CN103957549B - Node distribution optimization method for river local area wireless sensor based on genetic algorithm - Google Patents
Node distribution optimization method for river local area wireless sensor based on genetic algorithm Download PDFInfo
- Publication number
- CN103957549B CN103957549B CN201410202917.XA CN201410202917A CN103957549B CN 103957549 B CN103957549 B CN 103957549B CN 201410202917 A CN201410202917 A CN 201410202917A CN 103957549 B CN103957549 B CN 103957549B
- Authority
- CN
- China
- Prior art keywords
- river
- monitoring node
- monitoring
- represent
- node
- 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.)
- Active
Links
Landscapes
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention provides a node distribution optimization method for a river local area wireless sensor based on a genetic algorithm. The monitoring node distribution optimization problem of a river local area is converted into a 0/1 planning problem through the genetic algorithm. Reasonable distribution optimization selection of monitoring nodes is finally achieved through the genetic algorithm operations of inheritance, chiasma, variation and the like of chromosome and comparison of a fitness value. In addition, a plurality of evaluation indexes are adopted to conduct weighing and evaluation on the fitness value, the evaluation indexes are respectively the 'water level distribution difference ratio', the 'flow speed distribution difference ratio', the 'water quality distribution difference ratio', the 'channel slope gradient' and the 'river trend parameter', and in this way, a decision maker can flexibly adjust the index weight according to the difference of hydrology environments of the river local area so as to improve the adaptation of the algorithm.
Description
Technical field
The present invention relates to wireless senser distribution optimization method, specifically a kind of river local area radio based on genetic algorithm
Sensor node distribution optimization method.
Background technology
China is vast in territory, rivers are numerous and widely distributed, environment is complicated, and river basins hydrological data is adopted
Collection and management are an important foundation sex works, are both needed to put into a large amount of human and material resources every year.And with expanding economy, from
So environment has been subjected to serious destruction, and water pollution is increasingly serious, and the monitoring meanss for taking modernization are strengthened to river basins water
The collection of literary data and collect, become particularly important.
At present for the monitoring of river basins hydrographic data, relatively common is to arrange some prisons in river periphery ad-hoc location
Measuring point, then takes the mode of personal monitoring or monitoring of equipment, the hydrographic data required for periodic collection, then concentrate again into
Row analysis.In recent years, with the development of detection technique and sensor technology, occurred in that it is some can be to the river basins hydrology
Data carry out the electronic equipment of real time on-line monitoring, and by communication technology, are realized to river stream by building transmission network
The monitoring of domain hydrographic data and remote transmission.
However, for the such as river such as urban sewage mouth specific localized areas, local hydrological environment is excessively complicated, major embodiment
：On the one hand for riverbed longitudinal direction section, because it is class trapezium structure, the situation that there is upper and lower convection current, therefore in same position
Put the data detected by arranged monitoring device and do not possess concordance in riverbed longitudinal direction section, on the other hand for river table
Face, because discharge of river moves towards complicated, River Hydrology data can change with the riverbank width that runoff is moved towards, and not possess one
Cause property.Analyze visible for river specific localized areas by more than, be with anisotropic uneven in its regional extent
Body, therefore, for the selection of local monitor node should not be uniformly distributed design side using common big area monitoring network node
Method, mainly due to the requirement that can not meet river local hydrographic data monitoring completeness, globality, effectiveness, this is accomplished by
Reasonably optimizing is carried out to monitoring node distribution according to river local hydrographic data.
