Please use this identifier to cite or link to this item: https://idr.nitk.ac.in/jspui/handle/123456789/8872
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSreedhara, B.M.
dc.contributor.authorKuntoji, G.
dc.contributor.authorManu
dc.contributor.authorMandal, S.
dc.date.accessioned2020-03-30T10:22:55Z-
dc.date.available2020-03-30T10:22:55Z-
dc.date.issued2019
dc.identifier.citationAdvances in Intelligent Systems and Computing, 2019, Vol.741, , pp.383-392en_US
dc.identifier.urihttp://idr.nitk.ac.in/jspui/handle/123456789/8872-
dc.description.abstractScour is one of the major factors which affects directly on the durability and safety of the Bridge abutments. Based on the experimental data of Goswami in 2012, an effort is made to predict local scour by using a hybrid approach of Swarm Intelligence based algorithms which is today one of the powerful tools of optimization techniques. In this work, an intelligent model based on support vector machine in combination with the particle swarm optimization (PSO-SVM) technique is developed. The PSO-SVM models are developed with RBF, Polynomial and Linear kernel functions. The circular, rectangular, round-nosed, and sharp-nosed shapes of piers are considered in live bed scour condition. The scour depth around bridge piers is predicted by considering Sediment size, flow velocity, and time of flow as input parameters. Prediction accuracy of the models is evaluated using the model performance indicators such as Root Mean Square Error (RMSE, Correlation Coefficient (CC), Nash Succlift Error (NSE), etc. The results obtained from the model are compared with the measured scour depth to validate the reliability of the hybrid model. Based on the results, PSO based SVM model is found to be successful, reliable, and efficient in predicting the scour depth around the bridge pier. � Springer Nature Singapore Pte Ltd. 2019.en_US
dc.titlePSO-SVM approach in the prediction of scour depth around different shapes of bridge pier in live bed scour conditionen_US
dc.typeBook chapteren_US
Appears in Collections:2. Conference Papers

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.