g i ×0.589+b i ×0.114 (1)
S103, after removing the illumination influence on the gray level image set B obtained in the step S102, performing binarization processing to obtain a binary image set C = { C = { (C) } 1 ,C 2 ,…,C k Calculate each image C i Average fractal dimension and equivalent grain of middle flocsObtaining a morphological feature set D = { D ] of the effluent image 1 ,D 2 ,…,D k And R = { R = } 1 ,R 2 ,…,R k }。
The specific operation of step S103 is:
for the obtained gray image set B i Performing Gaussian blur processing to obtain an approximate global illumination distribution map; subtracting the global illumination distribution diagram from the original gray map to obtain an image without illumination influence, and performing binarization processing on the image; to floc binary image C i Obtaining the outline of the floc by using a findCounters method in OpenCV, and calculating the projection area S 'of the floc' i And circumference length of P' i Calculating fractal dimension D 'of flocs according to formulas (2) and (3)' i And equivalent particle diameter R' i 。
The average fractal dimension D of all flocs in the image was calculated using the same method i And average equivalent particle diameter R i Obtaining the morphological characteristics D = { D) of the effluent image set 1 ,D 2 ,…,D k And R = { R = } 1 ,R 2 ,…,R k }。
S104, calculating the gray level image set B obtained in the step S102 to obtain a gray level co-occurrence matrix set M = { M = { M = } 1 ,M 2 ,…,M k Formula (4), and then calculate each M according to formulas (5) and (6) i The texture entropy and the correlation of the water outlet image are obtained to obtain a texture feature set E = { E = of the water outlet image 1 ,E 2 ,…,E k And G = { G = } 1 ,G 2 ,…,G k }. Wherein L is a grayscale image B i The number of gray levels of (2).
The specific operation of step S104 is:
determining the gray level of the gray level co-occurrence matrix to be L, the metering direction to be 0 degree and the pixel step distance of single calculation to be 1. From the original grey scale chart B i The gray level co-occurrence matrix M of the formula (4) is calculated i 。
Calculating according to the obtained gray level co-occurrence matrix by the formulas (5) and (6) to obtain texture characteristic entropy E i And correlation G i 。
In the formula (6), L is the image gray level, m and n are the horizontal and vertical coordinate values of the gray level co-occurrence matrix, P (m, n) is the value of the gray level co-occurrence matrix at the m and n positions, and U m 、S m Respectively calculating the variance and the expected value of matrix element statistics on the gray level co-occurrence matrix row; u shape n 、S n The variance and expectation of the matrix element statistics on the gray level co-occurrence matrix column are respectively.
U m 、U n 、S m 、S n The calculation method is shown in formulas (7), (8), (9) and (10):
s105, introducing a RandomForest method from a sklern library of python, and adding a flocculant dosage sequence T = { T = T of the flocculant 1 ,T 2 ,…,T k And average fractal dimension D = { D } of flocs obtained in steps S103 and S104 1 ,D 2 ,…,D k Average equivalent particle diameter R = { R = } 1 ,R 2 ,…,R k Texture feature entropy E = { E = } 1 ,E 2 ,…,E k And correlation G = { G = } 1 ,G 2 ,…,G k As argument, with the corresponding effluent turbidity F = { F = } 1 ,F 2 ,…,F k And (5) constructing a learnable sample set as a dependent variable, and dividing the training set and the testing set according to the proportion of 7 by using a train _ test _ split method in a sklern library. Setting the number of initial decision trees of a random forest algorithm to be 50 and the maximum tree depth to be 10, and performing regression training and classification testing on a random forest classifier by using the constructed sample set to obtain a classifier between the flocculant dosing amount T, effluent image characteristics (morphological characteristics D and R, textural characteristics E and G) and an effluent turbidity value F.
S106, summarizing and summarizing according to historical data of effluent image characteristics (D, R, E and G), flocculant dosing quantity T and effluent turbidity F obtained by the working condition monitoring part to obtain a look-up table (table 1) of the dosing quantity, the effluent image characteristics, the effluent turbidity and the flocculation working condition; according to the water treatment standard, taking the turbidity value 1 of the effluent as a boundary threshold value, dividing the turbidity < =1 into 'qualified flocculation', and otherwise 'unqualified flocculation', and regulating and controlling the dosage of the flocculating agent.
S107, after the working condition monitoring part obtains the effluent image of the flocculation tank, morphological characteristics and textural characteristics of the image can be analyzed, the working condition lookup table obtained in the step S106 is inquired, whether the current effluent turbidity and the flocculation working condition are qualified or not can be judged, and a process regulation and control signal is sent to the flocculant dosing control part.
S108, in order to implement the technical steps, an image type analysis regulation and control device of flocculation working conditions is designed, the structure of the device is shown as figure 2, and the device comprises a working condition monitoring part 1, a main control center 2 storing an image type working condition analysis classifier and a flocculant dosing control part 3. The working condition monitoring component is used for sending the collected water outlet images of the flocculation tank of the water treatment unit under different flocculant dosing quantities into a master control center; the main control center is used for analyzing morphological characteristics and textural characteristics of flocs in the flocculation image, constructing a machine learning sample set, training and testing a random forest classifier, and obtaining classifiers among different flocculant dosing amounts, effluent image characteristics and effluent turbidity; the flocculant dosing control component is used for executing a master control center to send a signal for regulating and controlling the dosing amount of the flocculant; and controlling the current dosage of the flocculant.
The device judges the water quality by monitoring the water outlet image of the flocculation tank and automatically adjusts the dosage of the flocculating agent.
When the device works, the current effluent image of the flocculation tank collected by the working condition monitoring part is sent to a master control center for analyzing morphological characteristics and textural characteristics of the image to obtain D j 、R j 、E j 、G j Then the characteristics and the current flocculant dosage T are added j Inputting the water turbidity into a random forest classifier to predict the effluent turbidity F j And judging whether the current effluent turbidity and the flocculation working condition are qualified or not by referring to the working condition lookup table 1. If the flocculant is not qualified, the master control center sends a signal to the flocculant dosing control part to regulate and control the dosing amount of the flocculant; if the flocculant is qualified, keeping the current flocculant dosage unchanged.
TABLE 1 operating mode lookup table
As can be seen from the table 1, by adopting the method provided by the invention, whether the effluent turbidity is qualified or not can be judged according to the effluent floc image obtained by the monitoring part and the morphological characteristics and the textural characteristics in the image characteristics. According to the judgment result of the classifier, when the fractal dimension of the effluent floc image is 1.226-1.324, the equivalent particle size is 52.61-62.14, the texture entropy is 1.275-1.453, and the texture correlation is 0.148-0.175, the effluent turbidity corresponding to the effluent image is less than or equal to 1, the flocculation requirement is met, and the dosage of the flocculant is not required to be adjusted; when the fractal dimension of the effluent floc image is 1.324-2.226, the equivalent particle size is 62.14-74.82, the texture entropy is 1.453-1.871 and the texture correlation is 0.067-0.148, the effluent turbidity corresponding to the effluent image is greater than 1 and does not meet the flocculation requirement, the main control center sends a signal to the flocculant dosing component at the moment and the flocculant dosing amount is increased to meet the flocculation requirement.
The present invention is not limited to the above embodiments, and based on the technical solutions disclosed in the present invention, those skilled in the art can make some substitutions and modifications to some technical features without creative efforts based on the disclosed technical contents, and these substitutions and modifications are all within the protection scope of the present invention.
Claims (10)
1. An image type analytic regulation and control method for flocculation working condition is characterized by comprising the following steps:
acquiring water outlet images of a flocculation tank of the water treatment unit under different flocculant dosing amounts to obtain a floc image set; meanwhile, the effluent turbidity of the water outlet of the flocculation tank under the dosage of each flocculant is manually measured;
graying the floc image set to obtain a gray image set;
carrying out binarization processing on the obtained gray level image set after removing the influence of illumination to obtain a binary image set, and calculating the average fractal dimension and the equivalent particle size of flocs in each image in the binary image set to obtain a morphological feature set of a water outlet image;
respectively calculating texture feature entropies and correlations of the gray level co-occurrence matrix set and each gray level co-occurrence matrix set to the obtained gray level image set to obtain a texture feature set of the effluent image;
constructing a forest classifier model by using the flocculant dosing amount sequence, the morphological feature set and the textural feature set of the effluent image and the effluent turbidity of the water outlet of the flocculation tank, and training a random forest classifier; obtaining a regression classifier among the dosage of the flocculating agent, the effluent image characteristics and the effluent turbidity;
generating a flocculant dosing amount, effluent image characteristics and effluent turbidity flocculation working condition lookup table, judging whether the effluent turbidity and the working condition of the current flocculation tank are qualified or not, and if not, sending a signal to a flocculant dosing control part by a main control center to regulate the flocculant dosing amount; if the flocculant is qualified, keeping the current flocculant dosing amount unchanged.