At present, by open source literature can find with regard to the related patent application of sensor network coverage optimization method with
And Academic Periodical Papers, including the patent of invention " wireless sensor network based on genetic algorithm of Publication No. CN101459915
Coverage optimization method ", " a kind of wireless sensor network node covers excellent the patent of invention of Publication No. CN103237312A
Change algorithm ", Liu Yuying etc. is published in《Sensing technology journal》A kind of " wireless sensing based on genetic algorithm of the 6th phase in 2009
Device network node optimization method ", Zhang Shi etc. is published in《Northeastern University's journal (natural science edition)》4th phase in 2007 it is " wireless
The distribution optimization problem of mobile node in sensor network ", Lei Lin etc. is published in《Journal of UEST of China》2nd phase in 2009
" the wireless sensor network path optimization based on genetic algorithm ", and Lv Guanghui etc. is published in《Microcomputer ＆ its Application》2010
A kind of " wireless sensor network overlay model based on genetic algorithm " of the 15th phase of year.Although above-mentioned two patent applications and four
Piece open periodical document proposes relevant sensor network node coverage optimization method and network path optimization method, but should
Method its network coverage and routing problem when being intended to how to solve to dispose within the specific limits wireless senser itself, comments
Sentence the good and bad standard of method is whether to realize the coverage rate to wanting deployment region, or wireless biography whether is realized in region
Sensor network transmission route.Research and the content of the invention involved by it is not involved how to using the characteristic of institute's monitoring object
And build-in attribute is being monitored the optimization of node layout, thus the content of its achievement for being studied and patent application be can not
Enough meet or be applied to river regional area monitoring node layout optimization work.
Although above-mentioned patent application and open source literature give a kind of sensor placement that carries out in certain area and cover excellent
The method of change, but for the monitoring of the river such as river sewage draining exit regional area hydrographic data, due to local hydrological environment it is excessively multiple
It is miscellaneous, and in monitoring node a small range, the consistent feature of hydrographic data, the distribution of its monitoring node is more to be considered
It is that can monitoring node monitor the data for collecting and reflect the overall picture of river regional area hydrological environment, and each monitoring node
Whether the data for being collected have typicality, with above-mentioned two applications for a patent for invention be absorbed in it is important that inconsistent, and
And not yet there is the report about monitoring node distribution optimization method in river basins hydrologic monitoring field at present.
The content of the invention
The purpose of the present invention is the needs according to the regional area hydrographic data monitoring of the river such as above-mentioned river sewage draining exit, it is considered to
It is excessively complicated to local hydrological environment, and in monitoring node a small range, the consistent feature of hydrographic data, Yi Jijian
Survey node monitors the data for collecting will can reflect the overall picture of river regional area hydrological environment, and each monitoring node is collected
Data will have typicality requirement, there is provided a kind of river local area radio sensor node distribution optimization based on genetic algorithm
Method.
A kind of river local area radio sensor node distribution optimization method based on genetic algorithm, comprises the steps：
(1) optimization problem description：For the needs of river regional area monitoring node distribution, it is assumed that press grid in local
Shape is uniformly distributed N number of Hydrological Data Acquisition monitoring node；
(2) genetic algorithm parameter is initialized, and first generation coding is carried out to chromosome, here using binary system random coded
Mode, 0 expression does not use the monitoring node, and 1 represents using the monitoring node, then for N number of monitoring node generates length for N's
Binary string, is shown below：
X=[x1,x2,…,xN] xi={ 0,1 }
(3) fitness of each chromosome is calculated, fitness function is：
F (X)=w1f1(X)+w2f2(X)+w3f3(X)+w4β1+w5β2
Wherein, f1Represent water level distributional difference rate, f2Represent velocity flow profile variance rate, f3Represent water quality distributional difference rate, β1
Represent stream gradient than drop, β2Represent that river moves towards parameter, β1And β2For specific river region belongs to river basic geological study number
According to S={ S1,S2,…,SNRepresent monitoring node set, S*={ Sj,|xj=1 } subset of S, N (S are represented*) represent S*It is big
It is little,River level, flow velocity and the water quality of correspondence monitoring node position are represented respectively,It is right to represent respectively
Should the river mean water of all use monitoring node positions, mean flow rate and average water quality, w1、w2、w3、w4And w5Represent respectively
f1、f2、f3、β1And β2Weight.For stream gradient is than drop β1Where height, the flow velocity and flow in its correspondence river will increase
Plus, and river moves towards parameter beta2Size it is relevant with Coriolis force, it can also affect the flow and flow velocity in river；
(4) fitness after calculating is ranked up, and chooses optimal solution in proportion；
(5) according to wheel disc bet method, selective staining body is to chromosome population of future generation；
(6) intersection, mutation operation are performed to chromosome population of future generation, the present invention is using multiple-spot detection and uniform variation；
(7) if meeting end condition, terminate, otherwise return to step 3) proceed to calculate, at the end of calculating, its
Resulting optimal solution is the binary string that length is N, and the position number of 1 in the string represents the institute after distribution optimization is calculated
It is determined that the monitoring node sequence number that uses of needs, would know that for the prison adopted required for the monitoring of river regional area by it
Survey node.
