ThermoSysPro (TSP) is a library for the modeling and simulation of power plants and energy systems. It has been developed by EDF and it is released under open source license. When developing models with TSP it is necessary to ensure that they match reality. In practice, this operation is performed by adjusting the value of the parameters appearing in the model. This major step corresponds to model calibration. Calibration can be performed through various methods. A classical way to do so with Modelica models is by model inversion. The major inconvenience of this method, in addition of potential convergence problems for complex models, is that it is necessary to have exactly the same number of measurements as parameters to be calibrated, which is not often the case in practice. This paper shows how data assimilation techniques can robustly be used for calibration of complex TSP models avoiding the inconveniences associated to calibration by model inversion while ensuring an optimal use of the available measurements. A complex TSP model of the secondary loop of a Pressurized Water Reactor (PWR) is considered for this purpose.
To get reliable simulation results from a Modelica model it is important to parametrize and initialize the model using the best estimate of the state of the system. Commonly, this state estimation is done by inverse calculation on a square system of equations requiring as many known values as states to be computed. In practice this constraint is an important limitation and, in addition, this method does not provide any information on the uncertainties or confidence level associated to the estimated state.Taking advantage of the mathematical formulation of Modelica equations, this paper presents a new method to cope with the difficulties associated to the inverse calculation method. This approach adapts and extends the framework of data assimilation to provide a fullyintegrated Modelica tool, which efficiently can handle every type of state estimation problem for static models. This method has been successfully tested with simple and complex Modelica models. Finally, the Modelica implementation of this technique allows to easily extend it to further applications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.