CN102737288A - Radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality - Google Patents
Radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality Download PDFInfo
- Publication number
- CN102737288A CN102737288A CN2012102048383A CN201210204838A CN102737288A CN 102737288 A CN102737288 A CN 102737288A CN 2012102048383 A CN2012102048383 A CN 2012102048383A CN 201210204838 A CN201210204838 A CN 201210204838A CN 102737288 A CN102737288 A CN 102737288A
- Authority
- CN
- China
- Prior art keywords
- water quality
- neural network
- rbf neural
- data
- sample
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality. The method comprises the following steps of: first storing the data of each monitoring station into a database of a local server by using the remote transmission of an online water quality monitoring instrument; then performing normalization processing on a water quality sample sequence, calculating an autocorrelation coefficient to determine an input variable of an RBF neural network, and converting sample data into a standard dynamic sequence data format trained and predicted by the RBF neutral network; next searching for and determining an optimal value of a spreading coefficient spread of the RBF neural network by utilizing a differential evolution algorithm and taking a relative standard error as a target function to obtain an optimal prediction model; and finally sampling water quality data in real time, performing multi-step prediction by using the obtained optimal prediction model and adopting a single-point iteration method, and evaluating a water quality prediction result to realize an early warning function. The water quality can be intelligently warned.
Description
Technical field
The present invention relates to a kind of water quality prediction method, relate in particular to a kind of water quality multistep forecasting method based on the self-optimizing of RBF neural network parameter.
Background technology
Drinking water safety carries out the real-time estimate analysis concerning national economy to water quality, can effectively control and reduces the harm that water quality deterioration causes, and reaches the effective cognition to water quality deterioration, the target of control.In addition, the water quality multi-step prediction can win the more emergency response time for water factory timely and effectively.
Main both at home and abroad at present research single step water quality prediction method; The single step water quality prediction comprises modeling based on mechanism, based on two aspects of modeling of intelligence; Because Water Environment System is complicated and changeable; Its details mechanism can not be understood fully, and the latter has the good information processing power to the indefinite high dimensional nonlinear of mechanism system, in water quality prediction, has obtained in recent years using widely.Neural network has powerful self-organization, self study, parallel processing information and non-linear fault-tolerant ability; It is the focus of studying both at home and abroad; But only know the predicted value of next time point; Can not be for the water quality early-warning system provides the sufficiently long emergency processing time, so be necessary to introduce the method for multi-step prediction.
Summary of the invention
The objective of the invention is to propose a kind of water quality multistep forecasting method, not only satisfied the requirement of multi-step prediction precision, also realized intelligentized water quality early-warning well based on the self-optimizing of RBF neural network parameter.
The objective of the invention is to realize through following technical scheme: a kind of water quality multistep forecasting method based on the self-optimizing of RBF neural network parameter comprises the steps:
(1) through the remote transmission of online water quality monitoring instrument, the data of each monitoring station are deposited in the database of home server, form the water quality sample sequence, be convenient to water quality parameter is analyzed; The said good and bad water quality parameter of evaluating water quality that is used for comprises: pH, conductivity, turbidity, dissolved oxygen DO, temperature, ammonia nitrogen, chlorine residue and permanganate indices etc.;
(2) the water quality sample sequence being carried out normalization handles; Confirm the input variable of RBF neural network through calculating coefficient of autocorrelation; Sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction: the water quality sample sequence is for
; According to
data being carried out normalization handles; Wherein,
and
is respectively maximal value and minimum value in
; The calculating of coefficient of autocorrelation is shown below:
In the formula;
is autocorrelation function;
is coefficient of autocorrelation to be asked;
is the length of water quality time series
; N is a water quality sample ordinal number, and
is the water quality sample.
The scope of coefficient of autocorrelation is
; Set correlation coefficient threshold
; During as
;
is so the input vector of RBF neural network
is configured to
.Confirm the length
of training sample according to forecasting problem actual demand and sample data characteristics; Obtain the inputoutput pair
of network, sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction;
(3) with error criterion poor
as objective function; Utilize the differential evolution algorithm search to confirm the optimal value of RBF neural network spreading coefficient spread; Obtain optimum forecast model: utilize the newrbe function to set up the RBF neural network, utilize the differential evolution algorithm search to confirm the optimal value of RBF neural network spreading coefficient spread; The differential evolution algorithm is a kind of global optimization method based on the population evolution of floating-point code; The minimum problem of promptly finding the solution
; Wherein
is spread, objective function
be error criterion poor
expression as shown in the formula:
Wherein
is original water quality sequence;
is the water quality sequence of prediction, and
is training sample length;
(4) real-time sampling water quality data with the optimum prediction model that step 3 obtains, adopt the method for single-point iteration to realize multi-step prediction, and evaluating water quality predicts the outcome the effect of realization early warning; If exceed normal range then warning immediately.
The invention has the beneficial effects as follows; The present invention proposes to confirm through coefficient of autocorrelation the single-point process of iteration of RBF neural network input variable; Improve the effect of multi-step prediction, utilized the optimal value of differential evolution algorithm search spreading coefficient spread simultaneously, realized efficient intelligent robotization early warning.
Description of drawings
Fig. 1 utilizes coefficient of autocorrelation to confirm the method figure of RBF neural network input;
Fig. 2 is the sequence coefficient of autocorrelation figure of water quality parameter (is example with pH);
Fig. 3 utilizes the differential evolution algorithm to carry out the schematic diagram that the spread parameter is selected;
Fig. 4 is the differential evolution algorithm search figure as a result of spread;
Fig. 5 is the figure that predicts the outcome in 4 steps of water quality parameter (is example with pH);
Fig. 6 is 4 step predicated error curve maps of water quality parameter (is example with pH).
Embodiment
The step of water quality multistep forecasting method that the present invention is based on the self-optimizing of RBF neural network parameter is following:
1, through the remote transmission of online water quality monitoring instrument, the data of each monitoring station are deposited in the database of home server, form the water quality sample sequence, be convenient to water quality parameter is analyzed.
Being used for the good and bad water quality parameter of evaluating water quality mainly contains conventional Wucan number (pH, conductivity, turbidity, dissolved oxygen DO, temperature), ammonia nitrogen, chlorine residue, permanganate indices etc.; The different quality parameter sensors of various places monitoring station was whenever surveyed once at a distance from 15 minutes; Every survey one secondary data just is deposited into data in the database server; Set up the table in the database according to different websites and different quality parameter; This database is used for preserving all raw data from each monitoring station monitoring, is convenient to water quality parameter is carried out analyzing and processing.
2, the water quality sample sequence is carried out normalization and handle, confirm the input variable of RBF neural network, sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction through calculating coefficient of autocorrelation.
If the water quality sample sequence is
; The neural network of building in order to guarantee has enough input sensitivitys and good fitness for sample; According to
data are carried out normalization and handle, wherein maximal value and the minimum value in
and
difference
.Because coefficient of autocorrelation can be weighed the degree of correlation between the data on the time series different time points,, confirm the input vector of RBF neural network so adopt the autocorrelation analysis method to analyze and the most closely-related time series historical data of predicted value.Because the length of sequence is limited in the reality, so the calculating of coefficient of autocorrelation is suc as formula shown in (1):
In the formula (1);
is autocorrelation function;
is coefficient of autocorrelation to be asked;
is the length of water quality time series
; N is a water quality sample ordinal number, and
is the water quality sample.
The scope of coefficient of autocorrelation is
; Set correlation coefficient threshold
; During as
;
is so the input vector of RBF neural network
is configured to
.Confirm the length
of training sample according to forecasting problem actual demand and sample data characteristics; Obtain the inputoutput pair
of network, sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction.
3, with error criterion poor
as objective function; Utilize the differential evolution algorithm search to confirm the optimal value of RBF neural network spreading coefficient spread, obtain optimum forecast model.
Utilize the newrbe function to set up the RBF neural network; This function has automatic adjustment hidden layer unit number, and to make error be 0 function, but in RBF network operations process, need manually to change the degree of fitting that its spreading coefficient spread value is adjusted whole network; Manual adjustment often relatively blindly; Inefficiency not only, and differ and find optimum solution surely, so utilize the differential evolution algorithm search to confirm the optimal value of RBF neural network spreading coefficient spread.The differential evolution algorithm is a kind of global optimization method based on the population evolution of floating-point code; The minimum problem of promptly finding the solution
; Wherein
is spread, and objective function
is that error criterion poor
is represented suc as formula (2):
（2）
Wherein
is original water quality sequence;
is the water quality sequence of prediction, and
is training sample length.
Suppose that each has
individuals in generation
in the differential evolution algorithm;
representative is shown
, and wherein
is
individuals in
generation.
The concrete steps of selecting based on the spread parameter of differential evolution algorithm are following:
1) in parameter space, produces initial value at random for
individuals; With each group parameter training RBF neural network, obtain corresponding error criterion poor
then;
2) establish
;
3) utilize formula
to make a variation for current individuality
; In the middle of producing individual
, in the formula
;
4) middle individual
and the current individuality
that variation are obtained are hybridized; Obtain the candidate individual
of current individuality; If satisfied
goes up equally distributed random number and is not more than the hybridization factor; Then
, otherwise
;
5) it is poor to utilize new argument
training RBF neural network to obtain its error criterion; if
; Then
, otherwise
;
6)
; if
; Then be back to step 3), otherwise jump to step 7);
7) it is poor to find out minimum error criterion, and to note corresponding individuality be the spread value;
8) if the evolution number of times greater than 100, then export optimum spread value, otherwise be back to step 3).
4, real-time sampling water quality data with the optimum prediction model that step (3) obtains, adopt the method for single-point iteration to realize multi-step prediction, and evaluating water quality predicts the outcome the effect of realization early warning.
Sensor in real time sampling water quality data.In water quality early-warning, for advanced processing alert and definite emergency preplan, need obtain the predicted value of following water quality parameter in advance, adopt the method for single-point iteration to realize multi-step prediction, can shift to an earlier date the generation of foreseeing alert for more time.The method of single-point iteration specifically is embodied as; Water quality parameter predicted value
constantly that the optimum RBF neural network that adopts the front to build obtains
through
; Replace actual value to obtain the predicted value of any down with predicted value as the element of prediction input vector; Adopt this alternative manner can obtain the predicted value in following a plurality of moment; Along with the increase of predicting step number can make precision of prediction reduce; But be implied with the ductwork water quality parameter of some cycles property through the analysis of coefficient of autocorrelation method; Can be the numerical value of the special time of being separated by (Cycle Length) also as input variable; Avoid sample to replace the precision of prediction that actual value caused to descend, improve the effect of multi-step prediction by too much predicted value.With reference to GB5749-2006 " drinking water sanitary standard " regulation, the water quality prediction parameter is estimated in real time, if exceed normal range then report to the police immediately,, avoid the deterioration of alert so that relevant departments in time take appropriate measures.
Embodiment:
As shown in Figure 1; Get the somewhere pipe network water water quality parameter of on November 10th, 2011 to on-line monitoring on the 19th in November (is example with pH), real-time watch device monitoring in 15 minutes once has 960 data; Get 768 preceding 8 days data as the prediction training set, 192 back 2 days data are as test set.If time series is
; The neural network of building in order to guarantee for sample have enough input sensitivitys and with good fitness; According to
training set is carried out normalization and handle, wherein
and
is respectively the maximal value and the minimum value of training set.Calculate the coefficient of autocorrelation
of pH value training set sequence then; The threshold value
of setting related coefficient is 0.9; As shown in Figure 2; Coefficient of autocorrelation
during as
; Therefore the input variable of RBF neural network can be configured to
; The inputoutput pair of training set network is
, so just sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction.
As shown in Figure 3;
individuals of differential evolution algorithm makes a variation in parameter space, hybridizes, evolves; Every evolution generation is just with the new spread value training RBF neural network that produces; And it is poor to calculate the pairing error criterion of each parameter, judges whether to satisfy stop condition then, if the evolution number of times is greater than 100; Then export optimum spread value, otherwise the new parameter of generation that continues to evolve.As shown in Figure 4; Limited the locally optimal solution of differential evolution algorithm in can the quick lock in region of search, and finally confirm globally optimal solution just searched optimum solution after 4 times evolving; The spread optimal value that searches is 192, and minimum error criterion difference is 0.0079.
Water quality parameter predicted value
constantly that the optimum RBF neural network that adopts the front to build obtains
through
; Replace element acquisition
constantly the predicted value of actual value with predicted value as the prediction input vector; So iterate can obtain
constantly predicted value, wherein
is prediction step.Here choose
; Promptly predict the pH value after 1 hour; Final multi-step prediction result and graph of errors are respectively like Fig. 5 and shown in Figure 6; Can be known that by test result the water quality multistep forecasting method based on the self-optimizing of RBF neural network parameter of the present invention can be realized prediction well, the predicated error of 82% data is less than 0.03 in the prediction of 4 steps; Can obtain the following predicted value of water quality parameter constantly in advance through the method; With reference to GB5749-2006 " drinking water sanitary standard " regulation, the limit value of pH is to be not less than 6.5 and be not more than 8.5, and the water quality prediction parameter is estimated in real time; If report to the police not in normal range then immediately; So that relevant departments in time take appropriate measures, avoid the deterioration of alert to reach definite emergency preplan by the advanced processing alert, realize the effect of water quality early-warning.
Claims (1)
1. the water quality multistep forecasting method based on the self-optimizing of RBF neural network parameter is characterized in that, comprises the steps:
(1) through the remote transmission of online water quality monitoring instrument, the data of each monitoring station are deposited in the database of home server, form the water quality sample sequence, be convenient to water quality parameter is analyzed; The said good and bad water quality parameter of evaluating water quality that is used for comprises: pH, conductivity, turbidity, dissolved oxygen DO, temperature, ammonia nitrogen, chlorine residue and permanganate indices etc.;
(2) the water quality sample sequence being carried out normalization handles; Confirm the input variable of RBF neural network through calculating coefficient of autocorrelation; Sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction: the water quality sample sequence is for
; According to
data being carried out normalization handles; Wherein,
and
is respectively maximal value and minimum value in
; The calculating of coefficient of autocorrelation is shown below:
In the formula;
is autocorrelation function;
is coefficient of autocorrelation to be asked;
is the length of water quality time series
; N is a water quality sample ordinal number, and
is the water quality sample;
The scope of coefficient of autocorrelation is
; Set correlation coefficient threshold
; During as
;
is so the input vector of RBF neural network
is configured to
; Confirm the length
of training sample according to forecasting problem actual demand and sample data characteristics; Obtain the inputoutput pair
of network, sample data is converted into the standard dynamic sequence data layout of RBF neural metwork training prediction;
(3) with error criterion poor
as objective function; Utilize the differential evolution algorithm search to confirm the optimal value of RBF neural network spreading coefficient spread; Obtain optimum forecast model: utilize the newrbe function to set up the RBF neural network, utilize the differential evolution algorithm search to confirm the optimal value of RBF neural network spreading coefficient spread; The differential evolution algorithm is a kind of global optimization method based on the population evolution of floating-point code; The minimum problem of promptly finding the solution
; Wherein
is spread, objective function
be error criterion poor
expression as shown in the formula:
Wherein
is original water quality sequence;
is the water quality sequence of prediction, and
is training sample length;
(4) real-time sampling water quality data with the optimum prediction model that step 3 obtains, adopt the method for single-point iteration to realize multi-step prediction, and evaluating water quality predicts the outcome the effect of realization early warning; If exceed normal range then warning immediately.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012102048383A CN102737288A (en) | 2012-06-20 | 2012-06-20 | Radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012102048383A CN102737288A (en) | 2012-06-20 | 2012-06-20 | Radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102737288A true CN102737288A (en) | 2012-10-17 |
Family
ID=46992744
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012102048383A Pending CN102737288A (en) | 2012-06-20 | 2012-06-20 | Radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102737288A (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103018181A (en) * | 2012-12-14 | 2013-04-03 | 江苏大学 | Soft measurement method based on correlation analysis and ELM neural network |
| CN103136601A (en) * | 2013-03-14 | 2013-06-05 | 天津大学 | Regeneration water factory effluent residual chorine risk prediction method |
| CN103279671A (en) * | 2013-06-03 | 2013-09-04 | 南京大学 | Urban water disaster risk prediction method based on RBF (radial basis function) neural network-cloud model |
| CN103605909A (en) * | 2013-12-09 | 2014-02-26 | 重庆绿色智能技术研究院 | Water quality predication method based on grey theory and support vector machine |
| CN104007243A (en) * | 2014-01-13 | 2014-08-27 | 上海市长宁区卫生局卫生监督所 | Method for evaluating water quality of swimming pool by using back-propagation neural network model |
| CN103136601B (en) * | 2013-03-14 | 2016-11-30 | 天津大学 | Regeneration water factory effluent residual chorine risk prediction |
| CN106407691A (en) * | 2016-09-28 | 2017-02-15 | 中国地质大学（武汉） | Red tide biomass prediction method and device |
| CN106645620A (en) * | 2016-12-23 | 2017-05-10 | 江西理工大学 | Hybrid differential evolution based BOD soft measurement method for farmland water body in rare earth mine lot |
| CN106991491A (en) * | 2017-02-21 | 2017-07-28 | 南京邮电大学 | A kind of environment pollution monitoring system of the RBF neural based on genetic algorithm optimization |
| CN107300855A (en) * | 2017-06-27 | 2017-10-27 | 重庆科技学院 | A kind of shared direct drinking fountain water quality ensures management control method and system |
| CN107643335A (en) * | 2016-07-20 | 2018-01-30 | 复凌科技（上海）有限公司 | A kind of method for detecting water environment |
| CN107688871A (en) * | 2017-08-18 | 2018-02-13 | 中国农业大学 | A kind of water quality prediction method and device |
| CN108009636A (en) * | 2017-11-16 | 2018-05-08 | 华南师范大学 | Deep learning ANN Evolutionary method, apparatus, medium and computer equipment |
| CN108027885A (en) * | 2015-06-05 | 2018-05-11 | 渊慧科技有限公司 | Space transformer module |
| CN107229970B (en) * | 2017-06-27 | 2018-08-17 | 重庆科技学院 | The adaptive dynamic self study on-line monitoring system of shared direct drinking water quality |
| CN108665100A (en) * | 2018-05-09 | 2018-10-16 | 中国农业大学 | A kind of water quality prediction technique, system and device |
| CN108764520A (en) * | 2018-04-11 | 2018-11-06 | 杭州电子科技大学 | A kind of water quality parameter prediction technique based on multilayer circulation neural network and D-S evidence theory |
| CN108846423A (en) * | 2018-05-29 | 2018-11-20 | 中国农业大学 | Water quality prediction method and system |
| CN109032047A (en) * | 2018-10-12 | 2018-12-18 | 四川省建筑设计研究院 | A kind of instrument system of pipe gallery safety monitoring |
| CN109901542A (en) * | 2019-04-10 | 2019-06-18 | 上海上实龙创智慧能源科技股份有限公司 | A kind of sewage water treatment method based on Internet of Things, cloud platform and system |
| CN110852515A (en) * | 2019-11-15 | 2020-02-28 | 北京工业大学 | Water quality index prediction method based on mixed long-time and short-time memory neural network |
| CN111340176A (en) * | 2018-12-19 | 2020-06-26 | 富泰华工业（深圳）有限公司 | Neural network training method and device and computer storage medium |
| CN114660248A (en) * | 2020-12-22 | 2022-06-24 | 中国石油化工股份有限公司 | COD early warning method and device based on multi-step prediction strategy |
| CN115983506A (en) * | 2023-03-20 | 2023-04-18 | 华东交通大学 | Water quality early warning method and system and readable storage medium |
| CN116842856A (en) * | 2023-09-04 | 2023-10-03 | 长春工业大学 | Industrial process optimization method based on deep reinforcement learning |
| CN117309067A (en) * | 2023-11-30 | 2023-12-29 | 长春职业技术学院 | Water resource real-time monitoring method, system and electronic equipment |
| CN117524372A (en) * | 2023-11-16 | 2024-02-06 | 浙江大学 | Micro-channel metamaterial design method based on genetic algorithm, electronic equipment and medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070215556A1 (en) * | 2006-03-20 | 2007-09-20 | Sensis Corporation | System for detection and prediction of water nitrification |
-
2012
- 2012-06-20 CN CN2012102048383A patent/CN102737288A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070215556A1 (en) * | 2006-03-20 | 2007-09-20 | Sensis Corporation | System for detection and prediction of water nitrification |
Non-Patent Citations (2)
| Title |
|---|
| 路剑飞、陈子燊: "珠江口磨刀门水道盐度多步预测研究", 《水文》, vol. 30, no. 5, 25 October 2010 (2010-10-25), pages 69 - 74 * |
| 黄浙丰: "基于时序神经网络的藻类水华预测模型研究", 《中国优秀硕士学位论文全文数据库信息科技辑》, 15 July 2011 (2011-07-15) * |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103018181A (en) * | 2012-12-14 | 2013-04-03 | 江苏大学 | Soft measurement method based on correlation analysis and ELM neural network |
| CN103136601A (en) * | 2013-03-14 | 2013-06-05 | 天津大学 | Regeneration water factory effluent residual chorine risk prediction method |
| CN103136601B (en) * | 2013-03-14 | 2016-11-30 | 天津大学 | Regeneration water factory effluent residual chorine risk prediction |
| CN103279671A (en) * | 2013-06-03 | 2013-09-04 | 南京大学 | Urban water disaster risk prediction method based on RBF (radial basis function) neural network-cloud model |
| CN103279671B (en) * | 2013-06-03 | 2016-08-17 | 南京大学 | Municipal water disaster Risk Forecast Method based on RBF neural-cloud model |
| CN103605909A (en) * | 2013-12-09 | 2014-02-26 | 重庆绿色智能技术研究院 | Water quality predication method based on grey theory and support vector machine |
| CN104007243A (en) * | 2014-01-13 | 2014-08-27 | 上海市长宁区卫生局卫生监督所 | Method for evaluating water quality of swimming pool by using back-propagation neural network model |
| CN104007243B (en) * | 2014-01-13 | 2016-01-06 | 上海市长宁区卫生局卫生监督所 | Application back propagation artificial neural network model evaluates the method for swimming pool water |
| CN108027885A (en) * | 2015-06-05 | 2018-05-11 | 渊慧科技有限公司 | Space transformer module |
| CN108027885B (en) * | 2015-06-05 | 2022-07-01 | 渊慧科技有限公司 | Space transformer module |
| CN107643335A (en) * | 2016-07-20 | 2018-01-30 | 复凌科技（上海）有限公司 | A kind of method for detecting water environment |
| CN106407691B (en) * | 2016-09-28 | 2019-11-22 | 中国地质大学（武汉） | A kind of red tide plankton amount prediction technique and device |
| CN106407691A (en) * | 2016-09-28 | 2017-02-15 | 中国地质大学（武汉） | Red tide biomass prediction method and device |
| CN106645620A (en) * | 2016-12-23 | 2017-05-10 | 江西理工大学 | Hybrid differential evolution based BOD soft measurement method for farmland water body in rare earth mine lot |
| CN106645620B (en) * | 2016-12-23 | 2018-09-28 | 江西理工大学 | Using the rare-earth mining area farmland water body BOD flexible measurement methods of mixing differential evolution |
| CN106991491A (en) * | 2017-02-21 | 2017-07-28 | 南京邮电大学 | A kind of environment pollution monitoring system of the RBF neural based on genetic algorithm optimization |
| CN106991491B (en) * | 2017-02-21 | 2020-11-13 | 南京邮电大学 | RBF neural network's environmental pollution monitoring system based on genetic algorithm optimization |
| CN107300855A (en) * | 2017-06-27 | 2017-10-27 | 重庆科技学院 | A kind of shared direct drinking fountain water quality ensures management control method and system |
| CN107229970B (en) * | 2017-06-27 | 2018-08-17 | 重庆科技学院 | The adaptive dynamic self study on-line monitoring system of shared direct drinking water quality |
| CN107688871A (en) * | 2017-08-18 | 2018-02-13 | 中国农业大学 | A kind of water quality prediction method and device |
| CN107688871B (en) * | 2017-08-18 | 2020-08-21 | 中国农业大学 | Water quality prediction method and device |
| CN108009636B (en) * | 2017-11-16 | 2021-12-07 | 华南师范大学 | Deep learning neural network evolution method, device, medium and computer equipment |
| CN108009636A (en) * | 2017-11-16 | 2018-05-08 | 华南师范大学 | Deep learning ANN Evolutionary method, apparatus, medium and computer equipment |
| CN108764520B (en) * | 2018-04-11 | 2021-11-16 | 杭州电子科技大学 | Water quality parameter prediction method based on multilayer cyclic neural network and D-S evidence theory |
| CN108764520A (en) * | 2018-04-11 | 2018-11-06 | 杭州电子科技大学 | A kind of water quality parameter prediction technique based on multilayer circulation neural network and D-S evidence theory |
| CN108665100A (en) * | 2018-05-09 | 2018-10-16 | 中国农业大学 | A kind of water quality prediction technique, system and device |
| CN108846423A (en) * | 2018-05-29 | 2018-11-20 | 中国农业大学 | Water quality prediction method and system |
| CN109032047A (en) * | 2018-10-12 | 2018-12-18 | 四川省建筑设计研究院 | A kind of instrument system of pipe gallery safety monitoring |
| CN111340176A (en) * | 2018-12-19 | 2020-06-26 | 富泰华工业（深圳）有限公司 | Neural network training method and device and computer storage medium |
| CN109901542B (en) * | 2019-04-10 | 2020-08-07 | 上海上实龙创智能科技股份有限公司 | Sewage treatment method based on Internet of things, cloud platform and system |
| CN109901542A (en) * | 2019-04-10 | 2019-06-18 | 上海上实龙创智慧能源科技股份有限公司 | A kind of sewage water treatment method based on Internet of Things, cloud platform and system |
| CN110852515B (en) * | 2019-11-15 | 2022-05-06 | 北京工业大学 | Water quality index prediction method based on mixed long-time and short-time memory neural network |
| CN110852515A (en) * | 2019-11-15 | 2020-02-28 | 北京工业大学 | Water quality index prediction method based on mixed long-time and short-time memory neural network |
| CN114660248A (en) * | 2020-12-22 | 2022-06-24 | 中国石油化工股份有限公司 | COD early warning method and device based on multi-step prediction strategy |
| CN115983506A (en) * | 2023-03-20 | 2023-04-18 | 华东交通大学 | Water quality early warning method and system and readable storage medium |
| CN116842856A (en) * | 2023-09-04 | 2023-10-03 | 长春工业大学 | Industrial process optimization method based on deep reinforcement learning |
| CN116842856B (en) * | 2023-09-04 | 2023-11-14 | 长春工业大学 | Industrial process optimization method based on deep reinforcement learning |
| CN117524372A (en) * | 2023-11-16 | 2024-02-06 | 浙江大学 | Micro-channel metamaterial design method based on genetic algorithm, electronic equipment and medium |
| CN117524372B (en) * | 2023-11-16 | 2024-05-17 | 浙江大学 | Micro-channel metamaterial design method based on genetic algorithm, electronic equipment and medium |
| CN117309067A (en) * | 2023-11-30 | 2023-12-29 | 长春职业技术学院 | Water resource real-time monitoring method, system and electronic equipment |
| CN117309067B (en) * | 2023-11-30 | 2024-02-09 | 长春职业技术学院 | Water resource real-time monitoring method, system and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102737288A (en) | Radial basis function (RBF) neural network parameter self-optimizing-based multi-step prediction method for water quality | |
| Cho et al. | Prediction of contamination potential of groundwater arsenic in Cambodia, Laos, and Thailand using artificial neural network | |
| WO2019237440A1 (en) | Quantile probabilistic short-term power load prediction integration method | |
| CN108171259A (en) | A kind of wisdom tip water quality monitoring system and method based on NB-IoT | |
| CN106714336A (en) | Wireless sensor network temperature monitoring method based on improved Kriging algorithm | |
| KR101703981B1 (en) | Control methods of sewage processing facilities using rainfall forecasting | |
| Wang et al. | Supply-demand risk assessment and multi-scenario simulation of regional water-energy-food nexus: A case study of the Beijing-Tianjin-Hebei region | |
| CN109884892A (en) | Process industry system prediction model based on crosscorrelation time lag grey correlation analysis | |
| CN107293118B (en) | Short-time prediction method for traffic speed dynamic interval | |
| CN111191191A (en) | Construction method of combined model for accurately predicting deformation effect of concrete dam | |
| Kueh et al. | FORECASTING LONG TERM PRECIPITATION USING CUCKOO SEARCH OPTIMIZATION NEURAL NETWORK MODELS. | |
| CN102509173A (en) | Markov chain based method for accurately forecasting power system loads | |
| CN114399073A (en) | Ocean surface temperature field prediction method based on deep learning | |
| CN117522632A (en) | Water quality index prediction method based on deep learning | |
| CN115470850A (en) | Water quality abnormal event recognition early warning method based on pipe network water quality time-space data | |
| Bashar et al. | Investigation of factors affecting rural drinking water consumption using intelligent hybrid models | |
| CN109523077B (en) | Wind power prediction method | |
| CN116821695B (en) | Semi-supervised neural network soft measurement modeling method | |
| Choi et al. | Robust imputation method with context-aware voting ensemble model for management of water-quality data | |
| Kou et al. | Prediction intervals for wind power forecasting: using sparse warped Gaussian process | |
| Panjapornpon et al. | Energy efficiency and savings analysis with multirate sampling for petrochemical process using convolutional neural network-based transfer learning | |
| CN110287652A (en) | Method based on haloform concentration in radial neural network and grey relational grade analysis modeling and forecasting water system | |
| CN114386672B (en) | Environment big data Internet of things intelligent detection system | |
| CN107170235B (en) | Traffic flow prediction time sequence method based on depth feature extraction network | |
| CN105844077B (en) | A kind of testability growth test method based on timely correction strategy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20121017 |