River regional area monitoring node distribution optimization problem is converted into 0/1 planning and is asked by the present invention using genetic algorithm
Topic, by operatings of genetic algorithm such as the heredity of chromosome, intersection and variations, and passes through the relatively more final realization monitoring of fitness value
The reasonable layout optimum option of node.In addition, the algorithm of invention employs multiple evaluation indexes and assessment is weighted to fitness,
It is respectively " water level distributional difference rate ", " velocity flow profile variance rate ", " water quality distributional difference rate ", " stream gradient is than drop " and " river
Flow away to parameter ", this can be adjusted flexibly index weights for policymaker according to the difference of the hydrological environment of river regional area,
So as to the adaptability of boosting algorithm.
Description of the drawings
Fig. 1 is flow chart of the present invention based on the river local area radio sensor node distribution optimization method of genetic algorithm；
When Fig. 2 is to carry out example using the inventive method to calculate, fitness value shows with the effect of optimization that iterationses change
It is intended to.
Specific embodiment
Below in conjunction with the accompanying drawing in the present invention, the technical scheme in the present invention is clearly and completely described.
For the such as river such as urban sewage mouth specific localized areas, due to local hydrological environment it is excessively complicated, therefore can not
According to employed in the whole monitoring of river basins according to being such as monitored Node distribution cloth along discharge of river fixed range mode
The mode of office, this is larger mainly due to local hydrological Environment Changes, and a small amount of monitoring node cannot react local hydrological environment
Overall picture, and in monitoring node a small range, hydrographic data has concordance.Therefore, the cloth of monitoring node must be increased
Quantity is put, this is accomplished by the distribution to monitoring node and carries out reasonably optimizing, that is, reaching can reflect local hydrological environment overall picture
Purpose, reduces again the arrangement quantity of monitoring node, while the data that each monitoring node is collected should have typicality.
Generally, the problem of monitoring node distribution optimization model is described as follows：Hypothesis presses net in river regional area
Trellis is uniformly distributed N number of Hydrological Data Acquisition monitoring node, and monitoring node collection is combined into S={ S1,S2,…,SN, can adopt here
With calculating respectively in the case where the monitoring node is turned on or off, the hydrology that all monitoring nodes for enabling are collected
The diversity of data is estimated, and selection can most reflect regional area hydrological environment overall picture and institute's gathered data most typicality
Unlatching monitoring node combine as Node distribution optimum combination.
The present invention, as instrument, to river local monitor Node distribution optimization problem intelligence computation is carried out using genetic algorithm
Solve, and Node distribution optimization problem is converted into into 0/1 planning problem, in the hope of in the numerous prisons for assuming distributed in grid
Survey which actual selection in node arranges, in the number for guaranteeing to reflect river local hydrological environment overall picture and collected
In the case of according to most typicality, it is only necessary in the minimum monitoring node of some location arrangements.
In the present invention, by river regional area by it is latticed be uniformly distributed N number of Hydrological Data Acquisition monitoring node according to
Genetic algorithm basic theories, carries out binary chromosome coding, i.e., N number of monitoring node is encoded to the binary string that length is N, such as
Shown in following formula：
X=[x1,x2,…,xN] xi={ 0,1 }
Wherein, xi=0 expression does not use the monitoring node, xi=1 represents using the monitoring node.
The adaptive value of each chromosome in calculating the fitness function value of each chromosome coding respectively to evaluate colony.
In river local monitor Node distribution optimization problem, " water level distributional difference rate ", " velocity flow profile variance rate ", " water are employed
Matter distributional difference rate ", " stream gradient is than drop " and " moving towards parameter in river " five indexs, reflect respectively in river regional area
The diversity of each acquisition node hydrographic data, difference is bigger to illustrate, monitoring node higher to the reflection degree of hydrological environment overall picture
The data for being collected more have typicality.
