A linear and non-linear polynomial neural network modeling of dissolved oxygen content in surface water: Inter- and extrapolation performance with inputs' significance analysis
Само за регистроване кориснике
2018
Аутори
Šiljić-Tomić, AleksandraAntanasijević, Davor
Ristić, Mirjana
Perić-Grujić, Aleksandra
Pocajt, Viktor
Чланак у часопису (Објављена верзија)
Метаподаци
Приказ свих података о документуАпстракт
Accurate prediction of water quality parameters (WQPs) is an important task in the management of water resources. Artificial neural networks (ANNs) are frequently applied for dissolved oxygen (DO) prediction, but often only their interpolation performance is checked. The aims of this research, beside interpolation, were the determination of extrapolation performance of ANN model, which was developed for the prediction of DO content in the Danube River, and the assessment of relationship between the significance of inputs and prediction error in the presence of values which were of out of the range of training. The applied ANN is a polynomial neural network (PNN) which performs embedded selection of most important inputs during learning, and provides a model in the form of linear and non-linear polynomial functions, which can then be used for a detailed analysis of the significance of inputs. Available dataset that contained 1912 monitoring records for 17 water quality parameters was sp...lit into a "regular" subset that contains normally distributed and low variability data, and an "extreme" subset that contains monitoring records with outlier values. The results revealed that the non-linear PNN model has good interpolation performance (R-2 = 0.82), but it was not robust in extrapolation (R-2 = 0.63). The analysis of extrapolation results has shown that the prediction errors are correlated with the significance of inputs. Namely, the out-of-training range values of the inputs with low importance do not affect significantly the PNN model performance, but their influence can be biased by the presence of multi-outlier monitoring records. Subsequently, linear PNN models were successfully applied to study the effect of water quality parameters on DO content. It was observed that DO level is mostly affected by temperature, pH, biological oxygen demand (BOD) and phosphorus concentration, while in extreme conditions the importance of alkalinity and bicarbonates rises over pH and BOD.
Кључне речи:
Dissolved oxygen / Extrapolation / PNN / GMDH / Danube RiverИзвор:
Science of the Total Environment, 2018, 610, 1038-1046Издавач:
- Elsevier Science Bv, Amsterdam
Финансирање / пројекти:
- Развој и примена метода и материјала за мониторинг нових загађујућих и токсичних органских материја и тешких метала (RS-MESTD-Basic Research (BR or ON)-172007)
DOI: 10.1016/j.scitotenv.2017.08.192
ISSN: 0048-9697
PubMed: 28847097
WoS: 000411897700106
Scopus: 2-s2.0-85027850174
Институција/група
Tehnološko-metalurški fakultetTY - JOUR AU - Šiljić-Tomić, Aleksandra AU - Antanasijević, Davor AU - Ristić, Mirjana AU - Perić-Grujić, Aleksandra AU - Pocajt, Viktor PY - 2018 UR - http://TechnoRep.tmf.bg.ac.rs/handle/123456789/3974 AB - Accurate prediction of water quality parameters (WQPs) is an important task in the management of water resources. Artificial neural networks (ANNs) are frequently applied for dissolved oxygen (DO) prediction, but often only their interpolation performance is checked. The aims of this research, beside interpolation, were the determination of extrapolation performance of ANN model, which was developed for the prediction of DO content in the Danube River, and the assessment of relationship between the significance of inputs and prediction error in the presence of values which were of out of the range of training. The applied ANN is a polynomial neural network (PNN) which performs embedded selection of most important inputs during learning, and provides a model in the form of linear and non-linear polynomial functions, which can then be used for a detailed analysis of the significance of inputs. Available dataset that contained 1912 monitoring records for 17 water quality parameters was split into a "regular" subset that contains normally distributed and low variability data, and an "extreme" subset that contains monitoring records with outlier values. The results revealed that the non-linear PNN model has good interpolation performance (R-2 = 0.82), but it was not robust in extrapolation (R-2 = 0.63). The analysis of extrapolation results has shown that the prediction errors are correlated with the significance of inputs. Namely, the out-of-training range values of the inputs with low importance do not affect significantly the PNN model performance, but their influence can be biased by the presence of multi-outlier monitoring records. Subsequently, linear PNN models were successfully applied to study the effect of water quality parameters on DO content. It was observed that DO level is mostly affected by temperature, pH, biological oxygen demand (BOD) and phosphorus concentration, while in extreme conditions the importance of alkalinity and bicarbonates rises over pH and BOD. PB - Elsevier Science Bv, Amsterdam T2 - Science of the Total Environment T1 - A linear and non-linear polynomial neural network modeling of dissolved oxygen content in surface water: Inter- and extrapolation performance with inputs' significance analysis EP - 1046 SP - 1038 VL - 610 DO - 10.1016/j.scitotenv.2017.08.192 ER -
@article{ author = "Šiljić-Tomić, Aleksandra and Antanasijević, Davor and Ristić, Mirjana and Perić-Grujić, Aleksandra and Pocajt, Viktor", year = "2018", abstract = "Accurate prediction of water quality parameters (WQPs) is an important task in the management of water resources. Artificial neural networks (ANNs) are frequently applied for dissolved oxygen (DO) prediction, but often only their interpolation performance is checked. The aims of this research, beside interpolation, were the determination of extrapolation performance of ANN model, which was developed for the prediction of DO content in the Danube River, and the assessment of relationship between the significance of inputs and prediction error in the presence of values which were of out of the range of training. The applied ANN is a polynomial neural network (PNN) which performs embedded selection of most important inputs during learning, and provides a model in the form of linear and non-linear polynomial functions, which can then be used for a detailed analysis of the significance of inputs. Available dataset that contained 1912 monitoring records for 17 water quality parameters was split into a "regular" subset that contains normally distributed and low variability data, and an "extreme" subset that contains monitoring records with outlier values. The results revealed that the non-linear PNN model has good interpolation performance (R-2 = 0.82), but it was not robust in extrapolation (R-2 = 0.63). The analysis of extrapolation results has shown that the prediction errors are correlated with the significance of inputs. Namely, the out-of-training range values of the inputs with low importance do not affect significantly the PNN model performance, but their influence can be biased by the presence of multi-outlier monitoring records. Subsequently, linear PNN models were successfully applied to study the effect of water quality parameters on DO content. It was observed that DO level is mostly affected by temperature, pH, biological oxygen demand (BOD) and phosphorus concentration, while in extreme conditions the importance of alkalinity and bicarbonates rises over pH and BOD.", publisher = "Elsevier Science Bv, Amsterdam", journal = "Science of the Total Environment", title = "A linear and non-linear polynomial neural network modeling of dissolved oxygen content in surface water: Inter- and extrapolation performance with inputs' significance analysis", pages = "1046-1038", volume = "610", doi = "10.1016/j.scitotenv.2017.08.192" }
Šiljić-Tomić, A., Antanasijević, D., Ristić, M., Perić-Grujić, A.,& Pocajt, V.. (2018). A linear and non-linear polynomial neural network modeling of dissolved oxygen content in surface water: Inter- and extrapolation performance with inputs' significance analysis. in Science of the Total Environment Elsevier Science Bv, Amsterdam., 610, 1038-1046. https://doi.org/10.1016/j.scitotenv.2017.08.192
Šiljić-Tomić A, Antanasijević D, Ristić M, Perić-Grujić A, Pocajt V. A linear and non-linear polynomial neural network modeling of dissolved oxygen content in surface water: Inter- and extrapolation performance with inputs' significance analysis. in Science of the Total Environment. 2018;610:1038-1046. doi:10.1016/j.scitotenv.2017.08.192 .
Šiljić-Tomić, Aleksandra, Antanasijević, Davor, Ristić, Mirjana, Perić-Grujić, Aleksandra, Pocajt, Viktor, "A linear and non-linear polynomial neural network modeling of dissolved oxygen content in surface water: Inter- and extrapolation performance with inputs' significance analysis" in Science of the Total Environment, 610 (2018):1038-1046, https://doi.org/10.1016/j.scitotenv.2017.08.192 . .