CN114239387A - Organic biochemical reaction strain putting guidance method and system - Google Patents
Organic biochemical reaction strain putting guidance method and system Download PDFInfo
- Publication number
- CN114239387A CN114239387A CN202111443428.XA CN202111443428A CN114239387A CN 114239387 A CN114239387 A CN 114239387A CN 202111443428 A CN202111443428 A CN 202111443428A CN 114239387 A CN114239387 A CN 114239387A
- Authority
- CN
- China
- Prior art keywords
- water quality
- model
- sewage treatment
- modeling
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000005842 biochemical reaction Methods 0.000 title claims abstract description 22
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 205
- 239000010865 sewage Substances 0.000 claims abstract description 96
- 238000012544 monitoring process Methods 0.000 claims abstract description 60
- 230000007246 mechanism Effects 0.000 claims abstract description 48
- 238000004140 cleaning Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 10
- 238000006731 degradation reaction Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000004891 communication Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 9
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 8
- 229910052760 oxygen Inorganic materials 0.000 claims description 8
- 239000001301 oxygen Substances 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 2
- -1 UV254 Chemical compound 0.000 claims 1
- 238000005406 washing Methods 0.000 claims 1
- 230000008569 process Effects 0.000 description 11
- 230000000694 effects Effects 0.000 description 5
- 238000005273 aeration Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 230000001580 bacterial effect Effects 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 239000005416 organic matter Substances 0.000 description 2
- 230000035484 reaction time Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000004065 wastewater treatment Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Water Supply & Treatment (AREA)
- Mathematical Physics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Activated Sludge Processes (AREA)
Abstract
The invention discloses an organic biochemical reaction strain putting guidance method and a system, wherein the method comprises the following steps: collecting historical water quality monitoring data and performing data cleaning on part of the data; establishing a mechanism water quality modeling and rating model, and adjusting parameters of the mechanism water quality modeling and rating model through the cleaned historical water quality monitoring data; establishing an expert database based on a mechanism water quality modeling calibration model, wherein the expert database is used for simulating corresponding relations among different water quality situations, hydrodynamic situations, sewage treatment strain input amounts and water quality standard reaching time; establishing an AI rating modeling model through various AI models, and training the AI rating modeling model through data in an expert database; and guiding the sewage treatment strain input amount at the edge end of the sewage treatment site through a trained AI calibration modeling model. The invention establishes an expert database and integrates a multi-source heterogeneous model to guide the putting of the strains in the organic biochemical reaction, can guide the putting amount of the strains in the sewage treatment in real time at the field end, and has more practical guidance function.
Description
Technical Field
The invention belongs to the field of water resource monitoring and management, and particularly relates to a method and a system for guiding the putting of an organic biochemical reaction strain.
Background
In a sewage environment, soluble organic matters can cause the problems of color, smell, turbidity and the like of water quality, and play an important role in the migration and conversion of micro pollutants in a water body. Soluble organic matter is a substance of common interest in sewage treatment, and the treatment of such substances is particularly important in sewage treatment processes.
The conventional sewage treatment method for treating the soluble organic matters adopts an aeration biological filter method, and the method is also known as the method for treating the soluble organic matters with the best effect. Before and after sewage treatment, various water quality indexes are monitored, wherein the BOD water quality index detection principle is close to the aeration biological filter method principle, and the state of the aeration biological filter can be guided by the comprehensive organic matter monitoring index conditions such as BOD and the like. These methods tend to have hysteresis which is not conducive to rapid wastewater treatment.
In the prior art, a method for predicting and treating related sewage indexes by using an intelligent algorithm partially also appears, but most of the methods hardly produce actual effects in specific applications, input parameters are not judged, a model is easy to distort, the sewage treatment time cannot be effectively controlled, and the method basically has no actual application conditions.
Disclosure of Invention
In view of the above, the invention provides an organic biochemical reaction strain release guidance method and system, which are used for solving the problem that the prior art cannot guide the release amount of organic biochemical reaction strains in real time.
The invention discloses a method for guiding the putting of organic biochemical reaction strains in a first aspect, which comprises the following steps:
collecting historical water quality monitoring data and performing data cleaning on part of the data;
establishing a mechanism water quality modeling and rating model, and adjusting parameters of the mechanism water quality modeling and rating model through the cleaned historical water quality monitoring data;
establishing an expert database based on a mechanism water quality modeling calibration model, wherein the expert database is used for simulating corresponding relations among different water quality situations, hydrodynamic situations, sewage treatment strain input amounts and water quality standard reaching time;
establishing an AI calibration modeling model through a plurality of AI models, and training the AI calibration modeling model by taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in an expert database as input and corresponding sewage treatment strain input amount as output;
and guiding the sewage treatment strain input amount at the edge end of the sewage treatment site through a trained AI calibration modeling model.
Preferably, the historical water quality monitoring data comprises historical monitoring time sequence data of various water quality indexes, corresponding historical sewage treatment strain putting data, corresponding hydrodynamic historical monitoring time sequence data and corresponding historical degradation process data of the water quality indexes; the various water quality indicators include, but are not limited to, COD, dissolved oxygen, UV254, and BOD; the hydrodynamic historical monitoring time series data includes, but is not limited to, water level, flow rate, and flow profile data.
Preferably, the data cleansing specifically includes: and cleaning historical monitoring time sequence data and hydrodynamic historical monitoring time sequence data of various water quality indexes by a data cleaning tool, wherein the data cleaning tool comprises a filtering tool which is not limited to Carl diffusion, a filtering algorithm, a fuzzy theory and a principal component analysis.
Preferably, the adjusting of the parameter setting of the mechanism water quality modeling and calibration model by the washed historical water quality monitoring data specifically includes:
establishing a mechanism water quality modeling and rating model, wherein the mechanism water quality modeling and rating model comprises but is not limited to a water quality complete mixing model, CE-QUAL-W2, EFDC, WASP, DELFT3D or HEC-RAS;
inputting the cleaned historical monitoring time sequence data of various water qualities, corresponding historical sewage treatment strain putting data and corresponding cleaned historical monitoring time sequence data of hydrodynamic force into a mechanism water quality modeling calibration model, and acquiring water quality index simulation degradation process data output by the mechanism water quality modeling calibration model;
and carrying out calibration verification circulation through the historical degradation process data of the water quality indexes and the simulated degradation process of the water quality indexes, and adjusting the parameter setting of the mechanism water quality modeling calibration model until the model reaches a preset standard efficiency coefficient.
Preferably, the establishment of the expert database based on the mechanism water quality modeling calibration model is used for simulating the corresponding relationship among different water quality situations, hydrodynamic situations, sewage treatment strain input amount and water quality standard reaching time, and specifically comprises the following steps:
respectively setting the solubility of various water quality indexes under different water quality situations, the sewage treatment strain putting amount under different sewage treatment strain putting situations and situation parameters under different hydrodynamic situations;
inputting the solubility of various water quality indexes under different water quality situations, the sewage treatment strain input amount under different sewage treatment strain input situations and condition parameters under different hydrodynamic situations into a mechanism water quality modeling calibration model after parameter adjustment, and respectively outputting water quality index degradation process data under corresponding situations;
and analyzing the water quality standard-reaching time under the corresponding situation according to the data of the water quality index degradation process under the corresponding situation to obtain the corresponding relation among different water quality situations, hydrodynamic situations, the sewage treatment strain input amount and the water quality standard-reaching time.
Preferably, the establishing of the AI rating modeling model through a plurality of AI models takes different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in the expert database as input, takes corresponding sewage treatment strain input amount as output, and the training of the AI rating modeling model specifically comprises:
obtaining a plurality of AI models to form an AI rating modeling model, wherein the AI models comprise but are not limited to a support vector machine, a K nearest neighbor method, a random gradient descent, a multivariate linear regression, a multilayer perceptron, a decision tree, a back propagation neural network and a radial basis function network;
inputting different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in the expert database into each AI model respectively, calculating and outputting sewage treatment strain input analog quantity according to the output results of the AI models by an ensemble averaging method, carrying out calibration verification circulation according to the sewage treatment strain input analog quantity and the corresponding sewage treatment strain input quantity in the expert database, and adjusting the parameter setting of the AI calibration modeling model according to the corresponding sewage treatment strain input quantity in the expert database.
Preferably, the guiding of the sewage treatment strain input amount by the trained AI calibration modeling model at the edge of the sewage treatment site specifically comprises:
deploying the trained AI calibration modeling model at the edge end of each sewage treatment site to serve as an edge calculation AI model;
acquiring real-time monitoring time sequence data and hydrodynamic real-time monitoring time sequence data of various water quality indexes at an edge end and performing data cleaning;
setting the time for the expected water quality to reach the standard;
inputting the cleaned real-time monitoring time sequence data, hydrodynamic real-time monitoring time sequence data and expected water quality standard reaching time of various water quality indexes into an edge calculation AI model, and outputting a real-time guiding value of the sewage treatment strain input quantity.
In a second aspect of the present invention, a system for guiding the release of organic biochemical reaction strains is disclosed, the system comprising:
a data collection module: the system is used for collecting historical water quality monitoring data and cleaning partial data;
a parameter adjusting module: the water quality model establishing and calibrating method is used for establishing a mechanism water quality modeling and calibrating model, and parameters of the mechanism water quality modeling and calibrating model are adjusted through the cleaned historical water quality monitoring data;
an expert database establishing module: the method is used for establishing an expert database based on a mechanism water quality modeling calibration model and simulating corresponding relations among different water quality situations, hydrodynamic situations, sewage treatment strain input quantity and water quality standard reaching time;
a model construction module: the system is used for establishing an AI calibration modeling model through a plurality of AI models, and training the AI calibration modeling model by taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in an expert database as input and corresponding sewage treatment strain input amount as output;
a real-time guidance module: the method is used for guiding the sewage treatment strain input amount at the edge end of a sewage treatment site through a trained AI calibration modeling model.
In a third aspect of the present invention, an electronic device is disclosed, comprising: at least one processor, at least one memory, a communication interface, and a bus; the processor, the memory and the communication interface complete mutual communication through the bus; the memory stores program instructions executable by the processor, which program instructions are invoked by the processor to implement the method according to the first aspect of the invention.
In a fourth aspect of the invention, a computer-readable storage medium is disclosed, which stores computer instructions for causing a computer to implement the method according to the first aspect of the invention.
Compared with the prior art, the invention has the following beneficial effects:
1) the invention adds a data cleaning tool of water monitoring data, has the capacity of deeply exploring meaningful information of data, improves the fault-tolerant capacity of the model, establishes a mechanism water quality modeling calibration model, carries out digital twinning by means of the mechanism water quality modeling calibration model so as to simulate the corresponding relation between different water quality situations, hydrodynamic situations, sewage treatment strain input quantity and water quality standard reaching time, sets various situations to simulate mirror-phase twinning situations when the historical observation data are insufficient, establishes a situation expert database of multi-source heterogeneous data, and provides rich data support for organic biochemical reaction strain input guidance;
2) the invention uses a plurality of AI models to establish an AI calibration modeling model and is used for guiding the strain input amount, can solve the problem that a single model has weak generalization performance under different situations, and finally obtains a stable, reliable, practical and intelligent decision-making operation scheme through a set averaging method.
3) The invention deploys the trained AI calibration modeling model at each sewage treatment site edge end, has the function of decentralized edge calculation, can input AI model parameters into a site end chip, can perform distributed calculation and decision at the site end, can reduce the pressure of a central server, can improve the sewage treatment response speed of each edge end, reduces the hysteresis problem caused by conventional water quality measurement and strain input calculation, and improves the actual sewage treatment effect.
4) The method fully considers the influence of different water quality situations and different hydrodynamic force situations on the put-in amount of the strains for sewage treatment, simultaneously considers the requirement of the water quality standard-reaching time in practical application, controls the water quality standard-reaching time to be included in strain putting guidance, has more active operation, and can specifically guide the important treatment process in the sewage treatment link, including the setting of the water quality standard-reaching time and the reaction time, so as to achieve the practical guidance effect of sewage treatment.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a general flowchart of an organic biochemical reaction bacterial spawn feeding guidance method of the present invention;
FIG. 2 is a flow chart of the establishment steps of a mechanism water quality modeling calibration model;
FIG. 3 is a flow chart of expert database creation;
FIG. 4 is a flowchart of the AI rating modeling model building steps;
FIG. 5 is an AI rating modeling model architecture diagram;
FIG. 6 is a flowchart of an edge calculation application process.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
Referring to fig. 1, the present invention provides a method for guiding the release of organic biochemical reaction strains, which comprises:
step 1, collecting historical water quality monitoring data, wherein the historical water quality monitoring data is collected, and partial data is subjected to data cleaning.
The historical water quality monitoring data comprises historical monitoring time sequence data of various water quality indexes, corresponding historical sewage treatment strain putting data, corresponding hydrodynamic historical monitoring time sequence data and corresponding historical degradation process data of the water quality indexes; the various water quality indicators include, but are not limited to, COD, dissolved oxygen, UV254, and BOD; the hydrodynamic historical monitoring time series data includes, but is not limited to, water level, flow rate, and flow profile data.
The specific steps of step 1 are as follows:
step 1.1, collecting historical monitoring time sequence data of various water quality indexes, including but not limited to COD, dissolved oxygen, UV254, BOD and the like.
And step 1.2, inputting historical monitoring time series data of multiple water quality indexes collected in the step 1.1 into a data cleaning tool to correct unreasonable data, wherein the data cleaning tool comprises but is not limited to Carl diffusion filtering, a filtering algorithm, a fuzzy theory, principal component analysis and the like.
And step 1.3, collecting historical sewage treatment strain putting data corresponding to historical monitoring time sequence data of various water quality indexes.
And 1.4, collecting hydrodynamic historical monitoring time sequence data corresponding to the historical monitoring time sequence data of various water quality indexes, wherein the hydrodynamic historical monitoring time sequence data comprises data such as water level, flow rate and overflow section data.
And step 1.5, inputting the hydrodynamic historical monitoring time sequence data collected in the step 1.4 into the data cleaning tool to correct the problem of unreasonable data.
And step 1.6, collecting historical degradation process data of the water quality index corresponding to the historical sewage treatment strain putting data.
The collected historical water quality monitoring data can be used in the establishment process of the mechanism water quality modeling calibration model in the step 2.
And 2, establishing a mechanism water quality modeling and rating model, wherein the mechanism water quality modeling and rating model is established, and parameters of the mechanism water quality modeling and rating model are adjusted through the cleaned historical water quality monitoring data.
As shown in fig. 2, the specific steps of step 2 are as follows:
step 2.1, the invention uses 0-dimensional, 1-dimensional, 2-dimensional and 3-dimensional water quality models including but not limited to a water quality complete mixing model, CE-QUAL-W2, EFDC, WASP, DELFT3D, HEC-RAS and the like to model and rate the mechanism water quality. And inputting the achievements in the step 1.2, the step 1.3 and the step 1.5 into a mechanism water quality modeling calibration model, and outputting the data of the simulated degradation process of the water quality index.
And 2.2, carrying out a calibration verification circulation process through the historical degradation process data of the water quality index in the step 1.6 and the simulated degradation process data of the water quality index in the step 2.1, and adjusting the parameter setting of the mechanism water quality modeling calibration model until the model reaches the available standard efficiency coefficient of more than 0.5.
The mechanism water quality modeling calibration model for completing calibration verification can be used in the expert database establishing process of the step 3.
And 3, establishing an expert database, namely establishing the expert database based on a mechanism water quality modeling calibration model, wherein the expert database is used for simulating the corresponding relation among different water quality situations, hydrodynamic situations, sewage treatment strain input quantity and water quality standard reaching time.
As shown in fig. 3, the specific steps of step 3 are as follows:
and 3.1, establishing a plurality of water quality situations, and setting different water quality index solubilities including but not limited to COD (chemical oxygen demand), dissolved oxygen, UV254, BOD (biochemical oxygen demand) and the like.
And 3.2, establishing different sewage treatment strain putting situations and setting different strain putting quantities.
And 3.3, establishing different hydrodynamic situations, and setting parameters of different conditions including but not limited to water level, flow rate, overflow section data and the like.
And 3.4, inputting the achievements of the step 3.1, the step 3.2 and the step 3.3 into the mechanism water quality modeling calibration model which completes calibration verification in the step 2.
And 3.5, outputting the degradation process data of the water quality index under the corresponding situation by the step 3.4.
And 3.6, analyzing the water quality standard reaching time under the corresponding situation by the result of the step 3.5.
And 3.7, obtaining the corresponding relation among different water quality situations, different hydrodynamic force situations and different sewage treatment strain putting situations and the water quality standard reaching time through the step 3.1, the step 3.2, the step 3.3 and the step 3.6, and completing the establishment of the expert database.
The established expert library may be used in the AI model modeling process of step 4.
The invention carries out digital twinning by means of a mechanism water quality modeling calibration model so as to simulate the corresponding relation between different water quality situations, hydrodynamic force situations, sewage treatment strain input quantity and water quality standard reaching time, solves the problem that when historical observation data is insufficient, sets various situations to simulate the microscopic twin situation, establishes a situation expert database of multi-source heterogeneous data, and provides rich data support for organic biochemical reaction strain input guidance.
And 4, establishing an AI rating modeling model, namely establishing the AI rating modeling model through various AI models, and training the AI rating modeling model by taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in the expert database as input and corresponding sewage treatment strain input amount as output.
As shown in fig. 4, the specific steps of step 4 are as follows:
and 4.1, acquiring data of different water quality standard reaching times, various water quality situations and different hydrodynamic situations in the expert database completed in the step 3.
And 4.2, inputting the results of the step 4.1 into an AI rating modeling model. The architecture of the AI model is shown in fig. 5, and the various AI models include, but are not limited to, support vector machines, K-nearest neighbor, stochastic gradient descent, multivariate linear regression, multi-layer perceptrons, decision trees, back-propagation neural networks, radial basis function networks; and (4) respectively inputting the results obtained in the step (4.1) into each AI model, and calculating the output sewage treatment strain input analog quantity of the output AI model by a set-averaging method.
And 4.3, acquiring sewage treatment strain putting situations corresponding to the data of different water quality standard reaching times, various water quality situations and different hydrodynamic situations in the expert database in the step 3.
And 4.4, carrying out calibration verification circulation process through the sewage treatment strain putting situation of the step 4.3 and the sewage treatment strain putting analog quantity of the step 4.2, and adjusting parameter setting of the AI calibration modeling model through the corresponding sewage treatment strain putting quantity in the expert library, such as the influence weight coefficient of each AI model on the sewage treatment strain putting quantity, the parameters of the AI model, and the like. Until the model reaches a usable standard efficiency factor of above 0.5.
The AI rating modeling model that completed rating verification is used in the step 5 edge calculation application process.
The invention uses a plurality of AI models to establish the AI rating modeling model and is used for guiding the bacterial input amount, can solve the problem that a single model has weak generalization performance under different situations, adjusts the parameters of the AI rating modeling model by taking data in an expert database as training data, and finally obtains a stable, reliable, practical and intelligent decision-making operation scheme through a set averaging method.
And 5, an edge calculation application step, which is used for guiding the sewage treatment strain input amount through a trained AI calibration modeling model at the edge end of the sewage treatment site.
Deploying the trained AI rating modeling model at the edge end of each sewage treatment site as an edge calculation AI model, wherein the steps of the edge calculation application process are shown in FIG. 6, and the specific steps are as follows:
and 5.1, collecting various real-time water quality monitoring time sequence data including but not limited to COD, dissolved oxygen, UV254, BOD and the like.
And 5.2, inputting the various real-time water quality monitoring time sequence data collected in the step 5.1 into the data cleaning tool to correct the unreasonable data.
And 5.3, setting the time for the expected water quality to reach the standard.
And 5.4, collecting the hydrodynamic real-time monitoring time sequence data including but not limited to water level, flow rate, overflow section data and other data.
And 5.5, inputting the hydrodynamic real-time monitoring time sequence data collected in the step 5.4 into the data cleaning tool to correct the unreasonable data.
And 5.6, inputting the results of the step 5.2, the step 5.3 and the step 5.5 into the edge calculation AI model.
And 5.7, obtaining a real-time guide value of the sewage treatment strain input amount from the step 5.6.
The method fully considers the influence of different water quality situations and different hydrodynamic force situations on the put-in amount of the strains for sewage treatment, simultaneously considers the requirement of the water quality standard-reaching time in practical application, controls the water quality standard-reaching time to be included in strain putting guidance, has more active operation, and can specifically guide the important treatment process in the sewage treatment link, including the setting of the water quality standard-reaching time and the reaction time, so as to achieve the practical guidance effect of sewage treatment. Edge calculation is carried out on the edge end of the sewage treatment site through a trained AI calibration modeling model, the pressure of a data calculation center server is reduced through decentralization, meanwhile, the sewage treatment strain input amount is conveniently guided in real time at the site end, and the sewage treatment response speed is improved.
Corresponding to the embodiment of the method, the invention also provides an organic biochemical reaction strain release guidance system based on multi-source heterogeneous model fusion, which comprises the following steps:
a data collection module: the system is used for collecting historical water quality monitoring data and cleaning the data;
a parameter adjusting module: the water quality model establishing and calibrating method is used for establishing a mechanism water quality modeling and calibrating model, and parameters of the mechanism water quality modeling and calibrating model are adjusted through the cleaned historical water quality monitoring data;
an expert database establishing module: the method is used for establishing an expert database based on a mechanism water quality modeling calibration model and simulating corresponding relations among different water quality situations, hydrodynamic situations, sewage treatment strain input quantity and water quality standard reaching time;
a model construction module: the system is used for establishing an AI calibration modeling model through a plurality of AI models, and training the AI calibration modeling model by taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in an expert database as input and corresponding sewage treatment strain input amount as output;
a real-time guidance module: and guiding the sewage treatment strain input amount at the edge end of the sewage treatment site through a trained AI calibration modeling model.
The above method embodiments and system embodiments are in one-to-one correspondence, and please refer to the method embodiments for brief description of the system embodiments.
The present invention also discloses an electronic device, comprising: at least one processor, at least one memory, a communication interface, and a bus; the processor, the memory and the communication interface complete mutual communication through the bus; the memory stores program instructions executable by the processor, which invokes the program instructions to implement the methods of the invention described above.
The invention also discloses a computer readable storage medium which stores computer instructions for causing the computer to implement all or part of the steps of the method of the embodiment of the invention. The storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk, or an optical disk.
The above-described system embodiments are merely illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts shown as units may or may not be physical units, i.e. may be distributed over a plurality of network units. Without creative labor, a person skilled in the art can select some or all of the modules according to actual needs to achieve the purpose of the solution of the embodiment.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A method for guiding the putting of organic biochemical reaction strains is characterized by comprising the following steps:
collecting historical water quality monitoring data and performing data cleaning on part of the data;
establishing a mechanism water quality modeling and rating model, and adjusting parameters of the mechanism water quality modeling and rating model through the cleaned historical water quality monitoring data;
establishing an expert database based on a mechanism water quality modeling calibration model, wherein the expert database is used for simulating corresponding relations among different water quality situations, hydrodynamic situations, sewage treatment strain input amounts and water quality standard reaching time;
establishing an AI calibration modeling model through a plurality of AI models, and training the AI calibration modeling model by taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in an expert database as input and corresponding sewage treatment strain input amount as output;
and guiding the sewage treatment strain input amount at the edge end of the sewage treatment site through a trained AI calibration modeling model.
2. The method for guiding the release of organic biochemical reaction strains according to claim 1, wherein the historical water quality monitoring data comprises historical monitoring time sequence data of a plurality of water quality indexes, corresponding historical sewage treatment strain release data, corresponding hydrodynamic historical monitoring time sequence data, and corresponding historical degradation process data of the water quality indexes; the various water quality indicators include, but are not limited to, COD, dissolved oxygen, UV254, and BOD; the hydrodynamic historical monitoring time series data includes, but is not limited to, water level, flow rate, and flow profile data.
3. The method for guiding the release of organic biochemical reaction strains according to claim 2, wherein the step of performing data washing on the part specifically comprises: and cleaning historical monitoring time sequence data and hydrodynamic historical monitoring time sequence data of various water quality indexes by a data cleaning tool, wherein the data cleaning tool comprises but is not limited to Carl diffuse filtering, a filtering algorithm, fuzzy theory and principal component analysis.
4. The method for guiding the release of organic biochemical reaction strains according to claim 2, wherein the step of adjusting the parameters of the mechanism water quality modeling calibration model according to the washed historical water quality monitoring data specifically comprises the steps of:
establishing a mechanism water quality modeling and rating model, wherein the mechanism water quality modeling and rating model comprises but is not limited to a water quality complete mixing model, CE-QUAL-W2, EFDC, WASP, DELFT3D or HEC-RAS;
inputting the cleaned historical monitoring time sequence data of various water qualities, corresponding historical sewage treatment strain putting data and corresponding cleaned historical monitoring time sequence data of hydrodynamic force into a mechanism water quality modeling calibration model, and acquiring water quality index simulation degradation process data output by the mechanism water quality modeling calibration model;
and carrying out calibration verification circulation through the historical degradation process data of the water quality indexes and the simulated degradation process of the water quality indexes, and adjusting the parameter setting of the mechanism water quality modeling calibration model until the model reaches a preset standard efficiency coefficient.
5. The method for guiding the release of organic biochemical reaction strains according to claim 4, wherein the establishment of an expert database based on a mechanism water quality modeling and calibration model for simulating the corresponding relationship between different water quality situations, hydrodynamic situations, sewage treatment strain release amounts and water quality standard reaching time specifically comprises:
respectively setting the solubility of various water quality indexes under different water quality situations, the sewage treatment strain putting amount under different sewage treatment strain putting situations and situation parameters under different hydrodynamic situations;
inputting the solubility of various water quality indexes under different water quality situations, the sewage treatment strain input amount under different sewage treatment strain input situations and condition parameters under different hydrodynamic situations into a mechanism water quality modeling calibration model after parameter adjustment, and respectively outputting water quality index degradation process data under corresponding situations;
and analyzing the water quality standard-reaching time under the corresponding situation according to the data of the water quality index degradation process under the corresponding situation to obtain the corresponding relation among different water quality situations, hydrodynamic situations, the sewage treatment strain input amount and the water quality standard-reaching time.
6. The method for guiding the delivery of organic biochemical reaction strains according to claim 1, wherein the AI rating modeling model is established by a plurality of AI models, and the training of the AI rating modeling model specifically comprises the following steps, taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching times in an expert database as input, and taking corresponding sewage treatment strain delivery amount as output:
obtaining a plurality of AI models to form an AI rating modeling model, wherein the AI models comprise but are not limited to a support vector machine, a K nearest neighbor method, a random gradient descent, a multivariate linear regression, a multilayer perceptron, a decision tree, a back propagation neural network and a radial basis function network;
inputting different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in the expert database into each AI model respectively, calculating and outputting sewage treatment strain input analog quantity according to the output results of the AI models by an ensemble averaging method, carrying out calibration verification circulation according to the sewage treatment strain input analog quantity and the corresponding sewage treatment strain input quantity in the expert database, and adjusting the parameter setting of the AI calibration modeling model according to the corresponding sewage treatment strain input quantity in the expert database.
7. The method for guiding the release of organic biochemical reaction strains according to claim 6, wherein the guiding of the release amount of the strains in the sewage treatment by the trained AI calibration modeling model at the edge of the sewage treatment site specifically comprises:
deploying the trained AI calibration modeling model at the edge end of each sewage treatment site to serve as an edge calculation AI model;
acquiring real-time monitoring time sequence data and hydrodynamic real-time monitoring time sequence data of various water quality indexes at an edge end and performing data cleaning;
setting the time for the expected water quality to reach the standard;
inputting the cleaned real-time monitoring time sequence data, hydrodynamic real-time monitoring time sequence data and expected water quality standard reaching time of various water quality indexes into an edge calculation AI model, and outputting a real-time guiding value of the sewage treatment strain input quantity.
8. An organic biochemical reaction strain release guidance system, which is characterized in that the system comprises:
a data collection module: the system is used for collecting historical water quality monitoring data and cleaning partial data;
a parameter adjusting module: the water quality model establishing and calibrating method is used for establishing a mechanism water quality modeling and calibrating model, and parameters of the mechanism water quality modeling and calibrating model are adjusted through the cleaned historical water quality monitoring data;
an expert database establishing module: the method is used for establishing an expert database based on a mechanism water quality modeling calibration model and simulating corresponding relations among different water quality situations, hydrodynamic situations, sewage treatment strain input quantity and water quality standard reaching time;
a model construction module: the system is used for establishing an AI calibration modeling model through a plurality of AI models, and training the AI calibration modeling model by taking different water quality situations, hydrodynamic situations and corresponding water quality standard reaching time in an expert database as input and corresponding sewage treatment strain input amount as output;
a real-time guidance module: and guiding the sewage treatment strain input amount at the edge end of the sewage treatment site through a trained AI calibration modeling model.
9. An electronic device, comprising: at least one processor, at least one memory, a communication interface, and a bus;
the processor, the memory and the communication interface complete mutual communication through the bus;
the memory stores program instructions executable by the processor, the processor invoking the program instructions to implement the method of any one of claims 1-7.
10. A computer-readable storage medium storing computer instructions for causing a computer to implement the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111443428.XA CN114239387A (en) | 2021-11-30 | 2021-11-30 | Organic biochemical reaction strain putting guidance method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111443428.XA CN114239387A (en) | 2021-11-30 | 2021-11-30 | Organic biochemical reaction strain putting guidance method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114239387A true CN114239387A (en) | 2022-03-25 |
Family
ID=80752158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111443428.XA Pending CN114239387A (en) | 2021-11-30 | 2021-11-30 | Organic biochemical reaction strain putting guidance method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114239387A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117800425A (en) * | 2024-03-01 | 2024-04-02 | 宜宾科全矿泉水有限公司 | Water purifier control method and system based on artificial intelligence |
-
2021
- 2021-11-30 CN CN202111443428.XA patent/CN114239387A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117800425A (en) * | 2024-03-01 | 2024-04-02 | 宜宾科全矿泉水有限公司 | Water purifier control method and system based on artificial intelligence |
| CN117800425B (en) * | 2024-03-01 | 2024-06-07 | 宜宾科全矿泉水有限公司 | Water purifier control method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102854296B (en) | Sewage-disposal soft measurement method on basis of integrated neural network | |
| Hong et al. | Evolutionary self-organising modelling of a municipal wastewater treatment plant | |
| CN109858714B (en) | Tobacco shred quality inspection index prediction method, device and system based on improved neural network | |
| JPH07319509A (en) | Method and system for supporting process operation | |
| CN102262147A (en) | Soft measurement method and system for effluent chemical oxygen demand (COD) of waste water treatment system | |
| CN114297954A (en) | Big data intelligent analysis digital management and control platform based on sewage treatment plant | |
| CN102183621A (en) | Aquaculture dissolved oxygen concentration online forecasting method and system | |
| CN110045771B (en) | Intelligent monitoring system for water quality of fishpond | |
| CN107315775A (en) | A kind of index calculating platform and method | |
| CN107368707A (en) | Gene chip expression data analysis system and method based on US ELM | |
| CN117164103B (en) | Intelligent control method, terminal and system of domestic sewage treatment system | |
| Bagheri et al. | Modeling of effluent quality parameters in a submerged membrane bioreactor with simultaneous upward and downward aeration treating municipal wastewater using hybrid models | |
| CN111693667A (en) | Water quality detection system and method based on gated recursive array | |
| CN115795367A (en) | Algal bloom outbreak prediction method based on machine learning and application | |
| CN114239387A (en) | Organic biochemical reaction strain putting guidance method and system | |
| CN105701280A (en) | Southern America white-leg shrimp pond culture water quality prediction method | |
| CN109408896B (en) | Multi-element intelligent real-time monitoring method for anaerobic sewage treatment gas production | |
| CN114119277A (en) | Artificial intelligent neural network-based flocculation dosing decision analysis method | |
| CN113151842B (en) | Method and device for determining conversion efficiency of wind-solar complementary water electrolysis hydrogen production | |
| CN114861535A (en) | Water treatment process machine learning modeling method using CFD as data preprocessing | |
| CN117763352A (en) | Dosage optimization control method based on VSM-GRU total phosphorus prediction model | |
| CN106769748B (en) | Intelligent detection system for water permeability of membrane bioreactor-MBR (Membrane bioreactor) | |
| CN109241133A (en) | Data monitoring method, calculates equipment and storage medium at device | |
| CN116307383B (en) | Ecological balance-based land fine conservation improvement method and system | |
| CN117555303A (en) | Real-time control method for dosing pump of sewage treatment plant station based on machine learning |
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 |