Prediction of dissolved oxygen in reservoirs using adaptive network-based fuzzy inference system

Ranković, Vesna; Radulović, Jasna; Radojević, Ivana D.; Ostojić, Аleksandar; Čomić, Ljiljana R.

doi:10.2166/hydro.2011.084

Cited by 37 publications

(20 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These data are monitored regularly each month at eight different sites. The selection of an appropriate set of input variables from all possible input variables during artificial intelligence (AI) model development is important for obtaining high-quality model (Ranković et al 2012). Many of the described methods for input variable selection are based on heuristics, expert knowledge, statistical analysis or a combination of these.…”

Section: Study Areamentioning

confidence: 99%

“…Additionally, such models have analytical solutions, but they have boundary conditions as limitations (Basant et al 2010). Also, since a large number of factors affect the water quality, it has a complicated nonlinear relation with the variables; therefore, traditional data processing methods are no longer good enough for solving the problem (Xiang et al 2006;Ranković et al 2012). In recent years, several researches have been conducted on water quality forecast models (Palani et al 2008;Basant et al 2010;Faruk 2010).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models

Emamgholizadeh

Kashi

Marofpoor

et al. 2013

Int. J. Environ. Sci. Technol.

143

View full text Add to dashboard Cite

This paper describes the application of multilayer perceptron (MLP), radial basis network and adaptive neuro-fuzzy inference system (ANFIS) models for computing dissolved oxygen (DO), biochemical oxygen demand (BOD) and chemical oxygen demand (COD) levels in the Karoon River (Iran). Nine input water quality variables including EC, PH, Ca, Mg, Na, Turbidity, PO 4 , NO 3 and NO 2 , which were measured in the river water, were employed for the models. The performance of these models was assessed by the coefficient of determination R 2 , root mean square error and mean absolute error. The results showed that the computed values of DO, BOD and COD using both the artificial neural network and ANFIS models were in close agreement with their respective measured values in the river water. MLP was also better than other models in predicting water quality variables. Finally, the sensitive analysis was done to determine the relative importance and contribution of the input variables. The results showed that the phosphate was the most effective parameters on DO, BOD and COD.

show abstract

Section: Study Areamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models

Emamgholizadeh

Kashi

Marofpoor

et al. 2013

Int. J. Environ. Sci. Technol.

143

View full text Add to dashboard Cite

show abstract

“…Analysis of the impacts of different parameters on oxygen content in the reservoir Gruza was tested based on data from the database and other tools and models, which are all shown similar conclusions [24,25]. Some specifics and differences can be explained by the The examples show that IS SeLar can provide nec essary information to the users in the way that is the most suitable for them.…”

Section: Classifymentioning

confidence: 90%

Management information system of lakes and reservoirs

et al. 2012

Self Cite

View full text Add to dashboard Cite

488 1 Contemporary world is facing the fact that there is a discrepancy between different needs for drinking water and available resources of relatively clean water. There is a tendency for the proportion between the needs for and the availability of clean water to be vio lated under the influence of a rising number of popu lations as well as a greater contamination of water sys tems. Therefore, it is necessary to maintain the state of reservoirs, both natural and artificial ones, at a satis factory level. As lakes are evolutionary, natural and contemporary hydrographic objects, the research, especially if related to the living world and ecological relations, has to be constant, systematic, versatile and purposeful. Lake ecosystems are very susceptible to various influences, which is caused by a small degree of self regulation, wherefore the preventive lake pro tection is better than remediation of the harmful effects [13].The application of information technologies is very significant in managing lakes and reservoirs. Informa tion technologies enable merging all the relevant facts needed for managing as well as for showing them in the form best suitable for the user. For example, the unique information system of European lakes where 1 The article is published in the original. the data are organised within a relational database in five basic tables: station information, chemistry sam ple information, biology sample information, chemis try values (including variables such as pH or phospho rus) and biological values (such as biomass or abun dance per taxon) [22]. A special attention is given to the systems dealing with water quality. For example, information system for determining IBI (The Index of Biotic Integrity) for fish communities based on a greater number of input parameters [31], information system for managing water resources by integrating hydro chemical, hydrological, data about microbiol ogy, phytoplankton, zooplankton, macrophytes etc.[15], information system based on relational database containing a fuzzy models for water quality determin ing [19,20,21], real time information and control sys tem for decision support [11], etc.The objective of the information system (IS) SeLaR is creating an integral and purposeful informa tion system about lakes and reservoirs in Serbia. It should enable gathering in one place all the available data concerning lakes and reservoirs in Serbia, a rela tively easy access to scientific data and manipulation, a new, modern and contemporary insight into the state of lakes and hydroreservoirs in Serbia, as well as an Abstract-Information technologies provide a significant support in the management of lakes and reservoirs. Existing information systems mostly did not comprise all the necessary data and enabled collecting only a limited set of reports. More complex data analyses such as statistical analyses and data research were con ducted in separate environments and recquired additional preparation of the data. Information system of lakes and reservoirs in Serbia (SeLaR) was ...

show abstract

“…Therefore, traditional data processing and modelling methods required several input parameters which are hard to reach and make it a time consuming and expensive process are insufficient to solve the problems related to water quality (Ranković, Radulović, Radojević, Ostojić, & Čomić, 2012;Bayatzadeh Fard, Ghadimi, & Fattahi, 2017). The integration of different techniques and methods will contribute to the future of eco-environmental modelling (Chen, MoralesChaves, Li, & Mynett, 2006).…”

Section: Introductionmentioning

confidence: 99%

Untitled

Sönmez¹,

Kale²,

Ozdemir³

et al. 2018

Turk. J. Fish. Aquat. Sci.

View full text Add to dashboard Cite

Water quality is one of the main characteristics of a river system and prediction of water quality is the key factor in water resource management. Different physical, biological and chemical parameters including heavy metals can be used to assess river water quality. Evaluation of the water quality in the rivers is quite difficult and requires more time and effor t because of the fact that many factors affect water quality. Traditional data processing methods are insufficient to solve this problem. Therefore, in this study, an adaptive neuro-fuzzy inference system (ANFIS) model was developed to predict the concentrations of cadmium (Cd) in the Filyos River, Turkey. For this purpose, water samples collected at 7 sampling locations in the river during December 2014-2015 were used to develop ANFIS model. The available data set was apportioned into two separate sections for training and testing the ANFIS model. Developed models aimed to use the least parameters to estimate Cd concentration. As a result, a relatively higher correlation (R 2 =0.91) was found between observed and modelled Cd concentrations. The results indicated that the ANFIS model gave reasonable estimates for the concentrations of Cd with a high degree accuracy and robustness. In conclusion, this paper suggests that ANFIS methodology produce very successful findings and has the ability to predict Cd concentration in water resources. The outcomes of this research provide more information, simulation, and prediction about heavy metal concentration in natural aquatic ecosystems. Therefore, ANFIS can be used in further researches on water quality monitoring.

show abstract

Prediction of dissolved oxygen in reservoirs using adaptive network-based fuzzy inference system

Cited by 37 publications

References 35 publications

Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models

Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models

Management information system of lakes and reservoirs

Untitled

Contact Info

Product

Resources

About