CN109534486A - A kind of ship sewage treatment control forecasting system and prediction technique based on Stochastic Recursive wavelet neural network - Google Patents
A kind of ship sewage treatment control forecasting system and prediction technique based on Stochastic Recursive wavelet neural network Download PDFInfo
- Publication number
- CN109534486A CN109534486A CN201811573025.5A CN201811573025A CN109534486A CN 109534486 A CN109534486 A CN 109534486A CN 201811573025 A CN201811573025 A CN 201811573025A CN 109534486 A CN109534486 A CN 109534486A
- Authority
- CN
- China
- Prior art keywords
- neural network
- value
- stochastic
- sewage treatment
- srwnn
- 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
Classifications
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F3/00—Biological treatment of water, waste water, or sewage
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F2101/00—Nature of the contaminant
- C02F2101/10—Inorganic compounds
- C02F2101/16—Nitrogen compounds, e.g. ammonia
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F2103/00—Nature of the water, waste water, sewage or sludge to be treated
- C02F2103/008—Originating from marine vessels, ships and boats, e.g. bilge water or ballast water
-
- 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]
-
- 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
- C02F2209/14—NH3-N
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Microbiology (AREA)
- Hydrology & Water Resources (AREA)
- Environmental & Geological Engineering (AREA)
- Water Supply & Treatment (AREA)
- Chemical & Material Sciences (AREA)
- Organic Chemistry (AREA)
- Feedback Control In General (AREA)
- Activated Sludge Processes (AREA)
Abstract
The purpose of the present invention is to provide a kind of ship sewage treatment control forecasting systems and prediction technique based on Stochastic Recursive wavelet neural network, principle of wavelet analysis and recurrent neural network are combined, so that Stochastic Recursive wavelet neural network is inherited the historical trace characteristic of the excellent localization property of wavelet transformation and recurrent neural network completely, realizes powerful Function Fitting ability.The features such as particular for the nonlinearity of ship sewage treatment process, strong coupling, time-varying, large time delay and complexity, solves the problems such as low precision of traditional neural network prediction model, stability is low.Corresponding control strategy is proposed, while carrying out self-monitoring and diagnosis to ship sewage treatment equipment, also achieves self-regeneration, intelligence degree is high, further saves operating cost.This prediction technique can be good at predicting the contaminant removal efficiency in vessel sewage, and then provide practicable operation reserve to the processing of low temperature vessel sewage.
Description
Technical field
The present invention relates to a kind of Sewage from Ships processing forecasting system and its prediction techniques.
Background technique
International Maritime Organization (IMO) is increasingly stringenter the emission request of vessel sewage.Although sanitary sewage disposal technology
It is swift and violent in China's development with device, but there are many problems.Control problem especially during processing, due to sewage
The features such as there are nonlinearity, strong coupling, time-varying, large time delay and complexity in treatment process, traditional control method,
Such as switch control or PID control, single variable can only be controlled, and for complicated system, cannot achieve stabilization has
The control of effect.
For defect existing for traditional control method, in recent years, many scholars carry out control in sewage disposal process
Largely study and propose many mathematical models, such as ASM series model, fuzzy control model etc..But at present at sewage
Manage that process model structure is complicated, and undetermined parameter is excessive, can recognize difference, cannot dynamically reflect performance variable and control target it
Between connection, so being not used to On-line Control.Further, since sewage disposal process is had strong by influent quality, temperature and pH
The features such as coupling and nonlinearity, even more must be monitored and controlled sewage disposal process and propose challenge.Therefore, seek one
The effective sewage disposal process control method of kind is especially urgent.
The present invention devises the ship sewage treatment control forecasting system based on random regression wavelet neural network (SRWNN)
System is combined by building wavelet algorithm and recurrent neural network, using the method time of model prediction in sewage disposal process
Water pollutant (COD and total nitrogen) carries out real-time monitoring out, and ship sewage treatment system treatment effeciency is low when particular for low temperature
Problem proposes corresponding control strategy.The system is applied to the monitoring and control of sewage disposal process, improves the steady of system
Qualitative and reliability, while having ensured effluent quality and having reduced energy consumption.In addition, the system greatly reduces human factor pair
The interference and operator's bring operating cost of control process.
