CN1257110C - Flocculating dispensing compound control method - Google Patents
Flocculating dispensing compound control method Download PDFInfo
- Publication number
- CN1257110C CN1257110C CN 200410043892 CN200410043892A CN1257110C CN 1257110 C CN1257110 C CN 1257110C CN 200410043892 CN200410043892 CN 200410043892 CN 200410043892 A CN200410043892 A CN 200410043892A CN 1257110 C CN1257110 C CN 1257110C
- Authority
- CN
- China
- Prior art keywords
- value
- neurone
- turbidity
- raw water
- detected value
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Separation Of Suspended Particles By Flocculating Agents (AREA)
Abstract
The present invention relates to a chemical feeding control technology for water treatment, particularly to a compound control method for flocculation feeding. The compound control method comprises the steps: a turbidity setting value (5) of effluent in a sedimentation tank is subtracted from a turbidity detecting value of effluent in the sedimentation tank to obtain a turbidity deviation value x3 (6) of effluent in the sedimentation tank; the x3 carries out integral operation (7); the turbidity deviation value x3 (6) of effluent in the sedimentation tank is used, a flow rate detecting value x1 (3) of raw water and a turbidity detecting value x2 (4) of raw water are used as input values of neuronal operation (8); the flow rate detecting value x1 (3) of raw water and the turbidity detecting value x2 (4) of raw water are used as parameters of feedforward proportion operation (9); an integral operation result and a neuronal operation result (10) are added to obtain a middle parameter setting value (11); a middle parameter detecting value (12) is used; a middle parameter setting value (13) is subtracted from the middle parameter detecting value to obtain a middle parameter deviation value x (14); the x carries out proportional integral operation (15); and a proportional integral operation result and a feedforward proportion operation result (16) are multiplied to obtain the feeding amount of flocculating agents. The method can control a flocculation feeding technology with uncertainty and strong nonlinearity and can be used for water treatment in water works.
Description
Technical field:
The present invention relates to a kind of water treatment dispensing control techniques, be specifically related to a kind of flocculation dispensing composite control method.
Background technology:
The flocculation dispensing is the important step of purification of water quality, and how controlling adding of flocculation agent then is problem demanding prompt solution in the present water and wastewater industry.The dispensing control techniques that tradition is used the streaming current detector belongs to the back feedback control system that only detects intermediate parameters, has hysteresis characteristic, lag behind and pure hysteresis as capacity, reach new steady state from the generation of interference effect to controlled parameter and will experience considerable time.The quality of its control effect depends on the dependency of intermediate parameters and treatment process, and this dependency must in time be revised and could guarantee effluent quality along with the variation of raw water quality, the water yield and technological process changes in the practical application.Water treatment procedure adds from flocculation agent, at least need through more than the dozens of minutes through mixing, react, precipitating, the large time delay of system, non-linear causing can't be used conventional art and come the update the system set(ting)value according to effluent index, many water factories raw water turbidity and raw water flow change greatly, violent to system shock, only rely on the back feedback control can not meet the demands, thereby press for the advanced practical novel flocculation dosage control method of development.
Summary of the invention:
The purpose of this invention is to provide a kind of water treatment flocculation dispensing composite control method, it can realize the water treatment flocculation dispensing middle process with uncertain and strong nonlinearity is controlled in real time.The step of composite control method of the present invention is: get settling tank delivery turbidity detected value 1, get settling tank delivery turbidity set(ting)value 2; Settling tank delivery turbidity detected value is deducted settling tank delivery turbidity set(ting)value 5, obtain settling tank delivery turbidity deviate x
36, to x
3Carry out integral operation 7, operational formula is: f (x
3)=k
3∫ x
3Dt
3, k in the formula
3Be constant, t
3Be the time, f (x
3) be the integral operation result; Get settling tank delivery turbidity deviate x
36, get raw water flow detected value x
13, get raw water turbidity detected value x
24, all as the input value of neurone computing 8; The calculation step of described neurone computing 8 is: get raw water flow detected value x
13, get raw water turbidity detected value x
24, get neurone operation result 8-5, all as the input parameter of neurone positive model computing 8-1; Get settling tank delivery turbidity deviate x
36; Make settling tank delivery turbidity deviate x
3Additive operation 8-2 with neurone positive model operation result; Operation result is with reference to input value 8-3; Get raw water turbidity detected value x
24, get raw water flow detected value x
13, get settling tank delivery turbidity deviate x
36 and with reference to input value 8-3, all as the input parameter of neurone inversion model computing 8-4; Obtain neurone operation result 8-5, neurone operation result 8-5 constitutes closed loop control system as the input parameter value of neurone positive model computing 8-1 simultaneously; Get raw water flow detected value x
13, get raw water turbidity detected value x
24, all as the parameter of feedforward scale operation 9, operational formula is: f (x
1, x
2)=k
1X
1.x
2 n, k in the formula
1Drawn by test-results, n is an experience factor, is drawn f (x by test
1, x
2) be the result of feedforward scale operation; With integral operation f (x as a result
3) carry out addition 10 with the neurone operation result; Above-mentioned operation result is as intermediate parameters set(ting)value 11, get intermediate parameters detected value 12, the intermediate parameters detected value is deducted intermediate parameters set(ting)value 13, draw intermediate parameters deviate x14, x is carried out proportional integral computing 15, and operational formula is: f (x)=kx+k ∫ xdt, and k is a constant in the formula, t is the time, and f (x) is the proportional integral operation result; With the proportional integral operation result f (x) and the scale operation f (x as a result that feedovers
1, x
2) multiply each other 16; Its result is the dosage 17 of this moment flocculation agent.
The present invention has following beneficial effect: 1, the present invention be with water technology settling tank delivery turbidity detected value as final controlled target, make the settling tank delivery turbidity qualified by adjusting flocculant dosage.Utilize this composite control method can solve that existing water treatment dispensing Controlling System can not reflect rapidly that the water quality and the water yield change, retardation time is long,, system poor with resemble process is difficult for shortcomings such as stable, can realize the water treatment flocculation dispensing technology with uncertain and strong nonlinearity is controlled in real time, and have stronger antagonism interference rejection capability.2, neurone adopts the internal model control mode, promptly uses two neural networks, and one is the neurone positive model; One is the neurone inversion model, under the stable situation of flocculation administration system, the characteristics of utilizing self-study habit, the adaptivity of neural network and can approaching nonlinear function are arbitrarily set up intermediate parameters set(ting)value deviation predictive model, and neurone operation result and integral operation results added are revised the intermediate parameters set(ting)value automatically.3, use the water that this composite control method is handled, its water quality qualification rate reaches more than 99%, and can save flocculation dosing 20～40%, thereby this composite control method is better than traditional control method widely, and it has good environmental benefit and social benefit.
Description of drawings:
Fig. 1 is the schema of total body controlling means of the present invention, and Fig. 2 is the schema of neurone operational method, and Fig. 3 is the structural representation of neurone positive model, and Fig. 4 is the structural representation of neurone inversion model.
Embodiment:
The step of embodiment of the present invention is described: get settling tank delivery turbidity detected value 1, get settling tank delivery turbidity set(ting)value 2 (by artificial setting) in conjunction with Fig. 1, Fig. 2, Fig. 3, Fig. 4; Settling tank delivery turbidity detected value is deducted settling tank delivery turbidity set(ting)value 5, obtain settling tank delivery turbidity deviate x
36, to x
3Carry out integral operation 7, operational formula is: f (x
3)=k
3∫ x
3Dt
3, k in the formula
3Be constant, t
3Be the time, f (x
3) be the integral operation result; Get settling tank delivery turbidity deviate x
36, get raw water flow detected value x
13, get raw water turbidity detected value x
24, all as the input value of neurone computing 8; Get raw water flow detected value x
13, get raw water turbidity detected value x
24, all as the parameter of feedforward scale operation 9, operational formula is: f (x
1, x
2)=k
1X
1.x
2 n, k in the formula
1Drawn by test-results, n is an experience factor, is drawn f (x by test
1, x
2) be the result of feedforward scale operation; With integral operation f (x as a result
3) carry out addition 10 with the neurone operation result; Above-mentioned operation result is as intermediate parameters set(ting)value 11, get intermediate parameters detected value 12, the intermediate parameters detected value is deducted intermediate parameters set(ting)value 13, draw intermediate parameters deviate x14, x is carried out proportional integral computing 15, and operational formula is: f (x)=kx+k ∫ xdt, and wherein k is a constant, t is the time, and f (x) is the proportional integral operation result; With the proportional integral operation result f (x) and the scale operation f (x as a result that feedovers
1, x
2) multiply each other 16; Its result is the dosage 17 of this moment flocculation agent.Described neurone computing 8 comprises neurone positive model computing 8-1 and neurone inversion model computing 8-4, and neurone computing 8 is finished by following steps: get raw water flow detected value x
13, get raw water turbidity detected value x
24, get neurone operation result 8-5, all as the input parameter of neurone positive model computing 8-1; Get settling tank delivery turbidity deviate x
36; Make settling tank delivery turbidity deviate x
3Additive operation 8-2 with neurone positive model operation result; Operation result is with reference to input value 8-3; Get raw water turbidity detected value x
24, get raw water flow detected value x
13, get settling tank delivery turbidity deviate x
36 and with reference to input value 8-3, all as the input parameter of neurone inversion model computing 8-4; Obtain neurone operation result 8-5, neurone operation result 8-5 constitutes closed loop control system as the input parameter value of neurone positive model computing 8-1 simultaneously.
The neurone positive model is made up of input layer 18-1, middle layer 18-2, output layer 18-3: by raw water turbidity detected value x
24, raw water flow detected value x
13, intermediate parameters set(ting)value 11 is as the parameter of the input layer 18-1 of neurone positive model network 18, with the basic structure of three layers of feedforward BP neural network as neurone positive model network 18, through middle layer 18-2 computing, finally with settling tank delivery turbidity deviate x
36 parameters as output layer 18-3.The training method of neurone positive model network 18 is: choose representative water factory's service data and neurone positive model network 18 is trained or learn, under the prerequisite of the error precision that satisfies the requirement of settling tank delivery turbidity, determine the connection weight coefficient between neurone positive model network 18 each layers.
The neurone inversion model is made up of input layer 19-1, middle layer 19-2, output layer 19-3: by raw water turbidity detected value x
24, raw water flow detected value x
13, settling tank delivery turbidity deviate x
36 and with reference to the parameter of input value 8-3 as the input layer 19-1 of neurone inversion model network 19, with the basic structure of three layers of feedforward BP neural network as neurone inversion model network 19, through middle layer 19-2 computing, final with the parameter of intermediate parameters set(ting)value 11 as output layer 19-3.Neurone inversion model network 19 training methods are: with the input value of settling tank delivery turbidity deviate 6 as neurone inversion model network 19, and settling tank delivery turbidity deviate x
3The reference input value 8-3 that draws with the additive operation 8-2 of neurone positive model operation result is used for training, and learns the inverse characteristic of intermediate parameters object indirectly, makes neurone inversion model network 19 set up the inversion model of system by study.Still, choose representational water factory service data neurone inversion model network 19 is repeatedly trained, obtain the connection weight coefficient between each layer according to the general method of network training.
Claims (1)
1. a flocculation dispensing composite control method is characterized in that the step of composite control method is: get settling tank delivery turbidity detected value (1), get settling tank delivery turbidity set(ting)value (2); Settling tank delivery turbidity detected value is deducted settling tank delivery turbidity set(ting)value (5), obtain settling tank delivery turbidity deviate x
3(6), to x
3Carry out integral operation (7), operational formula is: f (x
3)=k
3∫ x
3Dt
3, k in the formula
3Be constant, t
3Be the time, f (x
3) be the integral operation result; Get settling tank delivery turbidity deviate x
3(6), get raw water flow detected value x
1(3), get raw water turbidity detected value x
2(4), all as the input value of neurone computing (8); The calculation step of described neurone computing (8) is: get raw water flow detected value x
1(3), get raw water turbidity detected value x
2(4), get neurone operation result (8-5), all as the input parameter of neurone positive model computing (8-1); Get settling tank delivery turbidity deviate x
3(6); Make settling tank delivery turbidity deviate x
3Additive operation (8-2) with neurone positive model operation result; Operation result is with reference to input value (8-3); Get raw water turbidity detected value x
2(4), get raw water flow detected value x
1(3), get settling tank delivery turbidity deviate x
3(6) with reference to input value (8-3), all as the input parameter of neurone inversion model computing (8-4); Obtain neurone operation result (8-5), neurone operation result (8-5) constitutes closed loop control system as the input parameter value of neurone positive model computing (8-1) simultaneously; Get raw water flow detected value x
1(3), get raw water turbidity detected value x
2(4), all as the parameter of feedforward scale operation (9), operational formula is f (x
1, x
2)=k
1X
1.x
2 n, k in the formula
1Drawn by test-results, n is an experience factor, is drawn f (x by test
1, x
2) be the result of feedforward scale operation; With integral operation f (x as a result
3) carry out addition (10) with the neurone operation result; Above-mentioned operation result is as intermediate parameters set(ting)value (11), get intermediate parameters detected value (12), the intermediate parameters detected value is deducted intermediate parameters set(ting)value (13), draw intermediate parameters deviate x (14), x is carried out proportional integral computing (15), and operational formula is: f (x)=kx+k ∫ xdt, and k is a constant in the formula, t is the time, and f (x) is the proportional integral operation result; With the proportional integral operation result f (x) and the scale operation f (x as a result that feedovers
1, x
2) multiply each other (16); Its result is the dosage (17) of this moment flocculation agent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410043892 CN1257110C (en) | 2004-09-24 | 2004-09-24 | Flocculating dispensing compound control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410043892 CN1257110C (en) | 2004-09-24 | 2004-09-24 | Flocculating dispensing compound control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1597548A CN1597548A (en) | 2005-03-23 |
| CN1257110C true CN1257110C (en) | 2006-05-24 |
Family
ID=34665505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410043892 Expired - Fee Related CN1257110C (en) | 2004-09-24 | 2004-09-24 | Flocculating dispensing compound control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1257110C (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4492473B2 (en) * | 2005-07-27 | 2010-06-30 | 株式会社日立製作所 | Flocculant injection control device and method |
| CN101825870B (en) * | 2010-05-18 | 2013-01-02 | 浙江浙大中控信息技术有限公司 | Method and system for controlling supply quantity of water-treatment flocculating agent |
| CN105923724B (en) * | 2016-05-16 | 2019-04-12 | 河南理工大学 | Underground coal mine waterpower punching coal slime water purification processing method |
| CN107055732A (en) * | 2017-05-17 | 2017-08-18 | 北京易沃特科技有限公司 | A kind of heavy metals minimizing technology and device |
| CN110426957B (en) * | 2019-07-31 | 2020-03-13 | 深圳信息职业技术学院 | Water plant dosing system self-adaptive sliding mode control method based on time delay observer |
-
2004
- 2004-09-24 CN CN 200410043892 patent/CN1257110C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1597548A (en) | 2005-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106873379B (en) | Sewage treatment optimal control method based on iterative ADP algorithm | |
| CN103197539B (en) | The method of wastewater disposal intelligent optimization control aeration quantity | |
| US20190359510A1 (en) | Cooperative optimal control method and system for wastewater treatment process | |
| CN110054274B (en) | Water purification flocculation precipitation dosing control method | |
| CN106406094B (en) | A kind of sewage treatment dissolved oxygen concentration tracking and controlling method based on two type fuzzy neural network of section | |
| CN106802563B (en) | A kind of sewage procedure optimization control method based on drosophila optimization and LSSVM | |
| CN101561661A (en) | Fuzzy control method and fuzzy controller | |
| CN104238527A (en) | Precise control method of sewage treatment plant aeration total quantity | |
| CN103570190B (en) | Method for controlling feeding amount of chemical phosphorus removal reagents of water reclamation plants based on fuzzy control | |
| CN103011356A (en) | Method for controlling automatic chemical dosing of high-turbidity water system | |
| CN103810309B (en) | A based on bounding theory2the soft-measuring modeling method of O urban sewage treatment process | |
| CN111498974A (en) | Intelligent dosing control system based on BP neural network | |
| CN105301960B (en) | A kind of control method of tap water flocculant dosage | |
| CN101825870A (en) | Method and system for controlling supply quantity of water-treatment flocculating agent | |
| CN110647037A (en) | Cooperative control method for sewage treatment process based on two-type fuzzy neural network | |
| CN107720946B (en) | Cascade control method in SBR sewage treatment process | |
| CN1257110C (en) | Flocculating dispensing compound control method | |
| CN111470628A (en) | Carbon source medicament adding equipment and adding method | |
| CN105807615A (en) | Fuzzy feedforward-feedback controller | |
| CN101993151A (en) | Loop control method for biochemical sewage treatment process | |
| CN108536106A (en) | A kind of aerating system dissolved oxygen based on Kalman filtering-extreme learning machine regulates and controls method online | |
| CN106354019B (en) | A kind of dissolved oxygen accuracy control method based on RBF neural | |
| CN111580381A (en) | Dissolved oxygen control method of dynamic event-driven control strategy | |
| CN102540897A (en) | Method for carrying out composite predictive control on coagulant dosage of water treatment | |
| CN1270219C (en) | Self-adaptive control administration apparatus and process thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20060524 Termination date: 20180924 |