Skip to Main content Skip to Navigation

Assimilation de données pour l'estimation de l'état hydraulique d'un aménagement hydroélectrique du Rhône équipé de la commande prédictive

Nelly Jean-Baptiste Dit Parny 1
1 LAAS-MAC - Équipe Méthodes et Algorithmes en Commande
LAAS - Laboratoire d'analyse et d'architecture des systèmes
Abstract : Electricity producer, CNR operates 19 dams and 19 hydroelectric power plants located along the French Rhone River. Most of the development projects are equipped with predictive control. This kind of control method has significantly improved the management of these development projects. However, there are situations that damage the calculation of this command. The aim of this research is to develop a methodology and algorithms to detect sources of errors, correct them and estimate the hydraulic states. The calculation of the predictive control occurs with each new acquisition of the observed data. For this, the stochastic data assimilation method as the Kalman filter seems to be the most suitable. The strong hypothesis of this method is the need to use a linear model, although the system studied is nonlinear. Thus, analyzing the impact of the linearization of the model and determining the application limits, was necessary. To ensure the effectiveness of this method, a study of the convergence of the filter was conducted. Then the observability and detectability notions were addressed to determine the sufficient hypothesis for convergence of the filter. To illustrate these studies, tests on twin experiments were conducted to see if the Kalman filter plays its role as a state estimator. These tests have been done for different kinds of situations which are considered as problematic for the regulation. The originality of this work was also to deal with real scenarios on the test platform for the regulation of the CNR. This tool has been used to make a comparison between the method of updating the observation that is currently used to regulate development projects of the Rhone and the one studied in this thesis.
Document type :
Complete list of metadata

Cited literature [50 references]  Display  Hide  Download
Contributor : Arlette Evrard <>
Submitted on : Monday, September 24, 2012 - 2:41:59 PM
Last modification on : Thursday, June 10, 2021 - 3:04:27 AM
Long-term archiving on: : Friday, December 16, 2016 - 4:39:58 PM


  • HAL Id : tel-00734758, version 1


Nelly Jean-Baptiste Dit Parny. Assimilation de données pour l'estimation de l'état hydraulique d'un aménagement hydroélectrique du Rhône équipé de la commande prédictive. Automatique. Université Paul Sabatier - Toulouse III, 2011. Français. ⟨tel-00734758⟩



Record views


Files downloads