| Autor: |
Debabrata Datta, U. K. Paul, Subrata Bera, Avinash J. Gaikwad, Dhanesh B. Nagrale |
| Rok vydání: |
2018 |
| Předmět: |
|
| Zdroj: |
Advances in Intelligent Systems and Computing ISBN: 9783319748078 |
| DOI: |
10.1007/978-3-319-74808-5_54 |
| Popis: |
Extreme value analysis is important for designing the engineering structures robust enough to withstand external hazards such as wind load, flood level, earthquake etc. Important use of routinely measured station data is made to obtain year wise maximum data of the required variable specifically related to an external hazard. As per common practice, the year wise extreme data are fitted with generalised extreme value distribution function to make predictions for various return periods. Model uncertainty with respect to the variation of model parameters is also estimated. Multiple models are developed for data from the measuring stations. A methodology for statistical aggregation of multiple models is developed and demonstrated considering data from four measuring stations. In this statistical aggregation method, the statistical property of the GEV model has been preserved. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