2. The image type analysis regulation and control method of flocculation working condition according to claim 1, characterized in that the working condition monitoring component is used to obtain the flocculation tank effluent image of the water treatment unit, and the sampling area is in the center position of the monitoring picture.
3. The image type analysis regulation and control method for the flocculation working condition according to claim 1, wherein the obtained gray level image set is subjected to Gaussian blur processing to obtain an approximate global illumination distribution map; subtracting the global illumination distribution graph from the original gray graph to obtain an image without illumination influence, and carrying out binarization processing on the image;
and calculating the projection area and the perimeter of the flocs in the binary flocculation image, the average fractal dimension and the average equivalent particle size of all the flocs in the image, and obtaining the morphological feature set of the effluent image.
4. The method as claimed in claim 3, wherein the floc binary image is subjected to findCounterers in OpenCV to obtain a floc profile, and a projected area S 'of the floc is calculated' i And circumference length of P' i Calculating fractal dimension D 'of flocs' i And equivalent particle diameter R' i (ii) a The average fractal dimension D of all flocs in the image was calculated using the same method i And equivalent particle diameter R i As morphological features of the floc image.
5. The image type analysis regulation and control method for the flocculation working condition according to claim 1, wherein texture feature entropy and correlation of the gray level co-occurrence matrix set and each gray level co-occurrence matrix set are respectively calculated for the obtained gray level image set, and the method comprises the following steps: and determining the gray level of the gray level co-occurrence matrix as L, the metering direction and the pixel step distance of single calculation, and calculating the gray level co-occurrence matrix from the original gray level graph.
6. The image type analysis regulation and control method for flocculation working condition according to claim 5, characterized in that texture characteristic entropy E of each gray level co-occurrence matrix set i And correlation G i The calculation is as follows:
wherein L is the image gray level, m and n are the horizontal and vertical coordinate values of the gray level co-occurrence matrix, P (m, n) is the value of the gray level co-occurrence matrix at the position of m and n, and U is the value of the gray level co-occurrence matrix at the position of m and n m 、S m Respectively calculating the variance and the expected value of matrix elements on the gray level co-occurrence matrix row; u shape n 、S n The variance and expectation of the matrix element statistics on the gray level co-occurrence matrix are respectively.
7. The image type analysis regulation and control method of the flocculation working condition according to claim 1, characterized in that a flocculant dosing amount sequence, a morphological feature set and a texture feature set of an obtained effluent image are used as independent variables, a learnable sample set is constructed by taking the corresponding outlet turbidity as a dependent variable, and the sample set is divided into a training set and a testing set according to a ratio of 7.
8. The method for image type analysis regulation and control of flocculation working conditions according to claim 1, wherein the number of initial decision trees and the maximum tree depth of a random forest algorithm are set, and the constructed sample set is used for regression training and classification testing of a random forest classifier to obtain the classifier between flocculant dosage, effluent image characteristics and effluent turbidity value.
9. The image type analysis regulation and control method for flocculation working conditions according to claim 1, characterized in that the turbidity value 1 of the effluent is taken as a boundary threshold value, the turbidity is less than or equal to 1 is classified as "qualified flocculation", otherwise "unqualified flocculation", and the dosage of the flocculating agent is regulated and controlled.
10. An image-based analytic regulation and control device for flocculation condition of any one of claims 1-9, comprising:
the working condition monitoring component is used for sending the collected water outlet images of the flocculation tank of the water treatment unit under different flocculant dosing quantities into a master control center;
the main control center is used for analyzing morphological characteristics and textural characteristics of flocs in the flocculation image, constructing a machine learning sample set, training and testing a random forest classifier, and obtaining classifiers among different flocculant dosing amounts, effluent image characteristics and effluent turbidity;
the flocculant dosing control part is used for executing a master control center to send a signal for regulating and controlling the dosing amount of the flocculant; and controlling the current dosage of the flocculant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211222711.4A CN115546720A (en) | 2022-10-08 | 2022-10-08 | Image type analytic regulation and control method and device for flocculation working condition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211222711.4A CN115546720A (en) | 2022-10-08 | 2022-10-08 | Image type analytic regulation and control method and device for flocculation working condition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115546720A true CN115546720A (en) | 2022-12-30 |
Family
ID=84730991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211222711.4A Pending CN115546720A (en) | 2022-10-08 | 2022-10-08 | Image type analytic regulation and control method and device for flocculation working condition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115546720A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115760852A (en) * | 2023-01-06 | 2023-03-07 | 青岛市城市规划设计研究院 | Marine sewage discharge treatment method |
| CN116768346A (en) * | 2023-08-23 | 2023-09-19 | 四川省每文环保科技有限公司 | Sewage treatment process control method based on pumping flocculation filtration |
| CN117023741A (en) * | 2023-08-14 | 2023-11-10 | 佛山市禅城区供水有限公司 | Multi-parameter load composite water treatment method and system in flocculation process |
-
2022
- 2022-10-08 CN CN202211222711.4A patent/CN115546720A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115760852A (en) * | 2023-01-06 | 2023-03-07 | 青岛市城市规划设计研究院 | Marine sewage discharge treatment method |
| CN117023741A (en) * | 2023-08-14 | 2023-11-10 | 佛山市禅城区供水有限公司 | Multi-parameter load composite water treatment method and system in flocculation process |
| CN117023741B (en) * | 2023-08-14 | 2024-03-19 | 佛山市禅城区供水有限公司 | Multi-parameter load composite water treatment method and system in flocculation process |
| CN116768346A (en) * | 2023-08-23 | 2023-09-19 | 四川省每文环保科技有限公司 | Sewage treatment process control method based on pumping flocculation filtration |
| CN116768346B (en) * | 2023-08-23 | 2023-12-12 | 四川省每文环保科技有限公司 | Sewage treatment process control method based on pumping flocculation filtration |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115546720A (en) | Image type analytic regulation and control method and device for flocculation working condition | |
| CN109074033B (en) | Method and system for optimizing coagulation and/or flocculation in water treatment processes | |
| CN111966053A (en) | Intelligent flocculant decision making system | |
| CN115619998B (en) | Image-based sewage grade identification method and device | |
| CN115147617B (en) | Intelligent monitoring method for sewage treatment based on computer vision | |
| CN107274403B (en) | Evaluation method for flotation surface quality | |
| JP7083269B2 (en) | Control method and control device for water treatment plants with agglomeration | |
| CN115760852B (en) | Marine sewage discharge treatment method | |
| CN116416555A (en) | Information processing method, device and equipment based on image analysis | |
| CN110378887A (en) | Screen defect inspection method, apparatus and system, computer equipment and medium | |
| CN116012701A (en) | Water treatment dosing control method and device based on alum blossom detection | |
| CN109509173A (en) | A kind of method of counting and counting device of bacterial clump | |
| CN112132700A (en) | Formula decision method for regulating and controlling water quality | |
| CN116485801B (en) | Rubber tube quality online detection method and system based on computer vision | |
| CN117455920B (en) | Artificial intelligence-based milk tea cup inferior product screening method and system | |
| JPH05263411A (en) | Object observation method and device | |
| CN113838081A (en) | Method and device for distinguishing color uniformity of flue-cured tobacco leaves based on machine vision | |
| CN112766384A (en) | Deep learning-based algae floc removal rate evaluation method | |
| CN109657731A (en) | A kind of anti-interference classification method of droplet digital pcr instrument | |
| CN117423046A (en) | Visual detection method for cement mortar stirring process based on image processing | |
| CN113505649A (en) | Tap water chlorination control method and device | |
| KR20190063188A (en) | Method for controlling water purification using real-time image analysis | |
| CN115661081B (en) | Automatic control water changing method for aquaculture based on image processing | |
| CN106898010A (en) | Particle copies the method and device planted | |
| CN110827267B (en) | Vinasse blending quality visual analysis method and system in brewing process of raw wine in white spirit factory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |