CN103235096A - Sewage water quality detection method and apparatus - Google Patents
Sewage water quality detection method and apparatus Download PDFInfo
- Publication number
- CN103235096A CN103235096A CN2013101319077A CN201310131907A CN103235096A CN 103235096 A CN103235096 A CN 103235096A CN 2013101319077 A CN2013101319077 A CN 2013101319077A CN 201310131907 A CN201310131907 A CN 201310131907A CN 103235096 A CN103235096 A CN 103235096A
- Authority
- CN
- China
- Prior art keywords
- water outlet
- outlet total
- concentration
- value
- sewage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Activated Sludge Processes (AREA)
Abstract
The present invention discloses a sewage water quality detection method and an apparatus, wherein data training is performed to pre-establish a prediction model based on a relevance vector machine, and the total nitrogen and the total phosphorus in sewage are predicted by using the prediction model. Compared to the manual detection method, the method and the apparatus have the following characteristics that: online prediction can be achieved, real-time monitoring and regulation are easily performed, and contribution is provided for real-time monitoring automation of sewage water quality. In addition, the prediction model adopted by the method and the apparatus is a soft measurement method based on the relevance vector machine, and has better applicability and higher prediction accuracy compared with a model established by adopting a neural network and support vector machine modeling method.
Description
Technical field
The present invention relates to the sewage detection technical field, particularly relate to a kind of sewage quality detection method and device.
Background technology
Quantity of wastewater effluent increases day by day along with the development of development of urbanization and industrial and agricultural production.China builds a large amount of sewage treatment plants in recent years, improves water resource environment, in order to avoid its further deterioration.Because sewage disposal process mechanism complexity for setting up good monitoring mechanism, guarantees good effluent quality, must in time monitor the water quality parameter in the sewage disposal process.
According to national relevant emission standards, total nitrogen, total phosphorus are the important indicators of weighing the water quality quality.Yet the detection of total nitrogen, total phosphorus at present is by manually finishing, and there is very big property time lag in manual detection, and the testing process complexity, error occurs easily.
Summary of the invention
Based on above-mentioned situation, the present invention proposes a kind of sewage quality detection method and device, to reduce people's participation, realize that rapidly and accurately sewage quality detects.
A kind of sewage quality detection method comprises step:
Monitoring sewage technological parameter;
With described sewage technological parameter input training in advance based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus;
The data of output water outlet total nitrogen or water outlet total phosphorus.
A kind of sewage quality pick-up unit comprises:
The Parameters Monitoring module is used for monitoring sewage technological parameter;
The model load module, be used for described sewage technological parameter input training in advance based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus;
The output module that predicts the outcome is used for the data of output water outlet total nitrogen or water outlet total phosphorus.
Sewage quality detection method of the present invention and device are trained the forecast model of setting up in advance based on the interconnection vector machine by data, adopt this forecast model that the total nitrogen in the sewage and total phosphorus are predicted.Compare the method for manual detection, this method can on-line prediction with device, is convenient to monitor in real time and regulate, for the real-time monitoring automation of sewage quality has been made contribution.In addition, the forecast model that this method and device adopt is based on the flexible measurement method of interconnection vector machine, compares the model that adopts neural network and support vector machine modeling method to set up, and has the precision of prediction of better applicability and Geng Gao.
Description of drawings
Fig. 1 is the schematic flow sheet of sewage quality detection method of the present invention;
Fig. 2 is the structural representation of sewage quality pick-up unit of the present invention.
Embodiment
The sewage quality testing process is a multivariate, multiple goal, comprises the complex process of magnanimity information processing at many levels, exists strong coupling with related between the various parameters.Some important index is difficult to on-line measurement, for example: the value of BOD5, N, P, in order to monitor these indexs better, can utilize the soft-measuring technique that emerges in recent years, mathematical model (soft-sensing model) between the process variable to be measured of namely utilizing the process variable of easily measuring and being difficult to directly measure, by various mathematical computations and method of estimation, realize the measurement of process variable to be measured with computer software.Support vector machine (SVM, Support Vector Machine) is a kind of soft-measuring modeling method based on Statistical Learning Theory, its target is constructed a decision function exactly, test data is as far as possible correctly classified and predict unknown classification, interconnection vector machine (RVM, Relevance Vector Machine) then be a kind of new computing method that grow up on the basis of support vector machine, both essence all is by means of kernel function the linear inseparable problem of lower dimensional space to be converted to linear partition problem in the higher dimensional space.Relative SVM model, RVM is the learning algorithm with sparse probability model that proposes under Bayesian frame, can obtain predicting like the probability.In the forecasting process of this paper total phosphorus TP and total nitrogen TN, compare with SVM, at first, the generalization ability that RVM is good can make to predict the outcome and reach the precision suitable with SVM; Secondly, in the RVM model, do not have the regularization coefficient, do not need to obtain this parameter by cross validation; Again, in the RVM solution procedure, kernel function needn't satisfy the Mercer condition; At last, after the RVM training is finished, have only the weights non-zero of minority basis function, more sparse than SVM, therefore more can analyze the uncertainty of representing with probability that predicts the outcome clearly.
To sum up, RVM makes up based on Bayesian frame, and its generalization ability is better than support vector machine, and its test duration is shorter, more is applicable to online detection.For this reason, this paper proposes to introduce this flexible measurement method of RVM, sets up the forecast model of important indicator total nitrogen (TN) and total phosphorus (TP) in the water quality.Explain the present invention in detail below in conjunction with accompanying drawing.
Sewage quality detection method of the present invention as shown in Figure 1, comprises the steps:
Step S101, monitoring sewage technological parameter;
Step S102, with described sewage technological parameter input training in advance based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus;
The data of step S103, output water outlet total nitrogen or water outlet total phosphorus.
It is the content of water outlet total nitrogen and water outlet total phosphorus that the purpose that sewage quality detects detects, in view of total nitrogen and total phosphorus are difficult to measure, therefore as described in step S101, monitor other technological parameters that are easy to measure earlier, these parameters are input in the forecast model, output namely is the data of total nitrogen or total phosphorus at last again.
Choosing of input variable in the forecast model (technological parameter), can be based on the mechanism of biological treatment and understanding and the experience of actual condition are determined, biological carbon and phosphorous removal is undertaken by two steps of nitrification and denitrification, in conjunction with the production Monthly Bulletin of Statistics Section of the major influence factors in the nitrification and denitrification course of reaction in the sewage disposal process and certain Sewage Plant and the data available in the technology operational factor form, water outlet TN forecast model input variable is by flow of inlet water, water inlet concentration of suspension SS and water inlet NH in this method
4 +-N value, aeration tank oxygen DO, temperature T, potential of hydrogen PH, redox electricity ORP, mixed liquor suspended solid, MLSS concentration MLSS, No
3 --N and aeration tank conductivity k and water outlet concentration of suspension SS, NH
4 +12 parameters such as-N are formed; Water outlet TP forecast model input variable is by flow of inlet water, water inlet concentration of suspension SS and aeration tank oxygen DO, temperature T, potential of hydrogen PH, redox electricity ORP, mixed liquor suspended solid, MLSS concentration MLSS, No
3 -10 parameters such as-N and aeration tank conductivity k and water outlet concentration of suspension SS are formed.
Described water outlet total nitrogen and the derivation of the forecast model of water outlet total phosphorus based on the interconnection vector machine of step S102 is as follows:
For given data set { x
i, t
i}
l I=1, x
i∈ R
d, t
i∈ R, d are the dimension of data centralization vector.Consider that desired value is a scalar and is subjected to noise effect, can the objective definition value be t=y (x, w)+ε, wherein ε is 0 for obeying average, variance is σ
2The Gaussian distribution noise.The Gaussian distributed of desired value, so its probability expression is:
p(t
i|y(x
i),σ
2)=N(y(x
i,w
i),σ
2) (1)
Wherein
Be a value that is determined by kernel function, l is the size of data set, t=(t
1, t
2..., t
l), x=(x
1, x
2..., x
l), w is weight vector, w=(w
0, w
1..., w
l),
Be kernel function, the kernel function of this paper is used radially basic kernel function commonly used.Suppose t
iIndependent same distribution, then the likelihood function of whole data set can be expressed as:
In the formula: w=(w
1, w
2..., w
l) be the weights of model, l is sample number, Φ is the kernel function mapping of input,
For all basis functions to the input x
iResponse.Under Bayesian frame, in order to improve the generalization ability of model, come the training pattern weight w with maximum likelihood method.Define each weights and obey Gauss's prior probability distribution, its expression formula is:
In the formula: α, α=(α
1, α
2..., α
l) be the hyper-function of the prior distribution of weight w.In conjunction with formula (2) and formula (3), the posterior probability of calculating weights according to bayesian criterion distributes:
p(w|t,α,σ)=N(μ,Σ) (4)
μ=α wherein
-2Σ Φ
TT, Σ=(α
-2Φ
TΦ+A)
-1, A=Dia (α
0, α
1..., α
l), wherein Dia is for producing with eigenwert (α
0, α
1..., α
l) matrix formed.The posteriority of formula (4) expression weights distributes and is determined by average μ and Σ.In order to estimate the weights model, to determine to estimate the optimal value of hyper-function α earlier.According to Bayesian frame, the likelihood of hyper-function distributes and can calculate by following formula:
Distribute by finding the solution maximum likelihood, can obtain its optimal value of hyper-function a
MPAnd σ
MP
For input value x
*, data set is trained the optimal value that obtains hyper-function, in conjunction with formula (1) and formula (5), the probability distribution that obtains the corresponding output of desired value is
Vectorial t in the formula
*Be x
*Predicted value, its average is
Variance is represented its uncertainty, and formula is
Above-mentioned average and variance, (x w)+defined formula of ε, obtains vectorial t to be updated to the desired value t=y of the initial definition of substitution
*Predictor formula:
In the formula, t
*The content of expression water outlet total nitrogen or water outlet total phosphorus, μ
TBe the posteriority weights mean value that calculates according to formula (4),
The expression kernel function,
For the obedience average is 0, variance is
Gaussian distribution.
According to the forecast model of formula (7) training total nitrogen and total phosphorus, form by known data sample training in advance, the process of training comprises the steps:
(1) technological parameter and the target component of monitoring sewage disposal process are set up sample data x, t, initialization hyper-function α
iWith the Gaussian distribution variances sigma
2
(2) sample data is input to water outlet TP and TN forecast model based on the interconnection vector machine as input, trains.
(3) calculate weights posteriority statistic μ, Σ according to sample data;
(4) calculate the error of the output of all forecast models and target component, if error is less than certain certain value, changes step (5), otherwise recomputate α
iAnd σ
2, change step (2);
(5) deletion α
iAssociated vector weights and the basis function of → ∞ are set up water outlet TP and TN forecast model based on the interconnection vector machine.
After forecast model trains, as described in step S102 and S103, technological parameter is imported forecast model, output total nitrogen or total phosphorus data.
Sewage quality pick-up unit of the present invention as shown in Figure 2, comprising:
The Parameters Monitoring module is used for monitoring sewage technological parameter;
The model load module, be used for described sewage technological parameter input training in advance based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus;
The output module that predicts the outcome is used for the data of output water outlet total nitrogen or water outlet total phosphorus.
As a preferred embodiment, if the data of output water outlet total phosphorus, then Shu Ru described sewage technological parameter comprises flow of inlet water, water inlet concentration of suspension, aeration tank dissolved oxygen content, temperature, potential of hydrogen, oxidation-reduction potential, mixed liquor suspended solid, MLSS concentration, No
3 --N value, aeration tank conductivity and water outlet concentration of suspension; If the data of output water outlet total nitrogen, then Shu Ru described sewage technological parameter comprises flow of inlet water, water inlet concentration of suspension, aeration tank dissolved oxygen content, temperature, potential of hydrogen, oxidation-reduction potential, mixed liquor suspended solid, MLSS concentration, No
3 --N value, aeration tank conductivity, water outlet concentration of suspension, and water inlet NH
4 +-N value and water outlet NH
4 +-N value.
As a preferred embodiment, describedly based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus be:
In the formula, t
*The content of expression water outlet total nitrogen or water outlet total phosphorus, μ
TExpression posteriority weights mean value,
The expression kernel function,
For the obedience average is 0, variance is
Gaussian distribution.
In order to verify forecast model, adopt the Guangzhou Sewage Plant to gather 1000 groups of data and carry out soft sensor modeling, choose 250 samples, after 3 σ rule pre-service, remain 222 samples, choose 200 samples, wherein preceding 170 samples are set up model as training sample, and back 30 samples are as the generalization ability of three kinds of models in the test sample book check table 1.
As can be seen from Table 1, the modeling effect of contrast neural network and support vector machine, the interconnection vector machine has lifting on every performance.Based on the water outlet TN model prediction AME 0.82 of interconnection vector machine, max value of error 2.11, root-mean-square error RMSE=1.05; Water outlet TP model prediction AME 0.15, max value of error 0.42, root-mean-square error RMSE=0.19; Its fitting precision is higher than the model of neural network and support vector machine foundation, has shown that the interconnection vector machine has better generalization ability under the small sample situation.
The above embodiment has only expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but can not therefore be interpreted as the restriction to claim of the present invention.Should be pointed out that for the person of ordinary skill of the art without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (6)
1. a sewage quality detection method is characterized in that, comprises step:
Monitoring sewage technological parameter;
With described sewage technological parameter input training in advance based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus;
The data of output water outlet total nitrogen or water outlet total phosphorus.
2. sewage quality detection method according to claim 1 is characterized in that,
If the data of output water outlet total phosphorus, then Shu Ru described sewage technological parameter comprises flow of inlet water, water inlet concentration of suspension, aeration tank dissolved oxygen content, temperature, potential of hydrogen, oxidation-reduction potential, mixed liquor suspended solid, MLSS concentration, No
3 --N value, aeration tank conductivity and water outlet concentration of suspension;
If the data of output water outlet total nitrogen, then Shu Ru described sewage technological parameter comprises flow of inlet water, water inlet concentration of suspension, aeration tank dissolved oxygen content, temperature, potential of hydrogen, oxidation-reduction potential, mixed liquor suspended solid, MLSS concentration, No
3 --N value, aeration tank conductivity, water outlet concentration of suspension, water inlet NH
4 +-N value and water outlet NH
4 +-N value.
3. sewage quality detection method according to claim 1 and 2 is characterized in that, describedly based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus is:
4. a sewage quality pick-up unit is characterized in that, comprising:
The Parameters Monitoring module is used for monitoring sewage technological parameter;
The model load module, be used for described sewage technological parameter input training in advance based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus;
The output module that predicts the outcome is used for the data of output water outlet total nitrogen or water outlet total phosphorus.
5. sewage quality pick-up unit according to claim 4 is characterized in that,
If the data of output water outlet total phosphorus, then Shu Ru described sewage technological parameter comprises flow of inlet water, water inlet concentration of suspension, aeration tank dissolved oxygen content, temperature, potential of hydrogen, oxidation-reduction potential, mixed liquor suspended solid, MLSS concentration, No
3 --N value, aeration tank conductivity and water outlet concentration of suspension;
If the data of output water outlet total nitrogen, then Shu Ru described sewage technological parameter comprises flow of inlet water, water inlet concentration of suspension, aeration tank dissolved oxygen content, temperature, potential of hydrogen, oxidation-reduction potential, mixed liquor suspended solid, MLSS concentration, No
3 --N value, aeration tank conductivity, water outlet concentration of suspension, water inlet NH
4 +-N value and water outlet NH
4 +-N value.
6. according to claim 4 or 5 described sewage quality pick-up units, it is characterized in that, describedly based on the water outlet total nitrogen of interconnection vector machine and the forecast model of water outlet total phosphorus be:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013101319077A CN103235096A (en) | 2013-04-16 | 2013-04-16 | Sewage water quality detection method and apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013101319077A CN103235096A (en) | 2013-04-16 | 2013-04-16 | Sewage water quality detection method and apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103235096A true CN103235096A (en) | 2013-08-07 |
Family
ID=48883147
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013101319077A Pending CN103235096A (en) | 2013-04-16 | 2013-04-16 | Sewage water quality detection method and apparatus |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103235096A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105372995A (en) * | 2015-12-17 | 2016-03-02 | 镇江市高等专科学校 | Measurement and control method for sewage disposal system |
CN105487526A (en) * | 2016-01-04 | 2016-04-13 | 华南理工大学 | FastRVM (fast relevance vector machine) wastewater treatment fault diagnosis method |
CN105675838A (en) * | 2016-04-15 | 2016-06-15 | 北京工业大学 | Intelligent A<2>/O process effluent total phosphorus detection method based on data driving |
CN106021884A (en) * | 2016-05-12 | 2016-10-12 | 武汉理工大学 | Neural network principle-based method for predicting effluent concentration of subsurface flow wetland |
CN103942457B (en) * | 2014-05-09 | 2017-04-12 | 浙江师范大学 | Water quality parameter time series prediction method based on relevance vector machine regression |
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 |
CN107977724A (en) * | 2016-10-21 | 2018-05-01 | 复凌科技(上海)有限公司 | A kind of water quality hard measurement Forecasting Methodology of permanganate index |
CN108074011A (en) * | 2017-11-02 | 2018-05-25 | 广州工程技术职业学院 | The monitoring method and system of a kind of sludge discharge |
CN109534486A (en) * | 2018-12-21 | 2019-03-29 | 哈尔滨工程大学 | A kind of ship sewage treatment control forecasting system and prediction technique based on Stochastic Recursive wavelet neural network |
CN110132629A (en) * | 2019-06-06 | 2019-08-16 | 浙江清华长三角研究院 | A method of utilizing SVM prediction rural domestic sewage treatment facility operation validity |
CN111207791A (en) * | 2020-01-08 | 2020-05-29 | 南宁市勘察测绘地理信息院 | Sewage parameter acquisition equipment and sewage well monitoring system |
CN112694171A (en) * | 2020-12-22 | 2021-04-23 | 上海上实龙创智能科技股份有限公司 | Aeration control method and device for sewage treatment, electronic equipment and storage medium |
CN112784476A (en) * | 2020-12-30 | 2021-05-11 | 浙江大学 | Soft measurement method and device for ammonia nitrogen in effluent of different process type agricultural sewage treatment facilities |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101718774A (en) * | 2009-11-09 | 2010-06-02 | 东南大学 | Diagnostic method for validity of online collected water quality data |
CN202075272U (en) * | 2011-03-30 | 2011-12-14 | 北京中拓百川投资有限公司 | Soft measurement system applied to sewage treatment |
-
2013
- 2013-04-16 CN CN2013101319077A patent/CN103235096A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101718774A (en) * | 2009-11-09 | 2010-06-02 | 东南大学 | Diagnostic method for validity of online collected water quality data |
CN202075272U (en) * | 2011-03-30 | 2011-12-14 | 北京中拓百川投资有限公司 | Soft measurement system applied to sewage treatment |
Non-Patent Citations (3)
Title |
---|
宋贤民等: "《基于支持向量机的出水TP软测量》", 《科技信息》 * |
曾甜玲等: "《相关向量机在污水硝氮检测中的应用》", 《光谱学与光谱分析》 * |
陈佳等: "《基于贝叶斯学习的关联向量机及其在软测量中的应用》", 《华东理工大学学报(自然科学版)》 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942457B (en) * | 2014-05-09 | 2017-04-12 | 浙江师范大学 | Water quality parameter time series prediction method based on relevance vector machine regression |
CN105372995A (en) * | 2015-12-17 | 2016-03-02 | 镇江市高等专科学校 | Measurement and control method for sewage disposal system |
CN105487526B (en) * | 2016-01-04 | 2019-04-09 | 华南理工大学 | A kind of Fast RVM sewage treatment method for diagnosing faults |
CN105487526A (en) * | 2016-01-04 | 2016-04-13 | 华南理工大学 | FastRVM (fast relevance vector machine) wastewater treatment fault diagnosis method |
CN105675838A (en) * | 2016-04-15 | 2016-06-15 | 北京工业大学 | Intelligent A<2>/O process effluent total phosphorus detection method based on data driving |
CN106021884A (en) * | 2016-05-12 | 2016-10-12 | 武汉理工大学 | Neural network principle-based method for predicting effluent concentration of subsurface flow wetland |
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 |
CN107977724A (en) * | 2016-10-21 | 2018-05-01 | 复凌科技(上海)有限公司 | A kind of water quality hard measurement Forecasting Methodology of permanganate index |
CN108074011A (en) * | 2017-11-02 | 2018-05-25 | 广州工程技术职业学院 | The monitoring method and system of a kind of sludge discharge |
CN109534486A (en) * | 2018-12-21 | 2019-03-29 | 哈尔滨工程大学 | A kind of ship sewage treatment control forecasting system and prediction technique based on Stochastic Recursive wavelet neural network |
CN110132629A (en) * | 2019-06-06 | 2019-08-16 | 浙江清华长三角研究院 | A method of utilizing SVM prediction rural domestic sewage treatment facility operation validity |
CN110132629B (en) * | 2019-06-06 | 2020-03-10 | 浙江清华长三角研究院 | Method for predicting operation effectiveness of rural domestic sewage treatment facility by using support vector machine |
CN111207791A (en) * | 2020-01-08 | 2020-05-29 | 南宁市勘察测绘地理信息院 | Sewage parameter acquisition equipment and sewage well monitoring system |
CN112694171A (en) * | 2020-12-22 | 2021-04-23 | 上海上实龙创智能科技股份有限公司 | Aeration control method and device for sewage treatment, electronic equipment and storage medium |
CN112694171B (en) * | 2020-12-22 | 2022-08-05 | 上海上实龙创智能科技股份有限公司 | Aeration control method and device for sewage treatment, electronic equipment and storage medium |
CN112784476A (en) * | 2020-12-30 | 2021-05-11 | 浙江大学 | Soft measurement method and device for ammonia nitrogen in effluent of different process type agricultural sewage treatment facilities |
CN112784476B (en) * | 2020-12-30 | 2022-07-19 | 浙江大学 | Soft measurement method and device for ammonia nitrogen in effluent of different process type agricultural sewage treatment facilities |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103235096A (en) | Sewage water quality detection method and apparatus | |
Sharafati et al. | The potential of new ensemble machine learning models for effluent quality parameters prediction and related uncertainty | |
Zhou et al. | Industrial structural upgrading and spatial optimization based on water environment carrying capacity | |
Liu et al. | An integrated system dynamics model developed for managing lake water quality at the watershed scale | |
CN103810537B (en) | A kind of Regional environment risk appraisal procedure based on water quality model | |
Regmi et al. | The future of WRRF modelling–outlook and challenges | |
CN111191191B (en) | Construction method of combined model for accurately predicting deformation effect of concrete dam | |
CN109492265B (en) | Wastewater effluent index prediction method based on dynamic nonlinear PLS soft measurement method | |
CN108920427B (en) | Dynamic estimation method for time-varying parameters of monthly scale hydrological model | |
CN110320335B (en) | Multi-element robust soft measurement method for effluent quality index of sewage treatment | |
CN102262147A (en) | Soft measurement method and system for effluent chemical oxygen demand (COD) of waste water treatment system | |
CN103942457A (en) | Water quality parameter time series prediction method based on relevance vector machine regression | |
Noori et al. | Active and online prediction of BOD 5 in river systems using reduced-order support vector machine | |
CN107895100B (en) | Drainage basin water quality comprehensive evaluation method and system | |
CN111553468A (en) | Method for accurately predicting effluent quality of sewage treatment plant | |
CN108595892A (en) | Soft-measuring modeling method based on time difference model | |
CN110132629B (en) | Method for predicting operation effectiveness of rural domestic sewage treatment facility by using support vector machine | |
Han et al. | Prediction of sludge bulking using the knowledge-leverage-based fuzzy neural network | |
Wang et al. | A full-view management method based on artificial neural networks for energy and material-savings in wastewater treatment plants | |
CN104914227B (en) | Sewage quality flexible measurement method based on many gaussian kernel self-optimizing Method Using Relevance Vector Machine | |
CN112183935A (en) | River water quality comprehensive evaluation method and system | |
CN111311026A (en) | Runoff nonlinear prediction method considering data characteristics, model and correction | |
Han et al. | A sludge volume index (SVI) model based on the multivariate local quadratic polynomial regression method | |
CN104063609B (en) | Method of assisting in judging pollution source monitoring data validity by utilizing neural network | |
CN117114190B (en) | River runoff prediction method and device based on mixed deep learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130807 |