CN112794550A - Method and system for solving COD standard exceeding of effluent of sewage treatment plant based on artificial intelligence - Google Patents
Method and system for solving COD standard exceeding of effluent of sewage treatment plant based on artificial intelligence Download PDFInfo
- Publication number
- CN112794550A CN112794550A CN202011422024.8A CN202011422024A CN112794550A CN 112794550 A CN112794550 A CN 112794550A CN 202011422024 A CN202011422024 A CN 202011422024A CN 112794550 A CN112794550 A CN 112794550A
- Authority
- CN
- China
- Prior art keywords
- cod
- effluent
- standard
- sludge
- mixed liquid
- 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.)
- Granted
Links
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F9/00—Multistage treatment of water, waste water or sewage
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F2209/00—Controlling or monitoring parameters in water treatment
- C02F2209/08—Chemical Oxygen Demand [COD]; Biological Oxygen Demand [BOD]
-
- 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
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W10/00—Technologies for wastewater treatment
- Y02W10/10—Biological treatment of water, waste water, or sewage
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Hydrology & Water Resources (AREA)
- Engineering & Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Water Supply & Treatment (AREA)
- Chemical & Material Sciences (AREA)
- Organic Chemistry (AREA)
- Activated Sludge Processes (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a method and a system for solving the problem that COD (chemical oxygen demand) of effluent of a sewage treatment plant exceeds standard based on artificial intelligence, wherein a training sample for model training is established according to measured values of the COD of the effluent, the COD of the influent, the influent flow, the reflux ratio of mixed liquid, the reflux ratio of sludge and dissolved oxygen in normal standard operation data; establishing a mathematical model and carrying out model training; obtaining a test sample, selecting test data of the effluent COD in the test sample reaching the standard, calling a trained mathematical model, and judging the fitting property of the mathematical model; selecting test data of the effluent COD in the test sample which does not reach the standard, calling the trained mathematical model, and matching with an artificial intelligent control system to automatically regulate and control the inflow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen so as to enable the effluent COD of the sludge treatment plant to reach the standard, so as to establish a functional relation among various parameters by monitoring various parameters, establish the model, and rapidly recover the effluent COD to reach the standard by judging and adjusting the method by using the artificial intelligent model when the effluent COD exceeds the standard.
Description
Technical Field
The invention belongs to the technical field of sewage treatment, and particularly relates to a method and a system for solving the problem that COD (chemical oxygen demand) of effluent of a sewage treatment plant exceeds standard based on artificial intelligence.
Background
The effluent standard of a sewage treatment plant is 'pollutant discharge standard of urban sewage treatment plant' (GB18918-2002) first-class A, and when the COD of the effluent exceeds the standard, parameters influencing the COD of the effluent need to be controlled in time so as to enable the COD of the effluent to reach the standard.
However, the sewage treatment process has the characteristics of high nonlinearity, time-varying property, uncertainty, strong coupling property and the like, and in the operation of the traditional sewage treatment plant, when the COD of the effluent of the sewage treatment plant exceeds the standard, the COD of the effluent is difficult to recover to normal within a short time by controlling and adjusting a single parameter.
Disclosure of Invention
In view of the above, in order to solve the above problems in the prior art, the present invention aims to provide a method and a system for solving the problem of the exceeding of the COD of the effluent of the sewage treatment plant based on artificial intelligence, so as to achieve the purpose of rapidly recovering the COD of the effluent to reach the standard by monitoring the parameters of the COD of the effluent of the sewage treatment plant, the COD of the effluent, the inflow water, the dissolved oxygen, the reflux ratio of the mixed liquid, the reflux ratio of the sludge, and the like, establishing a functional relationship among the parameters, thereby establishing a model, and intelligently judging and adjusting the method by using the artificial intelligence model when the COD of the effluent.
The technical scheme adopted by the invention is as follows: a method for solving the problem that COD (chemical oxygen demand) of effluent of a sewage treatment plant exceeds standard based on artificial intelligence comprises the following steps:
s1: establishing a training sample of model training according to measured values of water outlet COD, water inlet flow, mixed liquid reflux ratio, sludge reflux ratio and dissolved oxygen in normal standard-reaching operation data;
s2: establishing a mathematical model and carrying out model training;
s3: obtaining a test sample, selecting test data of effluent COD (chemical oxygen demand) in the test sample reaching the standard, calling a trained mathematical model, judging the fitting property of the mathematical model, and entering the next step if the fitting property is good; if the fitness is not good, return to S2;
s4: selecting test data of the effluent COD in the test sample which do not reach the standard, calling the trained mathematical model, and calculating the control quantity of the inflow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen through the mathematical model so as to meet the requirement that the predicted value of the effluent COD reaches the standard.
Further, the method further comprises:
s5: the control quantity of the water inlet flow, the mixed liquid reflux ratio, the sludge reflux ratio and the dissolved oxygen is calculated according to the mathematical model, and the water inlet flow, the mixed liquid reflux ratio, the sludge reflux ratio and the dissolved oxygen are automatically regulated and controlled through the artificial intelligent control system, so that the effluent COD of the sludge treatment plant reaches the standard.
Further, the mathematical model comprises an input layer, a hidden layer and an output layer, wherein the input data of the input layer are measured values of water inflow COD, water inflow, mixed liquid reflux ratio, sludge reflux ratio and dissolved oxygen, and are x respectively1、x2、x3、x4、x5And the input data are training samples: test sample 6: 1; the output data of the output layer is an effluent COD predicted value, and the effluent COD predicted value is yiAnd the measured value of COD of the effluent is y0。
Further, the method for training the mathematical model in step S1 includes:
s201: assigning a weight ω to input data for each training sampleiAggregating inputs intoWherein i is a natural number;
s202: introducing a deviation b, a threshold value theta, and defining f (x) as 1f(x)=0If f (x) is 1, the effluent COD reaches the standard; if f (x) is 0, the COD of the effluent exceeds the standard;
S204: introducing least squares functionsLet the initial deviation and the random value be (ω)0,b0) By an iterative method, calculate a gradient ^ E (ω)0,b0) And defineWherein j is a natural number, and epsilon is a learning rate;
s205: through learning training, the weight omega is adjustediThe deviation b and the threshold θ are optimally chosen such that f (x) is 1 and E is less than the set value, the current mathematical model is the trained model.
Further, in the step S3, half of the data in the test sample is COD standard exceeding data, and the other half of the data is COD standard reaching data.
Further, the mixed liquor reflux ratio is calculated by the ratio of the mixed liquor reflux amount to the water inlet flow, and the sludge reflux ratio is calculated by the ratio of the sludge reflux amount to the water inlet flow.
Further, in step S5, the artificial intelligence control system regulates and controls the water inflow rate through the rotation speed of the water inflow lift pump or the water inflow electric valve; the artificial intelligence control system regulates and controls the reflux ratio of the mixed liquid through the rotating speed of the mixed liquid reflux pump or the electric valve of the mixed liquid reflux; the artificial intelligent control system regulates and controls the sludge reflux ratio through the rotating speed of the sludge reflux pump or the sludge reflux electric valve; the artificial intelligence control system regulates and controls the content of the dissolved oxygen through the rotating speed of the blower or the gas transmission electric valve.
The invention also provides a system for solving the problem that the COD of the effluent of the sewage treatment plant exceeds the standard based on artificial intelligence, which comprises the following steps:
the system comprises a plurality of measuring points, a data acquisition unit and a data processing unit, wherein each measuring point respectively measures the operation data of a sewage treatment plant in real time;
the system comprises a plurality of control points, a plurality of control units and a control unit, wherein each control point respectively regulates and controls the operation data of a sewage treatment plant in real time;
the input data acquisition module is used for acquiring the operation data of each measuring point through the input data module so as to acquire a training sample and a test sample of the sewage treatment plant;
the data analysis module is used for carrying out fitting judgment on the test data of the effluent COD in the test sample which reaches the standard; calculating the test data of the effluent COD in the test sample which does not reach the standard through a data analysis module to obtain the control quantity of each control point;
and the artificial intelligence control module is used for respectively regulating and controlling each control point through the artificial intelligence control module according to each control quantity so as to enable the effluent COD of the sewage treatment plant to reach the standard.
Furthermore, each measuring point is respectively a water inlet pipe, a water outlet pipe, a mixed liquid return pipeline, a sludge return pipeline and an aerobic pool which are positioned in a sludge treatment plant, the water inlet pipeline is provided with a first COD online monitor and a water inlet electromagnetic flowmeter, the water outlet pipeline is provided with a second COD online monitor, the mixed liquid return pipeline is provided with a mixed liquid return electromagnetic flowmeter, the sludge return pipeline is provided with a sludge return electromagnetic flowmeter, and the aerobic pool is provided with an oxygen dissolving instrument;
first COD on-line monitoring appearance, second COD on-line monitoring appearance, water inlet electromagnetic flowmeter, mixed liquid backward flow electromagnetic flowmeter, mud backward flow electromagnetic flowmeter, dissolved oxygen appearance respectively with input data acquisition module communication connection.
Furthermore, each control point is respectively an inlet electric valve and an inlet lifting pump arranged on the water inlet pipeline, a mixed liquid reflux electric valve and a mixed liquid reflux pump arranged on the mixed liquid reflux pipeline, a sludge reflux electric valve and a sludge reflux pump arranged on the sludge reflux pipeline, and a gas transmission electric valve and a blower arranged on the air pipe of the aerobic pool, and the inlet electric valve, the inlet lifting pump, the mixed liquid reflux electric valve, the mixed liquid reflux pump, the sludge reflux electric valve, the sludge reflux pump, the gas transmission electric valve and the blower are respectively in communication connection with the artificial intelligence control module.
The invention has the beneficial effects that:
1. by adopting the method and the system for solving the problem that the COD of the effluent of the sewage treatment plant exceeds the standard based on artificial intelligence, when the COD of the effluent of the sewage treatment plant exceeds the standard, the optimal mathematical model for training is utilized to carry out real-time analysis, the operation data such as the water inflow, the dissolved oxygen, the mixed liquid reflux ratio, the sludge reflux ratio and the like are jointly controlled, the relevance among the data is found, the fastest and stable standard-reaching control quantity of the COD of the effluent is found out by replacing the artificial experience, the operation data is debugged according to the control quantity, the defect that the parameters such as the water inflow, the mixed liquid reflux ratio, the sludge reflux ratio and the dissolved oxygen cannot be dynamically processed is overcome, the optimal processing method for rapidly finding out the COD of the effluent when the COD exceeds the standard in the operation process of the sewage treatment plant is realized, and the.
Drawings
FIG. 1 is a schematic flow chart of a method for solving the COD standard exceeding of effluent water of a sewage treatment plant based on artificial intelligence, which is provided by the invention;
FIG. 2 is a neural network diagram of a mathematical model in the method for solving the COD standard exceeding of the effluent of a sewage treatment plant based on artificial intelligence provided by the invention;
FIG. 3 is a comparison graph between the effluent COD of the compliance data and the prediction result in the test sample by the method for solving the effluent COD superscript of the sewage treatment plant based on artificial intelligence provided by the invention.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar modules or modules having the same or similar functionality throughout. The embodiments described below with reference to the drawings are exemplary only for the purpose of explaining the present application and are not to be construed as limiting the present application. On the contrary, the embodiments of the application include all changes, modifications and equivalents coming within the spirit and terms of the claims appended hereto.
Example 1
In this embodiment, a method for solving the problem that the COD of the effluent of the sewage treatment plant exceeds the standard based on artificial intelligence is specifically provided, and the method aims to establish a model through the COD of the effluent, the COD of the influent, the influent flow, the reflux ratio of mixed liquid, the reflux ratio of sludge and dissolved oxygen, and in the operation process of the sewage treatment plant, when the COD of the effluent exceeds the standard, an optimal treatment method is quickly found and multi-parameter comprehensive control is realized, as shown in fig. 1, the method comprises the following steps:
s1: according to measured values of water outlet COD, water inlet flow, mixed liquid reflux ratio, sludge reflux ratio and dissolved oxygen in normal standard-reaching operation data, establishing a training sample for model training, wherein the training sample is mainly used for training a mathematical model;
wherein, the water inlet COD and the water outlet COD are obtained by respectively arranging a first COD on-line monitor and a second COD on-line monitor on the water inlet pipeline and the water outlet pipeline for measurement; the water inlet flow is obtained by installing a water inlet electromagnetic flowmeter on a water inlet pipeline for measurement; the mixed liquid reflux ratio is calculated by the ratio of the mixed liquid reflux amount to the water inlet flow, the mixed liquid reflux amount is obtained by installing a mixed liquid reflux electromagnetic flowmeter on a mixed reflux pipeline for measurement, and the mixed liquid reflux ratio is equal to the mixed liquid flow/the water inlet flow, and the system automatically calculates the mixed liquid reflux ratio; the sludge reflux ratio is calculated by the ratio of the sludge reflux amount to the water inflow amount, the sludge reflux amount is obtained by measuring a sludge reflux electromagnetic flowmeter arranged on a sludge reflux pipeline, and the system automatically calculates the sludge reflux ratio which is the sludge reflux amount/the water inflow amount; the measured value of the dissolved oxygen is obtained by installing an oxygen dissolving instrument in the aerobic tank for real-time measurement.
S2: establishing a mathematical model and carrying out model training; as shown in fig. 2, the mathematical model includes an input layer, a hidden layer and an output layer, the input data of the input layer are the measured values of the influent COD, the influent flow rate, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen, and are x respectively1、x2、x3、x4、x5And the input data are training samples: test sample 6: 1; the output data of the output layer is an effluent COD predicted value, and the effluent COD predicted value is yiAnd the measured value of COD of the effluent is y0。
According to first COD on-line monitoring appearance, second COD on-line monitoring appearance, dissolved oxygen appearance, mixed liquid backward flow electromagnetic flowmeter, mud backward flow electromagnetic flowmeter, the electromagnetic flowmeter of intaking respectively record normal water COD of going out when reaching standard, the COD of intaking, the flow of intaking, dissolved oxygen, mixed liquid reflux ratio, mud reflux ratio to this training sample is established. In this embodiment, the values of the training samples and the test samples are as follows:
the model training method mainly comprises the following steps:
s201: assigning a weight ω to input data for each training sampleiThe inputs are aggregated as:wherein i is a natural number; substituting into the first set of dataThe rest groups are analogized;
s202: introducing a deviation b, a threshold value theta, and defining f (x) as 1f(x)=0If f (x) is 1, the effluent COD reaches the standard; if f (x) is 0, the COD of the effluent exceeds the standard;
Substituting into the first set of data yi=σ(220ω1+1214ω2+200%ω3+80%ω4+2.22ω5+ b), the rest groups and so on;
s204: introducing least squares functionsSubstituting into the first set of dataThe rest groups are analogized;
let the initial deviation and the random value be (ω)0,b0) By an iterative method, calculate a gradient ^ E (ω)0,b0) And defineWherein j is a natural number, and epsilon is a learning rate; calculate gradient ^ E (ω)0,b0) When, defineAnd repeating the process… … and so on.
S205: through learning training, the weight omega is adjustediThe deviation b and the threshold value theta are optimally selected, so that f (x) is 1 and E is smaller than a set value, the current mathematical model is a trained model, and the predicted value of the effluent COD approaches to the measured value of the effluent COD infinitely at the moment.
S3: obtaining a test sample, wherein half of data in the test sample is required to be COD standard exceeding data, and the other half of data is COD standard reaching data;
selecting test data of effluent COD (chemical oxygen demand) in a test sample, calling a trained mathematical model, judging the fitting property of the mathematical model, and entering the next step if the fitting property is good; if the fitness is not good, the method returns to step S2, and the mathematical model continues to be trained. At this stepIn the step, it is necessary to determine the fitness of the trained mathematical model because the fitness of the model trained in the previous step is not always satisfied under the influence of the multi-parameter, and thus the fitness of the trained mathematical model needs to be determined. In the step, the judgment of the fitness is that after the test data of the effluent COD reaching the standard is selected, a least square function is carried outCalculating the value E, and if the value E is within a set value, indicating that the fitting performance of the model is good; otherwise, the fitting performance of the model cannot meet the requirement. The prediction results of the test data for effluent COD compliance in the test samples are shown in table 2 below:
s4: selecting test data that the effluent COD in the test sample does not reach the standard, namely selecting data that the effluent COD is more than 50mg/L, calling the trained mathematical model, and calculating the control quantity of the water inlet flow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen through the mathematical model, wherein the control quantity can meet the requirement that the effluent COD predicted value of the output data reaches the standard. The adjustment results of the test data for the excess COD of the effluent in the test sample are shown in the following table 3:
s5: the control quantity of the water inlet flow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen is calculated according to the mathematical model, and the water inlet flow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen are automatically regulated and controlled by the artificial intelligent control system, so that the COD of the effluent of the sludge treatment plant reaches the standard, namely the COD of the effluent is less than 50 mg/L.
In the embodiment, in the automatic regulation and control process, the artificial intelligent control system regulates and controls the water inlet flow through the rotating speed of the water inlet lifting pump or the water inlet electric valve; the artificial intelligence control system regulates and controls the reflux ratio of the mixed liquid through the rotating speed of the mixed liquid reflux pump or the electric valve of the mixed liquid reflux; the artificial intelligent control system regulates and controls the sludge reflux ratio through the rotating speed of the sludge reflux pump or the sludge reflux electric valve; the artificial intelligence control system regulates and controls the content of the dissolved oxygen through the rotating speed of the blower or the gas transmission electric valve.
The parameters of the rotating speed of the water inlet lifting pump or the rotating speed of the water inlet electric valve, the rotating speed of the mixed liquid reflux pump or the mixed liquid reflux electric valve, the rotating speed of the sludge reflux pump or the rotating speed of the sludge reflux electric valve, the rotating speed of the air blower or the rotating speed of the air transmission electric valve are controlled in an optimal range, so that the COD of the effluent can be ensured to reach the standard (the COD of the effluent is less than 50mg/L) when a sludge treatment plant operates.
Example 2
The invention also provides a system for solving the problem that COD (chemical oxygen demand) of effluent water of a sewage treatment plant exceeds standard based on artificial intelligence, the system is mainly based on a sludge AAO (anaerobic-anoxic-oxic) process treatment system, the sludge AAO process treatment system belongs to common general knowledge in the field, and details are not repeated here, and the system comprises:
measuring data part
The measurement data part comprises a plurality of measurement points, and each measurement point respectively measures the operation data of the sewage treatment plant in real time; each measuring point is respectively a water inlet pipe, a water outlet pipe, a mixed liquid return pipeline and a sludge return pipeline which are positioned in a sludge treatment plant, the water inlet pipeline is provided with a first COD online monitor and a water inlet electromagnetic flowmeter, the water outlet pipeline is provided with a second COD online monitor, the mixed liquid return pipeline is provided with a mixed liquid return electromagnetic flowmeter, the sludge return pipeline is provided with a sludge return electromagnetic flowmeter, and the aerobic tank is provided with an oxygen dissolving instrument; and the first COD online monitor, the second COD online monitor, the water inlet electromagnetic flow meter, the mixed liquid reflux electromagnetic flow meter, the sludge reflux electromagnetic flow meter and the dissolved oxygen meter are respectively in communication connection with the input data acquisition module to realize real-time transmission and feedback of the measured data.
② regulation control part
The regulation and control part comprises a plurality of control points, and each control point respectively regulates and controls the operation data of the sewage treatment plant in real time; the control points are respectively as follows: the system comprises a water inlet electric valve and a water inlet lifting pump which are arranged on a water inlet pipeline, a mixed liquid reflux electric valve and a mixed liquid reflux pump which are arranged on a mixed liquid reflux pipeline, a sludge reflux electric valve and a sludge reflux pump which are arranged on the sludge reflux pipeline, and a gas transmission electric valve and a blower which are arranged on an air pipe of an aerobic pool, wherein the water inlet electric valve, the water inlet lifting pump, the mixed liquid reflux electric valve, the mixed liquid reflux pump, the sludge reflux electric valve, the sludge reflux pump, the gas transmission electric valve and the blower are respectively in communication connection with an artificial intelligent control module; in example 1, the water inlet electric valve, the mixed liquid reflux electric valve, the sludge reflux electric valve and the gas transmission electric valve are controlled to respectively control the water inlet flow rate, the mixed liquid reflux flow rate, the sludge reflux flow rate and the aeration amount to the aerobic tank. Certainly, in practical application, the water inlet flow, the mixed liquid reflux flow, the sludge reflux flow and the aeration amount to the aerobic tank can be controlled by regulating the rotating speed of the water inlet lifting pump, the rotating speed of the mixed liquid reflux pump, the rotating speed of the sludge reflux pump and the rotating speed of the air blower respectively, the regulating effect can be achieved, and the water inlet lifting pump, the mixed liquid reflux pump, the sludge reflux pump and the air blower are respectively in communication connection with the artificial intelligent control module.
Input data acquisition module
The method comprises the steps that operation data of each measuring point are obtained through an input data obtaining module, so that training samples and testing samples of a sewage treatment plant are obtained; specifically, the training samples and the test samples are detailed in table 1 of example 1.
Data analysis module
When the data analysis module runs, the mathematical model in embodiment 1 is called, and the training process and method of the mathematical model are known by referring to embodiment 1, and are not described herein again. Fitting and judging the test data of the effluent COD in the test sample reaching the standard through a data analysis module, and if the fitting is good, continuing to execute subsequent logic; if the fitting is not good, retraining the mathematical model;
the control quantity of each control point is obtained by calculating the test data of the effluent COD in the test sample which does not reach the standard through the data analysis module, and the effluent COD predicted value output by the data analysis module can be ensured to reach the standard within the range of the control quantity.
Fifth artificial intelligence control module
Regulating and controlling each control point through an artificial intelligence control module according to each control quantity, issuing a control instruction by an artificial intelligence control system, and regulating and controlling the water inlet flow through a water inlet electric valve; regulating and controlling the reflux ratio of the mixed liquid through the rotating speed of the mixed liquid reflux pump or the electric valve of the mixed liquid reflux; the sludge reflux ratio is regulated and controlled through the rotating speed of the sludge reflux pump or the sludge reflux electric valve; the content of dissolved oxygen is regulated and controlled through the rotating speed of the air blower or the electric valve for gas transmission, and finally, the COD of the effluent of the sewage treatment plant reaches the standard.
It should be noted that, in the description of the present application, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present application, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present application includes other implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (10)
1. A method for solving the problem that COD (chemical oxygen demand) of effluent of a sewage treatment plant exceeds standard based on artificial intelligence is characterized by comprising the following steps:
s1: establishing a training sample of model training according to measured values of water outlet COD, water inlet flow, mixed liquid reflux ratio, sludge reflux ratio and dissolved oxygen in normal standard-reaching operation data;
s2: establishing a mathematical model and carrying out model training;
s3: obtaining a test sample, selecting test data of effluent COD (chemical oxygen demand) in the test sample reaching the standard, calling a trained mathematical model, judging the fitting property of the mathematical model, and entering the next step if the fitting property is good; if the fitness is not good, return to S2;
s4: selecting test data of the effluent COD in the test sample which do not reach the standard, calling the trained mathematical model, and calculating the control quantity of the inflow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen through the mathematical model so as to meet the requirement that the predicted value of the effluent COD reaches the standard.
2. The method for solving the COD standard exceeding of the effluent water of the sewage treatment plant based on the artificial intelligence according to claim 1, is characterized by further comprising the following steps:
s5: the control quantity of the water inlet flow, the mixed liquid reflux ratio, the sludge reflux ratio and the dissolved oxygen is calculated according to the mathematical model, and the water inlet flow, the mixed liquid reflux ratio, the sludge reflux ratio and the dissolved oxygen are automatically regulated and controlled through the artificial intelligent control system, so that the effluent COD of the sludge treatment plant reaches the standard.
3. The method for solving the COD standard exceeding of the effluent of the sewage treatment plant based on the artificial intelligence as claimed in claim 1, wherein the mathematical model comprises an input layer, a hidden layer and an output layer, the input data of the input layer are the measured values of the COD of the influent water, the influent water flow, the reflux ratio of the mixed liquid, the reflux ratio of the sludge and the dissolved oxygen, and are x1、x2、x3、x4、x5And the input data are training samples: test sample 6: 1; the outputThe output data of the layer is an effluent COD predicted value, and the effluent COD predicted value is set to be yiAnd the measured value of COD of the effluent is y0。
4. The method for solving the COD exceeding of the effluent of the sewage treatment plant based on the artificial intelligence as claimed in claim 3, wherein the training method for the mathematical model in the step S1 is as follows:
s201: assigning a weight ω to input data for each training sampleiAggregating inputs intoWherein i is a natural number;
s202: introducing a deviation b, a threshold value theta, defining If f (x) is 1, the effluent COD reaches the standard; if f (x) is 0, the COD of the effluent exceeds the standard;
S204: introducing least squares functionsLet the initial deviation and the random value be (ω)0,b0) By an iterative method, calculate a gradient ^ E (ω)0,b0) And defineWherein j is a natural number, and epsilon is a learning rate;
s205: through learning training, the weight omega is adjustediThe deviation b and the threshold θ are optimally chosen such that f (x) is 1 and E is less than the set value, the current mathematical model is the trained model.
5. The method for solving the COD overproof problem of the effluent of the sewage treatment plant based on the artificial intelligence of claim 1, wherein in the step S3, half of the data in the test sample are the COD overproof data, and the other half of the data are the COD standard data.
6. The method for solving the COD standard exceeding of the effluent of the sewage treatment plant based on the artificial intelligence as claimed in claim 1, wherein the mixed liquor reflux ratio is calculated by the ratio of the mixed liquor reflux amount to the influent flow rate, and the sludge reflux ratio is calculated by the ratio of the sludge reflux amount to the influent flow rate.
7. The method for solving the COD exceeding the standard of the effluent water of the sewage treatment plant based on the artificial intelligence according to claim 6, wherein in the step S5, the artificial intelligence control system regulates and controls the water inlet flow through the rotating speed of a water inlet lifting pump or a water inlet electric valve; the artificial intelligence control system regulates and controls the reflux ratio of the mixed liquid through the rotating speed of the mixed liquid reflux pump or the electric valve of the mixed liquid reflux; the artificial intelligent control system regulates and controls the sludge reflux ratio through the rotating speed of the sludge reflux pump or the sludge reflux electric valve; the artificial intelligence control system regulates and controls the content of the dissolved oxygen through the rotating speed of the blower or the gas transmission electric valve.
8. The utility model provides a system for solve sewage treatment plant effluent COD and exceed standard based on artificial intelligence which characterized in that, this system includes:
the system comprises a plurality of measuring points, a data acquisition unit and a data processing unit, wherein each measuring point respectively measures the operation data of a sewage treatment plant in real time;
the system comprises a plurality of control points, a plurality of control units and a control unit, wherein each control point respectively regulates and controls the operation data of a sewage treatment plant in real time;
the input data acquisition module is used for acquiring the operation data of each measuring point through the input data module so as to acquire a training sample and a test sample of the sewage treatment plant;
the data analysis module is used for carrying out fitting judgment on the test data of the effluent COD in the test sample which reaches the standard; calculating the test data of the effluent COD in the test sample which does not reach the standard through a data analysis module to obtain the control quantity of each control point;
and the artificial intelligence control module is used for respectively regulating and controlling each control point through the artificial intelligence control module according to each control quantity so as to enable the effluent COD of the sewage treatment plant to reach the standard.
9. The system for solving the COD exceeding standard of the effluent of the sewage treatment plant based on the artificial intelligence is characterized in that each measuring point is respectively positioned on an inlet pipe, an outlet pipe, a mixed liquid return pipeline, a sludge return pipeline and an aerobic tank of the sludge treatment plant, the inlet pipe is provided with a first COD online monitor and an inlet electromagnetic flowmeter, the outlet pipe is provided with a second COD online monitor, the mixed liquid return pipeline is provided with a mixed liquid return electromagnetic flowmeter, the sludge return pipeline is provided with a sludge return electromagnetic flowmeter, and the aerobic tank is provided with a dissolved oxygen meter;
first COD on-line monitoring appearance, second COD on-line monitoring appearance, water inlet electromagnetic flowmeter, mixed liquid backward flow electromagnetic flowmeter, mud backward flow electromagnetic flowmeter, dissolved oxygen appearance respectively with input data acquisition module communication connection.
10. The system for solving the COD exceeding the standard of the effluent of the sewage treatment plant based on the artificial intelligence as claimed in claim 8, wherein each control point comprises an electric water inlet valve and an electric water inlet lift pump arranged on the water inlet pipeline, an electric mixed liquid return valve and an electric mixed liquid return pump arranged on the mixed liquid return pipeline, an electric sludge return valve and an electric sludge return pump arranged on the sludge return pipeline, and an electric gas transmission valve and an air blower arranged on the air pipe of the aerobic pool, and the electric water inlet valve, the electric water inlet lift pump, the electric mixed liquid return valve, the electric mixed liquid return pump, the electric sludge return valve, the electric sludge return pump, the electric gas transmission valve and the air blower are respectively in communication connection with the artificial intelligence control.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011422024.8A CN112794550B (en) | 2020-12-08 | 2020-12-08 | Method and system for solving COD standard exceeding of effluent of sewage treatment plant based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011422024.8A CN112794550B (en) | 2020-12-08 | 2020-12-08 | Method and system for solving COD standard exceeding of effluent of sewage treatment plant based on artificial intelligence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112794550A true CN112794550A (en) | 2021-05-14 |
CN112794550B CN112794550B (en) | 2022-11-18 |
Family
ID=75806449
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011422024.8A Active CN112794550B (en) | 2020-12-08 | 2020-12-08 | Method and system for solving COD standard exceeding of effluent of sewage treatment plant based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112794550B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114671523A (en) * | 2021-12-31 | 2022-06-28 | 广东长天思源环保科技股份有限公司 | Sewage treatment energy-saving method based on predictive control |
CN115403222A (en) * | 2022-09-16 | 2022-11-29 | 广东海洋大学 | Breeding tail water treatment system and method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101334395A (en) * | 2008-08-07 | 2008-12-31 | 北京工业大学 | COD soft sensing process |
CN103472728A (en) * | 2013-09-22 | 2013-12-25 | 浙江大学 | Machine learning system and method for controlling pesticide incinerator injurant discharge to meet standard |
CN103728431A (en) * | 2014-01-09 | 2014-04-16 | 重庆科技学院 | Industrial sewage COD (chemical oxygen demand) online soft measurement method based on ELM (extreme learning machine) |
CN103809557A (en) * | 2013-12-30 | 2014-05-21 | 北京工业大学 | Neural network based sewage disposal process optimal control method |
CN105976028A (en) * | 2016-05-11 | 2016-09-28 | 深圳市开天源自动化工程有限公司 | Method for predicting effluent COD concentration in A2O sewage treatment process |
-
2020
- 2020-12-08 CN CN202011422024.8A patent/CN112794550B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101334395A (en) * | 2008-08-07 | 2008-12-31 | 北京工业大学 | COD soft sensing process |
CN103472728A (en) * | 2013-09-22 | 2013-12-25 | 浙江大学 | Machine learning system and method for controlling pesticide incinerator injurant discharge to meet standard |
CN103809557A (en) * | 2013-12-30 | 2014-05-21 | 北京工业大学 | Neural network based sewage disposal process optimal control method |
CN103728431A (en) * | 2014-01-09 | 2014-04-16 | 重庆科技学院 | Industrial sewage COD (chemical oxygen demand) online soft measurement method based on ELM (extreme learning machine) |
CN105976028A (en) * | 2016-05-11 | 2016-09-28 | 深圳市开天源自动化工程有限公司 | Method for predicting effluent COD concentration in A2O sewage treatment process |
Non-Patent Citations (2)
Title |
---|
王晶 等著: "《水源地水环境特征及水质预测模型与算法研究》", 31 May 2020 * |
葛东旭 编著: "《数据挖掘原理与应用》", 31 March 2020 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114671523A (en) * | 2021-12-31 | 2022-06-28 | 广东长天思源环保科技股份有限公司 | Sewage treatment energy-saving method based on predictive control |
CN114671523B (en) * | 2021-12-31 | 2024-06-04 | 广东长天思源环保科技股份有限公司 | Sewage treatment energy-saving method based on predictive control |
CN115403222A (en) * | 2022-09-16 | 2022-11-29 | 广东海洋大学 | Breeding tail water treatment system and method |
CN115403222B (en) * | 2022-09-16 | 2023-11-14 | 广东海洋大学 | Cultivation tail water treatment system and method |
Also Published As
Publication number | Publication date |
---|---|
CN112794550B (en) | 2022-11-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
AU2021101438A4 (en) | Adaptive control method and system for aeration process | |
CN106873379B (en) | Sewage treatment optimal control method based on iterative ADP algorithm | |
CN104360035B (en) | A kind of sewage total phosphorus TP flexible measurement method based on self-organization population-radial base neural net | |
CN112794550B (en) | Method and system for solving COD standard exceeding of effluent of sewage treatment plant based on artificial intelligence | |
CN108469507B (en) | Effluent BOD soft measurement method based on self-organizing RBF neural network | |
CN102122134A (en) | Method and system for wastewater treatment of dissolved oxygen control based on fuzzy neural network | |
CN110889085A (en) | Intelligent wastewater monitoring method and system based on complex network multiple online regression | |
CN108640276A (en) | A kind of sewage treatment plant AAO process optimization operation methods based on WEST models | |
CN104090488B (en) | The method that sewage plant controls dissolved oxygen, sludge loading and sludge age in real time automatically | |
CN104238527A (en) | Precise control method of sewage treatment plant aeration total quantity | |
CN103197539A (en) | Wastewater disposal intelligent optimization control aeration quantity method | |
CN114230110B (en) | Short-range intelligent phosphorus removal and medicine addition control method, equipment and system for sewage treatment | |
CN102411308A (en) | Adaptive control method of dissolved oxygen (DO) based on recurrent neural network (RNN) model | |
CN107986428A (en) | A kind of sewage disposal accurate aeration method | |
CN112782975A (en) | Sewage treatment aeration intelligent control method and system based on deep learning | |
CN109657790A (en) | A kind of Recurrent RBF Neural Networks water outlet BOD prediction technique based on PSO | |
CN111943292A (en) | Storage tank coupling high-load sewage treatment method, device and system | |
CN114671523B (en) | Sewage treatment energy-saving method based on predictive control | |
CN115793471A (en) | Adjustable control method and system based on sewage treatment monitoring | |
CN114956360B (en) | Sewage treatment aeration oxygenation control method based on hill climbing algorithm | |
CN114488973B (en) | Method and device for optimizing whole-flow technological parameters of regenerated water plant and electronic equipment | |
CN104345636A (en) | Dissolved-oxygen control method based on improved differential algorithm | |
CN116969616A (en) | Soft measurement-based integrated sewage treatment facility control method, device and equipment | |
CN113111576A (en) | Mixed coding particle swarm-long and short term memory neural network based soft measurement method for ammonia nitrogen in effluent | |
CN108563118A (en) | A kind of dissolved oxygen model predictive control method based on Adaptive Fuzzy Neural-network |
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 | ||
CB02 | Change of applicant information |
Address after: 610000 Shiyang Industrial Park, hi tech Zone, Chengdu, Sichuan Applicant after: Hankel Water Group Co.,Ltd. Address before: 610000 Shiyang Industrial Park, hi tech Zone, Chengdu, Sichuan Applicant before: SICHUAN HANKE ENVIRONMENTAL PROTECTION EQUIPMENT Co.,Ltd. |
|
CB02 | Change of applicant information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |