An efficient co-simulation and control approach to tackle complex multi-domain energetic systems: concepts and applications of the PEGASE platform

Mathieu Vallee; Roland Baviere; Valérie Seguin; Valéry Vuillerme; Nicolas Lamaison; Michael Nikhil Descamps; Antoine Aurousseau

arXiv · 2506.15195

An efficient co-simulation and control approach to tackle complex multi-domain energetic systems: concepts and applications of the PEGASE platform

Mathieu Vallee·Roland Baviere·Valérie Seguin·Valéry Vuillerme·Nicolas Lamaison·Michael Nikhil Descamps·Antoine Aurousseau

View original source ↗Read source document ↗Export metadata ↓

Abstract

In this paper, we present a novel research software, called PEGASE, suitable for the design, validation and deployment of advanced control strategies for complex multi-domain energy systems. PEGASE especially features a highly efficient cosimulation engine, together with integrated solutions for defining both rule-based control strategies and Model-Predictive Control (MPC). The main principle behind the PEGASE platform is divide-and-conquer. Indeed, rather than trying to solve a problem as a monolithic entity, which can be highly complex for multi-domain large-scale systems, it is often more efficient to decompose it into several domains or sub-problems, and to simulate them in a decoupled way. To provide its cosimulation capabilities, we based PEGASE on two main components. The first one is a framework for integrating simulation models, which can be either compatible with the FMI standard or interfaced through an Application Programming Interface (API). The second one is a multi-threaded sequencer enabling several simulation sequences with different time steps. To provide advanced control capabilities, we also equipped PEGASE with a framework for MPC combining a comprehensive management of predictions data and a modeler dedicated to the formulation of Mixed Integer Linear Programs. We implemented this framework in C++ providing low formulation and resolution times for typical applications. Connection to hardware is also available via standard industry protocols thereby allowing PEGASE to control real energy systems. In this paper, we show how these basic functionalities, combined with dedicated modeling tools, enable setting up simulation and control applications suitable for tackling the complexity of various kinds of energy systems. To illustrate this, we present four application examples from our recent research work. These examples cover several domains, from concentrated solar thermal plants to optimal control of district heating networks. The variety of examples demonstrates the robustness and genericity of the approach.

Explore related subjects

math.OC

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Mathieu Vallee, Roland Baviere, Valérie Seguin, Valéry Vuillerme, Nicolas Lamaison, Michael Nikhil Descamps, Antoine Aurousseau. 2025-06-18. An efficient co-simulation and control approach to tackle complex multi-domain energetic systems: concepts and applications of the PEGASE platform. https://arxiv.org/abs/2506.15195

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Genericity of Polyak-Lojasiewicz Inequalities for Entropic Mean-Field Neural ODEs

We address the behavior of idealized deep residual neural networks (ResNets), modeled via an optimal control problem set over continuity (or adjoint transport) equations. The continuity equations describe the statistical evolution of the features in the asymptotic regime where the layers of the network form a continuum. The velocity field is expressed through the network activation function, which is itself viewed as a function of the statistical distribution of the network parameters (weights and biases). From a mathematical standpoint, the control is interpreted in a relaxed sense, taking values in the space of probability measures over the set of parameters. We investigate the optimal behavior of the network when the cost functional arises from a regression problem and includes an additional entropic regularization term on the distribution of the parameters. In this framework, we focus in particular on the existence of stable optimizers --that is, optimizers at which the Hessian of the cost is non-degenerate. We show that, for an open and dense set of initial data, understood here as probability distributions over features and associated labels, there exists a unique stable global minimizer of the control problem. Moreover, we show that such minimizers satisfy a local Polyak--Lojasiewicz inequality, which can lead to exponential convergence of the corresponding gradient descent when the initialization lies sufficiently close to the optimal parameters. This result thus demonstrates the genericity (with respect to the distribution of features and labels) of the Polyak--Lojasiewicz condition in ResNets with a continuum of layers and under entropic penalization.

math.OC↗

arXiv2026

A regret minimization approach to fixed-point iterations

We propose a conversion scheme that turns regret minimizing algorithms into fixed point iterations, with convergence guarantees following from regret bounds. The resulting iterations can be seen as a grand extension of the classical Krasnoselskii--Mann iterations, as the latter are recovered by converting the Online Gradient Descent algorithm. This approach yields new simple iterations for finding fixed points of non-self operators. We also focus on converting algorithms from the AdaGrad family of regret minimizers, and thus obtain fixed point iterations with adaptive guarantees of a new kind. Numerical experiments on various problems demonstrate faster convergence of AdaGrad-based fixed point iterations over Krasnoselskii--Mann iterations.

math.OC↗

arXiv2026

Variational Analysis in Spectral Decomposition Systems

This work is concerned with the variational analysis of functions defined on Euclidean spaces whose values depend solely on certain invariants (``spectrum'') of their arguments, a class we term ``spectral functions.'' Building on our previous work \cite{PartI} on the convex analysis of such functions, we work in the abstract framework of spectral decomposition systems, which covers a wide range of previously studied settings, including eigenvalue decomposition of Hermitian matrices and singular value decomposition of rectangular matrices, and allows the derivation of new results in more general settings such as normal decomposition systems. The main results of this work provide constructive formulae for computing the regular, limiting, and Clarke subdifferentials of a spectral function in terms of the corresponding objects of the associated invariant function. Finally, we obtain a generalization of Lidski\uı's theorem on the spectrum of additive perturbations of Hermitian matrices to arbitrary spectral decomposition systems.

math.OC↗