The present paper aims at modeling suspended sediment load(SSL) using heuristic data driven methodologies, e.g. Gene Expression Programming(GEP) and Support Vector Machine(SVM) in three successive hydrometric stations...The present paper aims at modeling suspended sediment load(SSL) using heuristic data driven methodologies, e.g. Gene Expression Programming(GEP) and Support Vector Machine(SVM) in three successive hydrometric stations of Housatonic River in U.S. The simulations were carried out through local and cross-station data management scenarios to investigate the interrelations between the SSL values of upstream/downstream stations. The available scenarios were applied to predict SSL values using GEP to obtain the best models. Then, the best models were predicted by SVM approach and the obtained results were compared with those of GEP. The comparison of the results revealed that the SVM technique is more capable than the GEP for modeling the SSL through the both local and cross-station data management strategies. Besides, local application seems to be better than cross-station application for modeling SSL. Nevertheless, the cross-station application demonstrated to be a valid methodology for simulating SSL, which would be of interest for the stations with lack of observational data. Also, the prediction capability of conventional Sediment Rating Curve(SRC) method was compared with those of GEPand SVM techniques. The obtained results revealed the superiority of GEP and SVM-based models over the traditional SRC technique in the studied stations.展开更多
The present investigation has been fulfilled with a view to the future nuclear power station constriction near the Belene Island.Sediment loads and sediment transportation are important factors for the technical water...The present investigation has been fulfilled with a view to the future nuclear power station constriction near the Belene Island.Sediment loads and sediment transportation are important factors for the technical water supply of the nuclear station Belene.The paper deals with change of suspended sediment load at Bulgarian part of the Danube River downstream of the Iron Gates.Recent data on suspended sediment loads for the hydrometric gauge station at Svishtov have been collected,computed and presented.The results obtained discover the time variability of the sediment loads,climate change and the anthropogenic impact on the suspended sediments.The tendency of alteration and inter-annual variability of the suspended sediments are investigated and characteristics of average annual,monthly and maximal values are shown.In the material archive granulometric data are presented for this part of the Bulgarian stretch before the Iron Gates construction in the conditions of the natural sediment regime.There are statistical parameters of the studied characteristics of the river turbidity and discharge before and after the anthropogenic impact.The increase of the absolute minimal turbidity has been determined after the Iron Gate I as a result of activation of the bed processes and hydro-morphological changes.展开更多
文摘The present paper aims at modeling suspended sediment load(SSL) using heuristic data driven methodologies, e.g. Gene Expression Programming(GEP) and Support Vector Machine(SVM) in three successive hydrometric stations of Housatonic River in U.S. The simulations were carried out through local and cross-station data management scenarios to investigate the interrelations between the SSL values of upstream/downstream stations. The available scenarios were applied to predict SSL values using GEP to obtain the best models. Then, the best models were predicted by SVM approach and the obtained results were compared with those of GEP. The comparison of the results revealed that the SVM technique is more capable than the GEP for modeling the SSL through the both local and cross-station data management strategies. Besides, local application seems to be better than cross-station application for modeling SSL. Nevertheless, the cross-station application demonstrated to be a valid methodology for simulating SSL, which would be of interest for the stations with lack of observational data. Also, the prediction capability of conventional Sediment Rating Curve(SRC) method was compared with those of GEPand SVM techniques. The obtained results revealed the superiority of GEP and SVM-based models over the traditional SRC technique in the studied stations.
文摘The present investigation has been fulfilled with a view to the future nuclear power station constriction near the Belene Island.Sediment loads and sediment transportation are important factors for the technical water supply of the nuclear station Belene.The paper deals with change of suspended sediment load at Bulgarian part of the Danube River downstream of the Iron Gates.Recent data on suspended sediment loads for the hydrometric gauge station at Svishtov have been collected,computed and presented.The results obtained discover the time variability of the sediment loads,climate change and the anthropogenic impact on the suspended sediments.The tendency of alteration and inter-annual variability of the suspended sediments are investigated and characteristics of average annual,monthly and maximal values are shown.In the material archive granulometric data are presented for this part of the Bulgarian stretch before the Iron Gates construction in the conditions of the natural sediment regime.There are statistical parameters of the studied characteristics of the river turbidity and discharge before and after the anthropogenic impact.The increase of the absolute minimal turbidity has been determined after the Iron Gate I as a result of activation of the bed processes and hydro-morphological changes.