Summary of the invention
The purpose of the present invention is to provide can greatly reduce human factor to bring the interference of control process and operator
Operating cost a kind of ship sewage treatment control forecasting system and prediction technique based on Stochastic Recursive wavelet neural network.
The object of the present invention is achieved like this:
A kind of ship sewage treatment control forecasting system based on Stochastic Recursive wavelet neural network of the present invention, feature
Be: including buffer pool, biochemical treatment tank, water inlet pipe is connected to buffer pool, and buffer pool is connected to biochemical treatment tank by intermediate tube, biochemical
Processing pond outlet outlet pipe installs inlet regulating valve on water inlet pipe, flow sensor, biochemical treatment tank is installed on intermediate tube
Heating device is arranged in bottom, and aeration pump is connected below biochemical treatment tank, and temperature sensor, online is respectively set in biochemical treatment tank
COD sensor, online total nitrogen sensor.
A kind of ship sewage treatment control forecasting method based on Stochastic Recursive wavelet neural network of the present invention, feature
Be: prediction model based on wavelet neural network is divided into three layers: input layer, hidden layer and output layer；Prediction model SRWNN neural network is defeated
Enter for x (t)=[x1(t),x2(t),x3(t),x4(t)]T, x1It (t) is t moment biochemical treatment area temperature value, x2It (t) is t moment
Biochemical treatment area flow value, x3It (t) is t moment biochemical treatment area COD value, x4It (t) is t moment biochemical treatment area ammonia nitrogen value；
Model foundation is as follows:
(1) it initializes Stochastic Recursive Wavelet Neural Control device: determining that neural network is the connection type of 4-N-N-2,
I.e. input layer be 4, hidden layer, recurrence layer neuron be N, output layer neuron 2, N=2n+1, n is input layer
The number of neuron, learning rate from 0.02 to 0.2 between, initial connection weight wij、vj、cjIt is complied with standard on [- 1,1] uniformly
Distribution；Initial scale factor ajWith shift factor bjComply with standard normal distribution；Error ξ=10-5With maximum training iteration K=
100 objective function designed for ship sewage treatment process control prediction:
Wherein, cjIt is weight of the hidden layer to output layer, ψ (x) is the mother wavelet function of WNN model, small using Morlet
Wave function.The definition of Morlet wavelet function is by expression formula ψ (x)=cos (1.75x) exp {-x2/ 2 } it describes, ψa,bIt (x) is pair
The substrate wavelet function answered, as the activation primitive in hidden layer；
(2) SRWNN model performance index, loss function are defined are as follows:
Wherein E is the loss function of result test, and α is time intensive parameter, tnIt is training data any time value, tlIt is
The latest value of training data, γ are the preset values according to training data,It is the corresponding average value of training sample, output valve error
Are as follows:
Using gradient descent method loss function is minimized, until loss function is less than default error threshold ξ=10-5；
(3) it trains SRWNN model and carries out parameter update:
The gradient of input layer connection weight are as follows:
Wherein η is learning rate, Schistosomiasis control rate；For the derivative of SRWNN model activation primitive；
Correspondingly, the gradient of recurrence layer connection weight are as follows:
The connection weight gradient of hidden layer are as follows:
The two indices a of wavelet functionj、bjGradient be respectively as follows:
The parameter of SRWNN model more new formula is respectively as follows:
(4) the objective function error size for judging current time water outlet COD and ammonia nitrogen repeats step (3) as E > ξ；Such as
When fruit E < ξ, then the output y that step (1) calculates SRWNN controller is gone to1(t),y2(t)。
A kind of ship sewage treatment control forecasting method based on Stochastic Recursive wavelet neural network of the present invention can also wrap
It includes:
1, when detecting current time temperature value x1 (t) less than 15 DEG C of setting value, SRWNN controller output order is opened
Heating device；Until temperature value x1 (t) is increased to 30 DEG C, heating device is closed.
2, when water outlet COD predicted value y1 (t) is greater than the set value 125mg/L, water outlet TN predicted value y2 (t) is greater than setting
When 20mg/L, flow of inlet water regulating valve, reduction flow of inlet water to original 1/2 are opened, and open the power governor of aeration pump,
Increase air inflow, and then increases dissolved oxygen concentration.
Present invention has an advantage that
1, using Stochastic Recursive wavelet neural network predictive control algorithm, Stochastic Recursive wavelet neural network is constructed
(SRWNN) ship sewage treatment control forecasting system.It is combined by wavelet transformation with recurrent neural network, makes wavelet neural
Network inherits the excellent Time-Frequency Localization characteristic of wavelet transformation and the historical data memory characteristic of recurrent neural network completely, real
Strong nonlinearity is showed to approach, and then has overcome in sewage disposal process that there are nonlinearities, strong coupling, time-varying, large time delay
The influence of property and complexity to model.Compare the algorithm being currently employed in sewage treatment, such as FUZZY ALGORITHMS FOR CONTROL, BP nerve
The advantages that network and self-organizing radial base neural net scheduling algorithm have structure simple, fast convergence rate, precision is higher.
2, it among sewage disposal process control, is typically just controlled for indexs such as dissolved oxygen, pH, but for ship
Oceangoing ship sanitary sewage, temperature are essential reference factors.When ship oceangoing voyage in winter, the water temperature of sewage is often less than micro-
The temperature that biology can survive, at this time can damage biological treatment system.And it is based on the ship of wavelet neural network (WNN)
Sewage treatment control forecasting system compares other control technologies, joined the control of temperature.
3, most of algorithms applied in sewage treatment only considered single factors, such as dissolved oxygen concentration, pH at present
Value etc., can only play booster action to normal operation, can not provide control to the normal operation of entire sewage disposal device；It should
The ship sewage treatment control forecasting system of Stochastic Recursive wavelet neural network (SRWNN) is based on Stochastic Recursive Wavelet Neural Network
The predicted value of network algorithm proposes corresponding control strategy, compares other algorithms and control system on the market, our system
While playing self-monitoring and diagnosis to ship sewage treatment equipment, it is also proposed that control strategy solves ship under cryogenic conditions
The low problem of domestic sewage processing system treatment effeciency, and then self-regeneration is realized, intelligence degree is high, further saves
Operating cost.
Detailed description of the invention
Fig. 1 is Stochastic Recursive wavelet neural network structure chart；
Fig. 2 is ship sewage treatment system body figure；
Fig. 3 is process control schematic diagram；
Fig. 4 is water outlet COD prediction of result value；
Fig. 5 is water outlet COD resultant error value；
Fig. 6 is water outlet ammonia nitrogen prediction of result value；
Fig. 7 is water outlet ammonia nitrogen resultant error value.
Specific embodiment
It illustrates with reference to the accompanying drawing and the present invention is described in more detail:
In conjunction with Fig. 1-7, a kind of ship sewage treatment based on Stochastic Recursive wavelet neural network (SRWNN) of the present invention is controlled
Forecasting system, for the operation of Sewage from Ships processing system to be controlled and is predicted, process control schematic diagram such as Fig. 1,
The system includes 3 modules, respectively ship sewage treatment device ontology, sensor acquisition module and Stochastic Recursive wavelet neural
Network controller；
Ship sewage treatment device ontology diagram such as Fig. 2, total arrangement are divided into buffer pool and biochemical treatment tank, buffer pool water inlet
Mouth is connected with inlet regulating valve, for controlling amount of inlet water；Buffer area is connected with biochemical treatment area, is used for buffer system wastewater influent
Amount；Biochemical treatment area is connected with dispensing area by chemicals feed pump, for the harmful substance (COD and total nitrogen etc.) in vessel sewage of degrading
In addition, warming module built in biochemical treatment area, the too low destruction biosystem of temperature in order to prevent；
Sensor acquisition module is built in ship sewage treatment device biochemical treatment tank, includes temperature sensor, flow
Sensor, online COD sensor and online total nitrogen sensor, for Sewage from Ships processing system temperature, flow,
COD and total nitrogen index carry out real-time monitoring, and then obtain sensor signal and be passed to Stochastic Recursive Wavelet Neural Control device；This
Outside, heating device built in the device noumenon of system can acquire system real-time temperature values back according to temperature sensor, constantly
Power is adjusted, is run so that the ship sewage treatment system is under optimum temperature state.
Stochastic Recursive Wavelet Neural Control device is connected directly with sensor module, for by treated, sensor to be believed
Number to Sewage from Ships processing system carry out control and water quality prediction；
Built-in Stochastic Recursive wavelet neural network (SRWNN) prediction technique in Stochastic Recursive Wavelet Neural Control device,
The ship sewage treatment control forecasting system based on Stochastic Recursive wavelet neural network of design, wavelet neural network structure chart is as schemed
3, prediction model is divided into three layers: input layer, hidden layer and output layer；The input of prediction model SRWNN neural network is x (t)=[x1
(t), x2 (t), x3 (t), x4 (t)] T, x1 (t) is t moment biochemical treatment area temperature value, and x2 (t) is t moment biochemical treatment area stream
Magnitude, x3 (t) are t moment biochemical treatment area COD value, and x4 (t) is t moment biochemical treatment area ammonia nitrogen value；Model foundation is as follows:
(1) it initializes Stochastic Recursive Wavelet Neural Control device: determining that neural network is the connection type of 4-N-N-2,
I.e. input layer be 4, hidden layer, recurrence layer neuron be N, output layer neuron 2.The approximation of N is determined as N=
2n+1, n are the number of input layer.Learning rate from 0.02 to 0.2 between.Initial connection weight wij, vj, cj [- 1,
1] it complies with standard and is uniformly distributed on；Initial scale factor aj and shift factor bj complies with standard normal distribution；Error ξ=10-5
The objective function of ship sewage treatment process control prediction is designed for maximum training iteration K=100:
Wherein, cj is weight of the hidden layer to output layer, and ψ (x) is the mother wavelet function of WNN model, is used here
Morlet wavelet function.The definition of Morlet wavelet function is described by expression formula ψ (x)=cos (1.75x) exp {-x2/2 }.ψ
A, b (x) are corresponding substrate wavelet functions, as the activation primitive in hidden layer.
(2) SRWNN model performance index, loss function are defined are as follows:
Wherein E is the loss function of result test, and α (> 0) is time intensive parameter, and tn is training data any time value,
Tl is the latest value of training data, and γ is the preset value according to training data,It is the corresponding average value of training sample.Output valve
Error are as follows:
Using gradient descent method loss function is minimized, until loss function is less than default error threshold ξ=10-
5。
(3) it trains SRWNN model and carries out parameter update:
The gradient of input layer connection weight are as follows:
Wherein η is learning rate, Schistosomiasis control rate.For
The derivative of SRWNN model activation primitive.
Correspondingly, the gradient of recurrence layer connection weight are as follows:
The connection weight gradient of hidden layer are as follows:
The gradient of the two indices aj, bj of wavelet function are respectively as follows:
Therefore, the parameter of SRWNN model more new formula is respectively as follows:
(4) the objective function error size for judging current time water outlet COD and ammonia nitrogen repeats step (3) as E > ξ；Such as
When fruit E < ξ, then the output y1 (t) that step (1) calculates SRWNN controller is gone to, y2 (t) Fig. 4 is the water outlet COD result of system
Predicted value and true value comparison diagram, X-axis are the time, and unit is day, and Y axis is the predicted value y1 (t) and actual comparison of system,
Unit is mg/L；Fig. 5 is the water outlet COD resultant error value E1 of system, and X-axis is the time, and unit is day, and Y-axis is the prediction of system
Error；Fig. 6 is the water outlet ammonia nitrogen prediction of result value and true value comparison diagram of system, and X-axis is the time, and unit is day, and Y-axis is system
Predicted value y2 (t) and actual comparison, unit mg/L；When Fig. 7 is that water outlet COD resultant error value E2, the X axis of system is
Between, unit is day, and Y-axis is the prediction error of system, which demonstrates the effective of the Stochastic Recursive wavelet neural network algorithm
Property.
A kind of optimal control policy can be proposed by wavelet-neural network model, and it is timely that fast and easy finds out faulty equipment
It is excluded, realizes equipment self-regeneration, specific control strategy is as follows:
When detecting current time temperature value x1 (t) less than 15 DEG C of setting value, SRWNN controller output order, unlatching is set
Standby heating device, improves the bulk temperature of reactor；Until temperature value x1 (t) is increased to 30 DEG C, pass hull closure heating device is dirty
Water treatment facilities operate normally.
When the water outlet COD predicted value y1 (t) of continuous a period of time is greater than the set value 125mg/L, water outlet TN predicted value y2 (t)
When greater than setting 20mg/L, flow of inlet water regulating valve, reduction flow of inlet water to original 1/2 are opened, and open the function of aeration pump
Rate adjuster increases the air inflow of equipment, and then increases the dissolved oxygen concentration of equipment, enhances the removal effect of pollutant.
When water outlet COD predicted value y1 (t) is less than setting value 125mg/L, water outlet TN predicted value y2 (t) is less than setting value
When 20mg/L, sewage disposal device is operated normally.
Claims (4)
1. a kind of ship sewage treatment control forecasting system based on Stochastic Recursive wavelet neural network, it is characterized in that: including slow
Pond, biochemical treatment tank are rushed, water inlet pipe is connected to buffer pool, and buffer pool is connected to biochemical treatment tank by intermediate tube, and biochemical treatment tank goes out
Mouth is connected to outlet pipe, installs inlet regulating valve on water inlet pipe, and flow sensor, the setting of biochemical treatment bottom of pond portion are installed on intermediate tube
Heating device, biochemical treatment tank lower section connect aeration pump, and temperature sensor, online COD sensing are respectively set in biochemical treatment tank
Device, online total nitrogen sensor.
2. a kind of ship sewage treatment control forecasting method based on Stochastic Recursive wavelet neural network, it is characterized in that: small echo is refreshing
It is divided into three layers through Network Prediction Model: input layer, hidden layer and output layer；The input of prediction model SRWNN neural network is x (t)
=[x1(t),x2(t),x3(t),x4(t)]T, x1It (t) is t moment biochemical treatment area temperature value, x2It (t) is t moment biochemical treatment
Area's flow value, x3It (t) is t moment biochemical treatment area COD value, x4It (t) is t moment biochemical treatment area ammonia nitrogen value；
Model foundation is as follows:
(1) it initializes Stochastic Recursive Wavelet Neural Control device: determining that neural network is the connection type of 4-N-N-2, i.e., it is defeated
Enter layer neuron be 4, hidden layer, recurrence layer neuron be N, output layer neuron 2, N=2n+1, n be input layer nerve
Member number, learning rate from 0.02 to 0.2 between, initial connection weight wij、vj、cjIt complies with standard on [- 1,1] and uniformly divides
Cloth；Initial scale factor ajWith shift factor bjComply with standard normal distribution；Error ξ=10-5With maximum training iteration K=
100 objective function designed for ship sewage treatment process control prediction:
Wherein, cjIt is weight of the hidden layer to output layer, ψ (x) is the mother wavelet function of WNN model, using Morlet small echo letter
Number.The definition of Morlet wavelet function is by expression formula ψ (x)=cos (1.75x) exp {-x2/ 2 } it describes, ψa,bIt (x) is corresponding
Substrate wavelet function, as the activation primitive in hidden layer；
(2) SRWNN model performance index, loss function are defined are as follows:
Wherein E is the loss function of result test, and α is time intensive parameter, tnIt is training data any time value, tlIt is trained
The latest value of data, γ are the preset values according to training data,It is the corresponding average value of training sample, output valve error are as follows:
Using gradient descent method loss function is minimized, until loss function is less than default error threshold ξ=10-5；
(3) it trains SRWNN model and carries out parameter update:
The gradient of input layer connection weight are as follows:
Wherein η is learning rate, Schistosomiasis control rate；For the derivative of SRWNN model activation primitive；
Correspondingly, the gradient of recurrence layer connection weight are as follows:
The connection weight gradient of hidden layer are as follows:
The two indices a of wavelet functionj、bjGradient be respectively as follows:
The parameter of SRWNN model more new formula is respectively as follows:
(4) the objective function error size for judging current time water outlet COD and ammonia nitrogen repeats step (3) as E > ξ；If E <
When ξ, then the output y that step (1) calculates SRWNN controller is gone to1(t),y2(t)。
3. a kind of ship sewage treatment control forecasting side based on Stochastic Recursive wavelet neural network according to claim 2
Method, it is characterized in that: SRWNN controller output order is opened when detecting current time temperature value x1 (t) less than 15 DEG C of setting value
Open heating device；Until temperature value x1 (t) is increased to 30 DEG C, heating device is closed.
4. a kind of ship sewage treatment control based on Stochastic Recursive wavelet neural network according to claim 2 or 3 is pre-
Survey method, it is characterized in that: water outlet TN predicted value y2 (t), which is greater than, to be set when water outlet COD predicted value y1 (t) is greater than the set value 125mg/L
When determining 20mg/L, flow of inlet water regulating valve, reduction flow of inlet water to original 1/2 are opened, and open the power regulation of aeration pump
Device increases air inflow, and then increases dissolved oxygen concentration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811573025.5A CN109534486B (en) | 2018-12-21 | 2018-12-21 | Ship sewage treatment control prediction system and prediction method based on random recursion wavelet neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811573025.5A CN109534486B (en) | 2018-12-21 | 2018-12-21 | Ship sewage treatment control prediction system and prediction method based on random recursion wavelet neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109534486A true CN109534486A (en) | 2019-03-29 |
| CN109534486B CN109534486B (en) | 2023-07-25 |
Family
ID=65856419
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811573025.5A Active CN109534486B (en) | 2018-12-21 | 2018-12-21 | Ship sewage treatment control prediction system and prediction method based on random recursion wavelet neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109534486B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109901598A (en) * | 2019-04-08 | 2019-06-18 | 哈尔滨工程大学 | Autonomous underwater robot path following method based on stochastic model Prediction and Control Technology |
| CN110909922A (en) * | 2019-11-08 | 2020-03-24 | 四川大学 | Water resource efficiency detection and prediction method |
| CN111400651A (en) * | 2020-03-12 | 2020-07-10 | 干天伟 | Water environment regulation method and device based on water environment sampling |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101387632A (en) * | 2008-10-17 | 2009-03-18 | 北京工业大学 | Soft measurement method for biochemical oxygen demand BOD in process of sewage disposal |
| CN101833314A (en) * | 2010-03-30 | 2010-09-15 | 深圳达实智能股份有限公司 | Sewage treatment control system and sewage treatment control method |
| CA2706086A1 (en) * | 2010-05-31 | 2011-11-30 | Ecovu Analytics Inc. | Method and system for fluid purification and analysis |
| CN102616927A (en) * | 2012-03-28 | 2012-08-01 | 中国科学技术大学 | Adjusting method of technological parameters of sewage treatment and device |
| CN103235096A (en) * | 2013-04-16 | 2013-08-07 | 广州铁路职业技术学院 | Sewage water quality detection method and apparatus |
| US20140052422A1 (en) * | 2011-02-14 | 2014-02-20 | Jinquan Wan | Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network |
| CN103809557A (en) * | 2013-12-30 | 2014-05-21 | 北京工业大学 | Neural network based sewage disposal process optimal control method |
| CN104360035A (en) * | 2014-11-02 | 2015-02-18 | 北京工业大学 | Sewage TP soft measurement method based on self-organized particle swarm and radial basis function neural network |
| CN104376380A (en) * | 2014-11-17 | 2015-02-25 | 北京工业大学 | Ammonia concentration predicting method based on recursion self-organization neural network |
| 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 |
| CN106295800A (en) * | 2016-07-28 | 2017-01-04 | 北京工业大学 | A kind of water outlet total nitrogen TN intelligent detecting method based on recurrence Self organizing RBF Neural Network |
| CN106682316A (en) * | 2016-12-29 | 2017-05-17 | 北京工业大学 | Real-time effluent total-phosphorus monitoring system based on peak radial basis function neural network |
| CN107665363A (en) * | 2016-07-30 | 2018-02-06 | 复凌科技（上海）有限公司 | A kind of water quality hard measurement Forecasting Methodology of total phosphorus |
| CN108469507A (en) * | 2018-03-13 | 2018-08-31 | 北京工业大学 | A kind of water outlet BOD flexible measurement methods based on Self organizing RBF Neural Network |
-
2018
- 2018-12-21 CN CN201811573025.5A patent/CN109534486B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101387632A (en) * | 2008-10-17 | 2009-03-18 | 北京工业大学 | Soft measurement method for biochemical oxygen demand BOD in process of sewage disposal |
| CN101833314A (en) * | 2010-03-30 | 2010-09-15 | 深圳达实智能股份有限公司 | Sewage treatment control system and sewage treatment control method |
| CA2706086A1 (en) * | 2010-05-31 | 2011-11-30 | Ecovu Analytics Inc. | Method and system for fluid purification and analysis |
| US20140052422A1 (en) * | 2011-02-14 | 2014-02-20 | Jinquan Wan | Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network |
| CN102616927A (en) * | 2012-03-28 | 2012-08-01 | 中国科学技术大学 | Adjusting method of technological parameters of sewage treatment and device |
| CN103235096A (en) * | 2013-04-16 | 2013-08-07 | 广州铁路职业技术学院 | Sewage water quality detection method and apparatus |
| CN103809557A (en) * | 2013-12-30 | 2014-05-21 | 北京工业大学 | Neural network based sewage disposal process optimal control method |
| CN104360035A (en) * | 2014-11-02 | 2015-02-18 | 北京工业大学 | Sewage TP soft measurement method based on self-organized particle swarm and radial basis function neural network |
| CN104376380A (en) * | 2014-11-17 | 2015-02-25 | 北京工业大学 | Ammonia concentration predicting method based on recursion self-organization neural network |
| US20160140437A1 (en) * | 2014-11-17 | 2016-05-19 | Beijing University Of Technology | Method to predict the effluent ammonia-nitrogen concentration based on a recurrent self-organizing neural network |
| 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 |
| CN106295800A (en) * | 2016-07-28 | 2017-01-04 | 北京工业大学 | A kind of water outlet total nitrogen TN intelligent detecting method based on recurrence Self organizing RBF Neural Network |
| CN107665363A (en) * | 2016-07-30 | 2018-02-06 | 复凌科技（上海）有限公司 | A kind of water quality hard measurement Forecasting Methodology of total phosphorus |
| CN106682316A (en) * | 2016-12-29 | 2017-05-17 | 北京工业大学 | Real-time effluent total-phosphorus monitoring system based on peak radial basis function neural network |
| CN108469507A (en) * | 2018-03-13 | 2018-08-31 | 北京工业大学 | A kind of water outlet BOD flexible measurement methods based on Self organizing RBF Neural Network |
Non-Patent Citations (4)
| Title |
|---|
| 朱启兵等: "基于小波神经网络的污水处理厂出水水质预测", 《计算机工程与应用》 * |
| 朱启兵等: "基于小波神经网络的污水处理厂出水水质预测", 《计算机工程与应用》, vol. 43, no. 28, 31 October 2017 (2017-10-31), pages 15 - 17 * |
| 王义权: "极地航行船舶的生活污水处理装置效能研究及设计", 《中国优秀硕士论文全文数据库 工程科技Ⅰ辑》 * |
| 王义权: "极地航行船舶的生活污水处理装置效能研究及设计", 《中国优秀硕士论文全文数据库 工程科技Ⅰ辑》, 28 February 2022 (2022-02-28), pages 1 - 93 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109901598A (en) * | 2019-04-08 | 2019-06-18 | 哈尔滨工程大学 | Autonomous underwater robot path following method based on stochastic model Prediction and Control Technology |
| CN110909922A (en) * | 2019-11-08 | 2020-03-24 | 四川大学 | Water resource efficiency detection and prediction method |
| CN110909922B (en) * | 2019-11-08 | 2022-07-01 | 四川大学 | Water resource efficiency detection and prediction method |
| CN111400651A (en) * | 2020-03-12 | 2020-07-10 | 干天伟 | Water environment regulation method and device based on water environment sampling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109534486B (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109534486A (en) | A kind of ship sewage treatment control forecasting system and prediction technique based on Stochastic Recursive wavelet neural network | |
| US9747544B2 (en) | Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network | |
| CN109523094B (en) | Ship salty sewage treatment control prediction system and prediction method based on wavelet neural network | |
| Vyas et al. | Performance forecasting of common effluent treatment plant parameters by artificial neural network | |
| CN101334395B (en) | COD soft sensing process | |
| CN106295800A (en) | A kind of water outlet total nitrogen TN intelligent detecting method based on recurrence Self organizing RBF Neural Network | |
| de Canete et al. | Control and soft sensing strategies for a wastewater treatment plant using a neuro-genetic approach | |
| CN106986449A (en) | A kind of Sewage from Ships processing control system and control method | |
| CN108536106A (en) | A kind of aerating system dissolved oxygen based on Kalman filtering-extreme learning machine regulates and controls method online | |
| CN113274885B (en) | Membrane pollution intelligent early warning method applied to membrane sewage treatment | |
| CN108911301B (en) | Modularized organic wastewater treatment system | |
| Yetilmezsoy et al. | Adaptive neuro-fuzzy inference-based modeling of a full-scale expanded granular sludge bed reactor treating corn processing wastewater | |
| KR20150012709A (en) | Wastewater discharge system of the ship | |
| Naser et al. | Simulation of low TDS and biological units of Fajr industrial wastewater Treatment plant using artificial neural network and principal component analysis hybrid method | |
| Cohen et al. | An intelligent controller for automated operation of sequencing batch reactors | |
| Fu et al. | Dissolved oxygen control system based on the TS fuzzy neural network | |
| Ayaz et al. | Forecasting of heavy metal contamination in coastal sea surface waters of the karachi harbour area by neural network approach | |
| CN116332427A (en) | Water treatment device based on electrochemical method and application thereof | |
| Cao et al. | Deep learning based soft sensor for microbial wastewater treatment efficiency prediction | |
| Xu et al. | A novel dissolve oxygen control method based on fuzzy neural network | |
| CN114417622A (en) | Sewage treatment method, device, equipment and system | |
| Hamed et al. | Performance simulation of H-TDS unit of FAJR industrial wastewater treatment plant using a combination of neural network and principal component analysis | |
| Li et al. | Bibliometric analysis of artificial intelligence in wastewater treatment: Current status, research progress, and future prospects | |
| Husin et al. | Improving total nitrogen removal using a neural network ammonia-based aeration control in activated sludge process | |
| CN105550495A (en) | Method for evaluating ship navigation increasing resistance level |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |