CN108872508A - A kind of eutrophy quality evaluation method of GA-BP optimization TSFNN - Google Patents
A kind of eutrophy quality evaluation method of GA-BP optimization TSFNN Download PDFInfo
- Publication number
- CN108872508A CN108872508A CN201810434779.6A CN201810434779A CN108872508A CN 108872508 A CN108872508 A CN 108872508A CN 201810434779 A CN201810434779 A CN 201810434779A CN 108872508 A CN108872508 A CN 108872508A
- Authority
- CN
- China
- Prior art keywords
- neural network
- data
- optimization
- groups
- training
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 20
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 64
- 238000013528 artificial neural network Methods 0.000 claims abstract description 61
- 230000006978 adaptation Effects 0.000 claims abstract description 13
- 238000011156 evaluation Methods 0.000 claims abstract description 9
- 238000005259 measurement Methods 0.000 claims abstract description 6
- 239000003643 water by type Substances 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 34
- 238000004422 calculation algorithm Methods 0.000 claims description 26
- 230000002068 genetic effect Effects 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 13
- 230000035772 mutation Effects 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 6
- 238000001303 quality assessment method Methods 0.000 claims description 5
- 238000004088 simulation Methods 0.000 claims description 4
- 241001269238 Data Species 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000000977 initiatory effect Effects 0.000 claims description 3
- 210000004218 nerve net Anatomy 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 7
- XKMRRTOUMJRJIA-UHFFFAOYSA-N ammonia nh3 Chemical compound N.N XKMRRTOUMJRJIA-UHFFFAOYSA-N 0.000 abstract description 6
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 abstract description 6
- 239000001301 oxygen Substances 0.000 abstract description 6
- 229910052760 oxygen Inorganic materials 0.000 abstract description 6
- 238000012851 eutrophication Methods 0.000 abstract description 2
- 239000002352 surface water Substances 0.000 abstract description 2
- 230000007613 environmental effect Effects 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 230000001154 acute effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000003911 water pollution Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/18—Water
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses the eutrophy quality evaluation methods of GA-BP optimization TSFNN a kind of, belong to environmental monitoring field.The eutrophy quality evaluation method of GA-BP optimization TSFNN a kind of is by the different parameters of water quality come evaluating water quality grade.The present invention establishes the GA-BP neural network and TS fuzzy neural network of more hidden layers, using mean square error as the GA-BP neural network fit non-linear earth's surface water model of ideal adaptation angle value, PH, dissolved oxygen, turbidity, the ammonia nitrogen true value of water quality are accurately predicted by the measured value of sensor, to reach calibration measurement data, temperature drift, interference are eliminated, as the input of the foundation and TS fuzzy neural network that judge water standard grade.Because PH does not have a clear boundary in water grade evaluation criterion, TS fuzzy neural network carries out water grade evaluation according to dissolved oxygen, turbidity and the ammonia nitrogen value after calibration, waters where judging sensor whether eutrophication.
Description
Technical field
The present invention relates to water quality parameter calibrations and water grade to evaluate field, specifically designs a kind of GA-BP optimization TSFNN's
Eutrophy quality evaluation method.
Background technique
Now with rapid development of economy, the living environment of the mankind but rapidly deteriorates.Water is as Source of life, and nearly tens
It is also seriously polluted over year, especially the phenomenon that Water Eutrophication is particularly acute, the improvement of water pollution and quality of water environment
Monitoring be the departments concerned major responsibility.Connect since live water quality monitoring technology and remote sensing technology lack aquatic monitoring
Continuous property and accuracy, and electronic information technology and Internet technology are utilized, it is then real by the water quality monitoring system of node of sensor
Now to the real-time monitoring of a wide range of interior different zones surface water, evaluation result is obtained by sensing data, divides water for water body
Matter grade.Linear water quality model has complexity, variability and uncertainty.
Traditional quality evaluation method mainly has index method, Field Using Fuzzy Comprehensive Assessment.Traditional water quality monitoring system is commented
Valence method is easy to be influenced by subjective factor, it is caused to perform poor in Water Quality Evaluation.At the same time, it is traditional most
Small two multiplication algorithm can not be suitable for MIMO nonlinear systems.Artificial neural network is as a kind of information for imitating brain
Model is handled, the fitting of nonlinear model is realized by the modeling of parallel neural network, especially to multiple-input and multiple-output mould
Type has good identification effect.
Summary of the invention
Article proposes a kind of TS fuzzy neural network of modified GA-BP algorithm optimization to be applied to multi-parameter eutrophic water
In quality supervision survey and evaluation system, there is the spy of good identification effect to multiple-input and multiple-output model using artificial neural network
Point is handled by the data that algorithm acquires sensor, is eliminated many disturbing factor bring errors such as temperature drift, is connect
The data of nearly true water quality.The truthful data that system is obtained according to these predictions is allowed to come comprehensive descision water grade as ring
The reference frame of guarantor department facilitates centralized management.
The present invention is achieved by the following technical solutions:
1, a kind of eutrophy quality evaluation method of GA-BP optimization TSFNN, which is characterized in that include the following steps:
S101:National water environment quality standard (GB3838-2002) is obtained, building is based on GA-BP Optimization of Fuzzy mind
Through network model, and initial time genetic algorithm, BP neural network and TS fuzzy neural network, the initiation parameter of set algorithm;
After the completion of S101, into S102;
S102:Training number from 2000 groups of data of sensor historic data decimation of multi-parameter water quality system as model
According to the data calibration part of training pattern, using training data prediction mean square error MSE as ideal adaptation during training
Angle value obtains the initial threshold and weight of network according to optimum individual.200 groups of data are randomly selected again as test data, are used
Network performance after test training;After the completion of S102, into S103;
S103:Even distribution pattern interpolation mark at equal intervals is carried out according to water environment quality standard (GB3838-2002)
The mode of quasi- data obtains 400 groups of data, wherein I to II class, II to Group III, III to IV class and IV to V class data each 100
Group.Water quality assessment obscure portions neural network of 350 groups of data for training pattern is randomly selected, 50 groups of data are used for Test Network
Whether network has good generalization ability and judgment accuracy;After the completion of S103, into S104
S104:The BP neural network model part for obtaining fuzzy neural network model and genetic algorithm optimization collectively constitutes one
Kind optimizes the quality evaluation method of TSFNN according to the GA-BP of multi-parameter.Using 200 groups of test datas as GA-BP neural network
Input data is calibrated, and is obtained after prediction result again as the input data of TS fuzzy neural network to current sensor
The water quality in waters carries out grade evaluation where node.
2, the eutrophy quality evaluation method of GA-BP optimization TSFNN according to claim 1 a kind of, feature exist
In, in S101 include establish the more hidden layer neural networks of GA-BP, include the following steps:
S201:According to sensor measurement to water quality data dimension can determine that input node and output node are 4, use
The structure of double hidden layers, the mean square error emulated by different node in hidden layer, absolute error, absolute error percentage it is big
It is small, determine that neural network structure is 4-7-6-4.After the completion of S201.Into S202;
S202:Parameter in BP neural network is set, including:Learning rate, the number of iterations, training objective allow to miss
Difference;The parameter of GA algorithm is set, including:Group size N, genetic algebra G, crossover probability Pc, mutation probability Pm.It is completed to S202
Afterwards, into S203;
S203:Choose training data of 2000 groups of data as GA-BP neural network, and by the mean square error of training data
As ideal adaptation angle value, is intersected, mutation operation and calculates fitness value.Into circulation, until acquiring optimum individual.To
S203 is completed, into S204;
S204:The power threshold value in optimum individual is obtained, BP neural network is assigned to as initial power threshold value, utilizes 2000 groups
Training data calculates network error, and the power threshold value of network is constantly updated in training, adjusts until meeting.Simulation and prediction obtains
As a result.
3, the eutrophy quality evaluation method of GA-BP optimization TSFNN according to claim 2 a kind of, feature exist
In signified in S203 to use mean square error MSE as ideal adaptation angle value, calculation is:Above formula explanation, mean square errorBy the variance of point estimation
And deviationSquare two parts composition.
Compared with prior art, the beneficial effects of the invention are as follows:
1, the present invention is a kind of artificial neural network of more hidden layers, is had very to the nonlinear model of multiple-input and multiple-output
Good identification effect can effectively be fitted complicated multi-parameter water quality model, be predicted by the water quality data that sensor measures
To true water quality data.
2, the present invention optimizes BP neural network using genetic algorithm.Genetic algorithm is as a kind of parallel chess game optimization immediately
Algorithm filters out optimum individual in line with the principle of " survival of the fittest ".Fitness value letter in using mean square error as genetic algorithm
Number, is obtained, the precision of prediction of this method is significantly larger than BP neural network in fit procedure by Matlab simulation results
Precision of prediction, provide more efficiently data for water quality assessment.
3, it uses the individual adaptation degree using mean square error as genetic algorithm to calculate function in the present invention, establishes more hidden layers
GA-BP neural network calibrate the measurement data of sensor, eliminate temperature drift and hardware interfere bring error.It is tied according to emulation
The present invention is compared to traditional GA-BP neural network known to fruit, and precision of prediction is higher, and mean square error is smaller, error and its percentage
Than lower.
4, the present invention is a kind of eutrophy water quality evaluation algorithms of multi input.Multiple sensors measure multiple-quality water parameter
Data, and calibrated by neural network, by being more nearly PH, dissolved oxygen, ammonia nitrogen, the turbidity water quality data of true value,
The water grade in current measurement waters is judged according to fuzzy neural network.It is more accurate compared to single parameter evaluation method.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, the accompanying drawings in the following description is only implementation of the invention
Example, to those skilled in the art, without creative efforts, can also set according to these attached drawings
Count other similar circuit figures.
Fig. 1 is a kind of flow chart of the eutrophy quality evaluation method of GA-BP optimization TSFNN;
Fig. 2 is BP neural network forecast error and chart of percentage comparison in algorithm;
Fig. 3 is GA-BP neural network forecast error and chart of percentage comparison in algorithm;
Fig. 4 is modified GA-BP neural network forecast error and chart of percentage comparison in algorithm;
Fig. 5 is middle grade prediction-error image after algorithm optimization;
Specific embodiment
Of the invention is described in detail with preferred embodiment with reference to the accompanying drawing, so that advantages and features of the invention
It can be easier to be readily appreciated by one skilled in the art.
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention, but the present invention can be with
Implemented using other than the one described here other way, those skilled in the art can be without prejudice to intension of the present invention
In the case of do similar popularization, therefore the present invention is not limited by the specific embodiments disclosed below.
It please refers to shown in Fig. 1:
1, a kind of eutrophy quality evaluation method of GA-BP optimization TSFNN, which is characterized in that include the following steps:
S101:National water environment quality standard (GB3838-2002) is obtained, building is based on GA-BP Optimization of Fuzzy mind
Through network model, and initial time genetic algorithm, BP neural network and TS fuzzy neural network, the initiation parameter of set algorithm;
After the completion of S101, into S102;
S102:Training number from 2000 groups of data of sensor historic data decimation of multi-parameter water quality system as model
According to the data calibration part of training pattern, using training data prediction mean square error MSE as ideal adaptation during training
Angle value obtains the initial threshold and weight of network according to optimum individual.200 groups of data are randomly selected again as test data, are used
Network performance after test training;After the completion of S102, into S103;
S103:Even distribution pattern interpolation mark at equal intervals is carried out according to water environment quality standard (GB3838-2002)
The mode of quasi- data obtains 400 groups of data, wherein I to II class, II to Group III, III to IV class and IV to V class data each 100
Group.Water quality assessment obscure portions neural network of 350 groups of data for training pattern is randomly selected, 50 groups of data are used for Test Network
Whether network has good generalization ability and judgment accuracy;After the completion of S103, into S104
S104:The BP neural network model part for obtaining fuzzy neural network model and genetic algorithm optimization collectively constitutes one
Kind optimizes the quality evaluation method of TSFNN according to the GA-BP of multi-parameter.Using 200 groups of test datas as GA-BP neural network
Input data is calibrated, and is obtained after prediction result again as the input data of TS fuzzy neural network to current sensor
The water quality in waters carries out grade evaluation where node.
2, include establishing the more hidden layer neural networks of GA-BP in S101, include the following steps:
S201:According to sensor measurement to water quality data dimension can determine that input node and output node are 4, use
The structure of double hidden layers, the mean square error emulated by different node in hidden layer, absolute error, absolute error percentage it is big
It is small, determine that neural network structure is 4-7-6-4.After the completion of S201.Into S202;
S202:Parameter in BP neural network is set, including:Learning rate, the number of iterations, training objective allow to miss
Difference;
S202:Parameter in BP neural network is set, including:Learning rate, the number of iterations, training objective allow to miss
Difference;The parameter of GA algorithm is set, including:Group size N, genetic algebra G, crossover probability Pc, mutation probability Pm.It is completed to S202
Afterwards, into S203;
S203:Choose training data of 2000 groups of data as GA-BP neural network, and by the mean square error of training data
As ideal adaptation angle value, is intersected, mutation operation and calculates fitness value.Into circulation, until acquiring optimum individual.To
S203 is completed, into S204;
S204:The power threshold value in optimum individual is obtained, BP neural network is assigned to as initial power threshold value, utilizes 2000 groups
Training data calculates network error, and the power threshold value of network is constantly updated in training, adjusts until meeting.Simulation and prediction obtains
As a result.
3, signified in S203 to use mean square error MSE as ideal adaptation angle value, calculation is:Above formula explanation, mean square errorBy the variance of point estimation
And deviationSquare two parts composition.
It please refers to shown in Fig. 2:
Fig. 2 is common BP neural network in the concrete condition being fitted to water quality model.Wherein to four parameter predictions
When, the error of PH and dissolved oxygen maintains essentially within 0.2, and the error of ammonia nitrogen is smaller relative to other three kinds of parameter errors, turbid
The fluctuating error of degree is larger.Four percentage error is all larger than 2%.
It please refers to shown in Fig. 3:
Fig. 3 is that the BP network of GA optimization is fitted and the specific prediction case of four parameters in complicated water quality model.Four
Error fluctuates between 0.1 and 0.2 substantially, turbidity prediction error still fluctuate it is larger, but it is whole be significantly less than be not optimised before.Four
Prediction percentage error of a water quality parameter in GA-BP network is controlled 1.5% to 2.0% or so, illustrates the network pair
Water quality model has good Training Capability and predictive ability.
It please refers to shown in Fig. 4:
Fig. 4 is that the modified GA-BP network using mean square error MSE as ideal adaptation angle value is fitted water quality model
Situation afterwards.Neural network forecast absolute error fluctuation amplitude reduces, whole that more steady situation is presented.Wherein water quality parameter PH, molten
The prediction error of solution oxygen and ammonia nitrogen substantially remains within ± 0.1, and the prediction error of turbidity is most of within ± 0.1, but
Fluctuate larger, discrete error close ± 0.2.It please refers to shown in Fig. 5:
The water grade prediction fluctuating error that initial data obtains is larger.When TS fuzzy neural network prediction error value is greater than
When 1, will affect system evaluation as a result, then influence result sample number be 25, accuracy 87.5%, and pass through modified GA-
Before water grade prediction error after the BP network optimization is less than optimization on the whole, fluctuating error reduces and major part can be stablized
0.4 hereinafter, and influence result sample number be 0, accuracy reaches 100%, improves 14.28% compared with the former.
In order to verify this method verifying the case where calibrating to the fitting of non-linear water quality model and water quality data, pass through
Matlab software calculates three kinds of different neural networks to the prediction percentage error of four kinds of different quality parameters, is shown in Table 1 institute
Show:
1 water quality parameter of table predicts percentage error
By table 1 it is found that when BP neural network is fitted water quality model, the percentage error of four kinds of parameters is all larger than
2%, when using GA-BP neural network, the percentage error of four water quality parameters is controlled 1.About 5%, precision of prediction
21.9%, 30.2%, 38.4%, 23.9% has been respectively increased compared to BP neural network;When use modified GA-BP nerve net
When network, compared to the GA-BP network before improvement, percentage error reduces 25.7%, 29.2%, 23.2%, 25.6% respectively
The principle of eutrophy quality evaluation method of GA-BP optimization TSFNN of the present embodiment a kind of is:It is how hidden by establishing
BP neural network containing layer recognizes complicated water quality model, and using using the mean square error of test data as ideal adaptation angle value
GA algorithm BP neural network is optimized, obtain optimal initial power threshold value of the power threshold value as BP neural network, improve it
Fitting degree and generalization ability.TS fuzzy neural network is built according to the PH, dissolved oxygen, turbidity of water quality after calibration and ammonia nitrogen data
The vertical model for being suitable for local water quality assessment, thus judge waters where sensor node water grade and whether eutrophy
Change.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use this hair
It is bright.Various modifications to these embodiments will be readily apparent to those skilled in the art, defined herein
General Principle can be realized in other embodiments without departing from the spirit and scope of the present invention.Therefore, originally
Invention is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein
Consistent widest scope.
Claims (3)
1. a kind of eutrophy quality evaluation method of GA-BP optimization TSFNN, which is characterized in that include the following steps:
S101:National water environment quality standard (GB3838-2002) is obtained, building is based on GA-BP Optimization of Fuzzy nerve net
Network model, and initial time genetic algorithm, BP neural network and TS fuzzy neural network, the initiation parameter of set algorithm;To
After the completion of S101, into S102;
S102:Training data from 2000 groups of data of sensor historic data decimation of multi-parameter water quality system as model, instruction
The data calibration part for practicing model, using training data prediction mean square error MSE as ideal adaptation angle value during training,
The initial threshold and weight of network are obtained according to optimum individual.200 groups of data are randomly selected again as test data, for testing
Network performance after training;After the completion of S102, into S103;
S103:Even distribution pattern interpolation criterion numeral at equal intervals is carried out according to water environment quality standard (GB3838-2002)
According to mode obtain 400 groups of data, wherein I to II class, II to Group III, III to IV class and each 100 groups of IV to V class data.With
Machine chooses the water quality assessment obscure portions neural network that 350 groups of data are used for training pattern, and 50 groups of data are for test network
It is no to have good generalization ability and judgment accuracy;After the completion of S103, into S104
S104:The BP neural network model part for obtaining fuzzy neural network model and genetic algorithm optimization collectively constitutes a kind of
According to the quality evaluation method of the GA-BP optimization TSFNN of multi-parameter.Using 200 groups of test datas as the input of GA-BP neural network
Data are calibrated, and are obtained after prediction result again as the input data of TS fuzzy neural network to current sensor node
The water quality in place waters carries out grade evaluation.
2. the eutrophy quality evaluation method of GA-BP optimization TSFNN according to claim 1 a kind of, which is characterized in that
Include establishing the more hidden layer neural networks of GA-BP in S101, includes the following steps:
S201:According to sensor measurement to water quality data dimension can determine that input node and output node are 4, using double hidden
Structure containing layer, the mean square error emulated by different node in hidden layer, absolute error, the size of absolute error percentage,
Determine that neural network structure is 4-7-6-4.After the completion of S201.Into S202;
S202:Parameter in BP neural network is set, including:The allowable error of learning rate, the number of iterations, training objective;If
The parameter of GA algorithm is set, including:Group size N, genetic algebra G, crossover probability Pc, mutation probability Pm.After the completion of S202, into
Enter S203;
S203:Choose training data of 2000 groups of data as GA-BP neural network, and using the mean square error of training data as
Ideal adaptation angle value is intersected, mutation operation and calculates fitness value.Into circulation, until acquiring optimum individual.To S203
It completes, into S204;
S204:The power threshold value in optimum individual is obtained, BP neural network is assigned to as initial power threshold value, utilizes 2000 groups of training
Data calculate network error, and the power threshold value of network is constantly updated in training, adjust until meeting.Simulation and prediction obtains result.
3. the eutrophy quality evaluation method of GA-BP optimization TSFNN according to claim 2 a kind of, which is characterized in that institute in S203
Refer to that using mean square error MSE, calculation is as ideal adaptation angle value:
Above formula explanation, mean square errorBy the variance of point estimationAnd deviationSquare two parts composition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810434779.6A CN108872508A (en) | 2018-05-08 | 2018-05-08 | A kind of eutrophy quality evaluation method of GA-BP optimization TSFNN |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810434779.6A CN108872508A (en) | 2018-05-08 | 2018-05-08 | A kind of eutrophy quality evaluation method of GA-BP optimization TSFNN |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108872508A true CN108872508A (en) | 2018-11-23 |
Family
ID=64333121
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810434779.6A Pending CN108872508A (en) | 2018-05-08 | 2018-05-08 | A kind of eutrophy quality evaluation method of GA-BP optimization TSFNN |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108872508A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109655594A (en) * | 2018-12-30 | 2019-04-19 | 杭州铭展网络科技有限公司 | A kind of water quality acquisition methods |
CN109740286A (en) * | 2019-01-21 | 2019-05-10 | 北京工业大学 | A kind of Water Quality Forecasting Model of Lake construction method of hybrid optimization BP neural network |
CN110208478A (en) * | 2019-06-20 | 2019-09-06 | 南京芊玥机器人科技有限公司 | A kind of solar energy unmanned boat carrying water environment monitoring system |
CN110738307A (en) * | 2019-09-16 | 2020-01-31 | 天津大学 | Application of BP neural network model in prediction of water quality of peach forest mouth reservoir |
CN111104860A (en) * | 2019-11-19 | 2020-05-05 | 浙江工业大学 | Unmanned aerial vehicle water quality chromaticity monitoring method based on machine vision |
CN113642109A (en) * | 2021-08-17 | 2021-11-12 | 湖南大学 | Multi-sensor scheme evaluation and optimization system and method based on BP neural network |
CN113971517A (en) * | 2021-10-25 | 2022-01-25 | 中国计量大学 | GA-LM-BP neural network-based water quality evaluation method |
CN115053685A (en) * | 2022-06-07 | 2022-09-16 | 舒城县农业科学研究所 | Water and fertilizer management and control system for aquatic vegetable planting |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106951983A (en) * | 2017-02-27 | 2017-07-14 | 浙江工业大学 | Injector performance Forecasting Methodology based on the artificial neural network using many parent genetic algorithms |
-
2018
- 2018-05-08 CN CN201810434779.6A patent/CN108872508A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106951983A (en) * | 2017-02-27 | 2017-07-14 | 浙江工业大学 | Injector performance Forecasting Methodology based on the artificial neural network using many parent genetic algorithms |
Non-Patent Citations (2)
Title |
---|
崔东文;: "水质综合评价的LM-BP神经网络通用模型应用", 水资源保护, vol. 29, no. 06, pages 18 - 25 * |
张维砚;沈蓓雷;童琰;胡雪芹;徐春燕;由文辉;薄芳芳;: "基于GA-BP模型的景观小水体富营养化评价方法", 中国环境科学, vol. 31, no. 04, pages 674 - 679 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109655594A (en) * | 2018-12-30 | 2019-04-19 | 杭州铭展网络科技有限公司 | A kind of water quality acquisition methods |
CN109740286A (en) * | 2019-01-21 | 2019-05-10 | 北京工业大学 | A kind of Water Quality Forecasting Model of Lake construction method of hybrid optimization BP neural network |
CN110208478A (en) * | 2019-06-20 | 2019-09-06 | 南京芊玥机器人科技有限公司 | A kind of solar energy unmanned boat carrying water environment monitoring system |
CN110738307A (en) * | 2019-09-16 | 2020-01-31 | 天津大学 | Application of BP neural network model in prediction of water quality of peach forest mouth reservoir |
CN111104860A (en) * | 2019-11-19 | 2020-05-05 | 浙江工业大学 | Unmanned aerial vehicle water quality chromaticity monitoring method based on machine vision |
CN111104860B (en) * | 2019-11-19 | 2022-02-15 | 浙江工业大学 | Unmanned aerial vehicle water quality chromaticity monitoring method based on machine vision |
CN113642109A (en) * | 2021-08-17 | 2021-11-12 | 湖南大学 | Multi-sensor scheme evaluation and optimization system and method based on BP neural network |
CN113642109B (en) * | 2021-08-17 | 2023-07-04 | 湖南大学 | Multi-sensor scheme evaluation and optimization system and method based on BP neural network |
CN113971517A (en) * | 2021-10-25 | 2022-01-25 | 中国计量大学 | GA-LM-BP neural network-based water quality evaluation method |
CN115053685A (en) * | 2022-06-07 | 2022-09-16 | 舒城县农业科学研究所 | Water and fertilizer management and control system for aquatic vegetable planting |
CN115053685B (en) * | 2022-06-07 | 2023-11-28 | 舒城县农业科学研究所 | Fertilizer water management and control system for aquatic vegetable planting |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108872508A (en) | A kind of eutrophy quality evaluation method of GA-BP optimization TSFNN | |
CN106841075B (en) | COD ultraviolet spectra on-line checking optimization method neural network based | |
Mulia et al. | Hybrid ANN–GA model for predicting turbidity and chlorophyll-a concentrations | |
CN101793887B (en) | Construction method of fuzzy neural network expert system for water quality assessment in turbot culture | |
CN102183621B (en) | Aquaculture dissolved oxygen concentration online forecasting method and system | |
CN108898215B (en) | Intelligent sludge bulking identification method based on two-type fuzzy neural network | |
CN108876021B (en) | Medium-and-long-term runoff forecasting method and system | |
CN109816094A (en) | Optical dissolved oxygen sensor non-linear temperature compensation method based on neural network L-M algorithm | |
CN116630122B (en) | Lake ecological hydraulic regulation and control method and system based on hydrologic-ecological response relation | |
CN110824915B (en) | GA-DBN network-based intelligent monitoring method and system for wastewater treatment | |
CN104700153A (en) | PH (potential of hydrogen) value predicting method of BP (back propagation) neutral network based on simulated annealing optimization | |
CN109409441A (en) | Based on the coastal waters chlorophyll-a concentration remote sensing inversion method for improving random forest | |
CN102183802B (en) | Short-term climate forecast method based on Kalman filtering and evolution modeling | |
CN108764473A (en) | A kind of BP neural network water demands forecasting method based on correlation analysis | |
CN114154401B (en) | Soil erosion modulus calculation method and system based on machine learning and observation data | |
CN111510500A (en) | Water quality monitoring data calibration method based on edge calculation | |
CN108960334A (en) | A kind of multi-sensor data Weighted Fusion method | |
CN108595892A (en) | Soft-measuring modeling method based on time difference model | |
CN110428107A (en) | A kind of corn yield remote sensing prediction method and system | |
CN110309609A (en) | A kind of architecture indoor air quality evaluation method based on rough set and wavelet neural network | |
CN108460462A (en) | A kind of Interval neural networks learning method based on interval parameter optimization | |
CN109033509A (en) | A kind of Fusion Model construction method based on dam deformation analysis | |
CN114970377A (en) | Method and system for field flood forecasting based on Xinanjiang and deep learning coupling model | |
CN103637800B (en) | Based on the method for the bioelectrical impedance analysis of eight sections of impedance models | |
CN115859789A (en) | Method for improving inversion accuracy of polar atmosphere temperature profile |
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 | ||
DD01 | Delivery of document by public notice |
Addressee: Yang Lijuan Document name: Notice of patent application for invention entering the substantive examination stage |
|
DD01 | Delivery of document by public notice | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181123 |
|
WD01 | Invention patent application deemed withdrawn after publication |