In the present invention, using f1To represent water level distributional difference rate, using f2To represent velocity flow profile variance rate, adopt
f3To represent water quality distributional difference rate, using β1Stream gradient is represented than drop, using β2Represent that river moves towards parameter, β2And β2For
Specific river region belongs to river basic geological study data.For monitoring node set S={ S1,S2,…,SN, S*={ Sj,|xj=
1 } in the case of representing the subset of S, f1、f2And f3Definition be respectively：
Wherein, N (S*) represent S*Size,River level, the stream of correspondence monitoring node position are represented respectively
Speed and water quality,River mean water, the mean flow rate peace of all use monitoring node positions of correspondence are represented respectively
Equal water quality.By calculating f respectively1、f2And f3And combine β1、β2, can calculate the final fitness value of chromosome is：
F (X)=w1f1(X)+w2f2(X)+w3f3(X)+w4β1+w5β2
Wherein w1、w2、w3、w4And w5F is represented respectively1、f2、f3、β1And β2Weight, and w1+w2+w3+w4+w5=1.Separately
Outward, due toDimension it is different, need to be normalized operation here in calculating process, that is, be transformed into [0,
1] it is interval.Final chromosome fitness f can be obtained through above-mentioned computing, the value is bigger, represent corresponding node point
Cloth scheme is more excellent.
In the present invention, using the next progeny population of roulette selection, and using multiple-spot detection and uniform mutation algorithm pair
Chromosome in colony is changed.
Roulette is that the probability of selective staining body i and the fitness value of chromosome are directly proportional, the higher chromosome of fitness
Selected probability is higher, and can repeat in next filial generation.
The way of multiple-spot detection is to produce random binary sequence of the length for N, is selected based on the random binary sequence
Cross point, be 1 position on intersect, be that 0 position does not intersect.The size for assuming N is 20, and two dyeing before intersecting
Body is respectively：
The random binary sequence of generation is [00101100010101110100], then intersect two new chromosomes of generation
For：
The way of uniform variation be each gene in each chromosome in colony produce a random number ρ ∈ [0,
1], if the random number is less than mutation probability ρm, then the gene for selecting the chromosome enters row variation.Assume the chromosome to be made a variation
For：
Mutation operator selects the gene of the 2nd of the chromosome the, 8 to enter row variation, then the chromosome of new generation is：
WhereinFor 0/1 newly-generated random number.
Calculate through the optimization of above-mentioned loop iteration, at the end of calculating, the optimal solution obtained by it is that length is entered for the two of N
System string, the position number of 1 in the string represents the monitoring node sequence number for needing to use determined by after distribution optimization is calculated,
Would know that for the monitoring node adopted required for the monitoring of river regional area by it.
In order to further illustrate the specific implementation process of the present invention, here based on certain city contaminating enterprises river is arranged
Dirty saliva text data monitoring needs to provide the test example that node layout's optimization is monitored using inventive algorithm.In this example
In, with contaminating enterprises' river sewage draining exit as border center, delimit the monitored area of 1000m × 1000m.It is assumed that according to uniform grid
Shape installs 100 monitoring nodes, and optimized choice is done to the layout of monitoring node used here as the optimized algorithm of the present invention.Heredity is calculated
The parameter setting of method is：Population size is 50, mutation probability ρm=0.05, Cycle Evolution iteration is typically passed through in genetic algorithm
Mode computing is carried out to initial chromosome colony to find out optimum chromosome, the algebraically of genetic iteration computing is set here
For 150, water level distributional difference rate, velocity flow profile variance rate, water quality distributional difference rate, stream gradient move towards parameter than drop, river
Weight be respectively w1=0.125, w2=0.125, w3=0.5, w4=0.125 and w5=0.125.It is average that 20 suboptimization are calculated
Fitness value (diversity value) is 0.9458, and the average monitoring node quantity for needing is 15.6.Fig. 2 is given in this example, is adopted
Be iterated optimization with genetic algorithm, its average fitness value (diversity value) with the change curve of the increase of iterationses,
With the increase of iterationses, its average fitness value (diversity value) constantly increases, corresponding after showing iterative calculation every time
Node distribution scheme is more excellent.When iterating to for 150 generations, its average fitness value (diversity value) has reached maximum, shows now
Node distribution scheme is optimum.
In addition, in order to the inventive method is contrasted with the effect of common river local monitor node selection method,
Herein for preferred 16 monitoring nodes of institute in example, further using institute during usual river local monitor node selection
Using parameter, including " water-head ", " current difference ", and respectively weighting weight is w1=0.5, w2=0.5, it is selected through calculating
Monitoring node " water-head ", " current difference " parameter diversity value be 0.7654.So as to find out and the method phase in the present invention
Than adopting " water-head " in usual river local monitor node selection method, " current difference " parameter to be monitored by foundation
The ability of Node distribution reflection river local hydrographic data less than in the inventive method with " water level distributional difference rate ", " flow velocity
Distributional difference rate ", " water quality distributional difference rate ", " stream gradient is than drop " and " moving towards parameter in river " parameter are supervised by foundation
Survey the ability that Node distribution reflects river local hydrographic data.
Therefore, can be seen that and optimized choice is laid out using the inventive method from optimization measuring and calculating and parameter comparison result,
Reflect that the ability of the sewage draining exit hydrographic data overall picture and the typicality of institute's gathered data are made according to monitoring node institute gathered data
For criterion, the design of river regional area monitoring node distribution optimization can be well realized.
Claims (1)
1. a kind of river local area radio sensor node distribution optimization method based on genetic algorithm, it is characterised in that including as follows
Step：
(1) optimization problem description：For the needs of river regional area monitoring node distribution, it is assumed that by latticed equal in local
It is even to be distributed N number of Hydrological Data Acquisition monitoring node；
(2) genetic algorithm parameter is initialized, and first generation coding is carried out to chromosome, here using binary system random coded side
Formula, 0 expression does not use the monitoring node, and 1 represents using the monitoring node, then for N number of monitoring node generates length for the two of N
System string, is shown below：
X=[x1,x2,…,xN] xi={ 0,1 }
(3) fitness of each chromosome is calculated, fitness function is：
F (X)=w1f1(X)+w2f2(X)+w3f3(X)+w4β1+w5β2
Wherein, f1Represent water level distributional difference rate, f2Represent velocity flow profile variance rate, f3Represent water quality distributional difference rate, β1Represent
Stream gradient is than drop, β2Represent that river moves towards parameter, β1And β2For specific river region belongs to river basic geological study data, S=
{S1,S2,…,SNRepresent monitoring node set, S*={ Sj,|xj=1 } subset of S, N (S are represented*) represent S*Size,River level, flow velocity and the water quality of correspondence monitoring node position are represented respectively,Correspondence is represented respectively
The river mean water of all use monitoring node positions, mean flow rate and average water quality, w1、w2、w3、w4And w5Represent respectively
f1、f2、f3、β1And β2Weight；
(4) fitness after calculating is ranked up, and chooses optimal solution in proportion；
(5) according to wheel disc bet method, selective staining body is to chromosome population of future generation；
(6) intersection, mutation operation are performed to chromosome population of future generation；
(7) if meeting end condition, terminate, otherwise return to step 3) proceed to calculate, at the end of calculating, its gained
The optimal solution for arriving is the binary string that length is N, and the position number of 1 in the string is represented and determined after distribution optimization is calculated
The monitoring node sequence number that uses of needs, would know that for the monitoring section adopted required for the monitoring of river regional area by it
Point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410202917.XA CN103957549B (en) | 2014-05-14 | 2014-05-14 | Node distribution optimization method for river local area wireless sensor based on genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410202917.XA CN103957549B (en) | 2014-05-14 | 2014-05-14 | Node distribution optimization method for river local area wireless sensor based on genetic algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103957549A CN103957549A (en) | 2014-07-30 |
| CN103957549B true CN103957549B (en) | 2017-05-10 |
Family
