Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling

ISBN-13: 9781138027466 / Angielski / Miękka / 2015 / 200 str.

Opis

This study concentrates on: (i) committee modelling of hydrological models, (ii) hybrid committee hydrological models, (iii) influence of sampling strategies on prediction uncertainty of hydrological models, (iv) uncertainty prediction using machine learning techniques, (v) committee of predictive uncertainty models and (vi) flood inundation model and their uncertainty. This thesis is a contribution to hydroinformatics, which aims to connect various scientific disciplines: hydrological modelling, hydrodynamic modelling, multi-model averaging, data driven models, hybrid hydrological models, uncertainty analysis and high performance computing. The conclusions drawn allow for advancing the theory and practice of hydrological and integrated modelling. The developed software components is made available for public use and can be used by the researchers and practitioners to advance the mentioned areas further.

Spis treści

Summary
1 Introduction
2 Conceptual and data-driven hydrological modelling
3 Committees of hydrological models
4 Hybrid committees of hydrological models
5 Model parametric uncertainty and effects of sampling strategies
6 Prediction of uncertainty by machine learning techniques
7 Committees of models predicting models' uncertainty
8 Integration of hydrological and hydrodynamic models and their uncertainty in inundation modelling
9 Conclusions and recommendations

O autorze

Nagendra Kayastha graduated in Civil Engineering (MSc) from the St. Petersburg State University of Means of Communication, St. Petersburg, Russia, in 1997, with a specialization in Bridge and Tunnel Engineering. He was engaged as a consulting engineer in Morrison Knudson International Inc.(USA), for the Kaligandaki ‘A’ Hydropower Project in western Nepal. After completion of this project, he continued working at a consulting company in Nepal and was assigned to various national and international water-related projects. In 2005, he joined the MSc degree programme in Water Science and Engineering, specializing in Hydroinformatics at UNESCO-IHE Institute for Water Education, Delft, The Netherlands. His MSc research topic was on the "Novel approaches to uncertainty analysis of hydrological model" which covered new methods for uncertainty prediction of hydrological models using machine learning techniques. After completion of his study, he joined the special research programme at Hydroinformatics and Knowledge Management Department. He was involved in projects of the Delft Cluster Research Programme and in the EU project FLOODsite. He joined the PhD programme of the UNESCO-IHE and the Delft University of Technology, under the supervision of Professor Dimitri Solomatine with the co-supervision of Professor Ann van Griensven (Vrije Universiteit, Brussels) in 2010. He contributed to several research projects, namely EnviroGRIDs, WeSenseIT, and MyWater and has assisted in the Master programme in Hydroinformatics. He published more than 15 technical papers in international journals and conferences.