Poster (Scientific congresses and symposiums)
Bayesian methods for predicting and modelling winter wheat biomass
Mansouri, Majdi; Dumont, Benjamin; Destain, Marie-France
2014Modelling climate change impacts on crop production for food security
 

Files


Full Text
Poster_Oslo.pdf
Publisher postprint (140.85 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Crop model; Bayesian methods; Particle Filters
Abstract :
[en] The objectives of this paper are threefold. The first objective is to propose to use an Improved Particle Filtering (IPF) based on minimizing Kullback-Leibler divergence for crop models' predictions. The performances of the proposed technique are compared with those of the conventional Particle Filtering (PF) for improving nonlinear crop model predictions. The main novelty of this task is to develop a Bayesian algorithm for nonlinear and non-Gaussian state and parameter estimation with better proposal distribution. The second objective is to investigate the effects of practical challenges on the performances of state estimation algorithms PF and IPF. Such practical challenges include (i) the effect of measurement noise on the estimation performances and (ii) the number of states and parameters to be estimated. The third objective is to use the state estimation techniques PF and IPF for updating prediction of nonlinear crop model in order to predict winter wheat biomass. PF and IPF are applied at a dynamic crop model with the aim to predict a state variable, namely the winter wheat biomass, and to estimate several model parameters. Furthermore, the effect of measurement noise (e.g., different signal-to-noise ratios) on the performances of PF and IPF is investigated. The results of the comparative studies show that the IPF provides a significant improvement over the PF because, unlike the PF which depends on the choice of sampling distribution used to estimate the posterior distribution, the IPF yields an optimum choice of the sampling distribution, which also accounts for the observed data.
Disciplines :
Engineering, computing & technology: Multidisciplinary, general & others
Author, co-author :
Mansouri, Majdi ;  Université de Liège - ULiège > Sciences et technologie de l'environnement > Agriculture de précision
Dumont, Benjamin  ;  Université de Liège - ULiège > Sciences et technologie de l'environnement > Agriculture de précision
Destain, Marie-France ;  Université de Liège - ULiège > Sciences et technologie de l'environnement > Agriculture de précision
Language :
English
Title :
Bayesian methods for predicting and modelling winter wheat biomass
Publication date :
February 2014
Event name :
Modelling climate change impacts on crop production for food security
Event organizer :
FACCE JPI Knowledge Hub
Event place :
Oslo, Norway
Event date :
10-12 February 2014
Audience :
International
Name of the research project :
Filtering method-based state and parameter estimation for crop models
Funders :
F.R.S.-FNRS - Fonds de la Recherche Scientifique [BE]
Available on ORBi :
since 30 January 2014

Statistics


Number of views
114 (17 by ULiège)
Number of downloads
107 (6 by ULiège)

Bibliography


Similar publications



Contact ORBi