ID=51334752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410202917.XA Active CN103957549B (en) | 2014-05-14 | 2014-05-14 | Node distribution optimization method for river local area wireless sensor based on genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103957549B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104270773B (en) * | 2014-10-17 | 2017-07-07 | 长江水利委员会长江科学院 | Basin sensor covering optimization method based on genetic algorithm multiple-objection optimization |
| CN111698306B (en) * | 2020-06-04 | 2021-08-06 | 中国科学院地理科学与资源研究所 | Hydrological real-time flow data acquisition and processing method based on Internet of things |
| CN112070309B (en) * | 2020-09-10 | 2021-03-30 | 西南民族大学 | Intelligent milk collection platform system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103175513A (en) * | 2013-03-01 | 2013-06-26 | 戴会超 | System and method for monitoring hydrology and water quality of river basin under influence of water projects based on Internet of Things |
| WO2013116527A1 (en) * | 2012-01-31 | 2013-08-08 | The University Of Toledo | Methods and devices for detection and measurement of analytes |
-
2014
- 2014-05-14 CN CN201410202917.XA patent/CN103957549B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013116527A1 (en) * | 2012-01-31 | 2013-08-08 | The University Of Toledo | Methods and devices for detection and measurement of analytes |
| CN103175513A (en) * | 2013-03-01 | 2013-06-26 | 戴会超 | System and method for monitoring hydrology and water quality of river basin under influence of water projects based on Internet of Things |
Non-Patent Citations (1)
| Title |
|---|
| 基于遗传算法的多传感器优化布站研究;周磊;《中国优秀硕士论文全文数据库》;20130615;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103957549A (en) | 2014-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chacon-Hurtado et al. | Rainfall and streamflow sensor network design: a review of applications, classification, and a proposed framework | |
| Emamgholizadeh et al. | Prediction the groundwater level of bastam plain (Iran) by artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) | |
| CN104270773B (en) | Basin sensor covering optimization method based on genetic algorithm multiple-objection optimization | |
| Chitsazan et al. | Forecasting groundwater level by artificial neural networks as an alternative approach to groundwater modeling | |
| Fadaee et al. | Suspended sediment prediction using integrative soft computing models: on the analogy between the butterfly optimization and genetic algorithms | |
| Sreekanth et al. | Comparison of FFNN and ANFIS models for estimating groundwater level | |
| Tapoglou et al. | Groundwater-level forecasting under climate change scenarios using an artificial neural network trained with particle swarm optimization | |
| Wolfs et al. | Development of discharge-stage curves affected by hysteresis using time varying models, model trees and neural networks | |
| Shirzad et al. | Pipe failure rate prediction in water distribution networks using multivariate adaptive regression splines and random forest techniques | |
| Currell et al. | A new assessment framework for transience in hydrogeological systems | |
| Haghighi et al. | Reliability-based layout design of sewage collection systems in flat areas | |
| Yilmaz et al. | Runoff estimation by machine learning methods and application to the Euphrates Basin in Turkey | |
| Nie et al. | Simulation and uncertainty analysis for groundwater levels using radial basis function neural network and support vector machine models | |
| Sahoo et al. | Efficacy of ANFIS-GOA technique in flood prediction: a case study of Mahanadi river basin in India | |
| Fan et al. | Development of PCA-based cluster quantile regression (PCA-CQR) framework for streamflow prediction: Application to the Xiangxi river watershed, China | |
| Houben et al. | Numerical Modeling of the Near‐Field Hydraulics of Water Wells | |
| CN103957549B (en) | Node distribution optimization method for river local area wireless sensor based on genetic algorithm | |
| Dehghanian et al. | Evaluation of coupled ANN-GA model to prioritize flood source areas in ungauged watersheds | |
| Qiu et al. | Seepage Monitoring Models Study of Earth‐Rock Dams Influenced by Rainstorms | |
| Rosin et al. | A committee evolutionary neural network for the prediction of combined sewer overflows | |
| Xiong et al. | Prediction of flooding in the downstream of the Three Gorges Reservoir based on a back propagation neural network optimized using the AdaBoost algorithm | |
| Kumar et al. | Adaptive neuro-fuzzy interface system based performance monitoring technique for hydropower plants | |
| Yadav et al. | Application of Artificial Neural Network and Genetic Algorithm Based Artificial Neural Network Models for River Flow Prediction. | |
| Narayana Reddy et al. | A hybrid artificial intelligence and semi-distributed model for runoff prediction | |
| Kikon et al. | ANFIS-based soft computing models for forecasting effective drought index over an arid region of India |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |