3 présentations

• 
  15h30 - 15h55

  Optimization of energy consumption in multi-unit residential heating systems under economic demand response programs.

  • Shaival Nagarsheth, prés., Universite du Quebec a Trois-Rivieres
  • David Camilo Toquica Cardenas, Universite du Quebec a Trois-Rivieres
  • Kodjo Agbossou, Universite du Quebec a Trois-Rivieres
  • Nilson Henao, Universite du Quebec a Trois-Rivieres

  Economic demand response (DR) programs are aimed at elastic users who control their loads to modulate consumption patterns. In cold weather regions, controlling space heating systems represents a sizeable opportunity for DR, as they leverage the inner thermal capacitance of buildings to shift consumption without greatly affecting users' comfort. Multi-unit residential heating is especially appealing due to its high energy demand and controllability options. This work proposes an optimization problem for managing a multi-unit heating system considering a day-ahead DR program. The problem is formulated for a centralized controller with complete information and system observability. Then, the thermal dynamics of an entire building are represented in a state-space model to keep problem tractability and convexity. Besides, a fuzzy logic set is added to distribute the energy requirements among building units based on inward heat transfers and users' comfort valuations. The proposed approach is tested on a case study with actual data from Trois-Rivieres, QC. CA. The results expose the demand-side flexibility potential of multi-unit heating systems. Thus, future works can elaborate on DR strategies relying on trustworthy demand control capabilities.

• 
  15h55 - 16h20

  System and Control Theoretic Considerations for Power Network Optimization: Low Inertia Stability and DC Line Placement

  • Mohammad Pirani, prés., University of Ottawa

  This presentation explores two optimization challenges in power networks, focusing on generation and transmission. The first part focuses on the challenge of low inertia stability in power networks. With the transition from bulk synchronous generators to distributed generation connected via power electronics, system inertia diminishes, raising concerns about grid stability. While various metrics exist for assessing grid stability and performance, there is a lack of theoretical comparisons among these metrics. We will discuss a rigorous system-theoretic explanation of performance metrics for low-inertia stability. Our findings reveal that optimizing inertia based on different system norms can yield conflicting outcomes relative to system inertia, underscoring the significant influence of the chosen performance metric on low-inertia performance. The second part of the presentation addresses the optimal placement of DC lines within an AC grid. We investigate how the presence of DC lines impacts the controllability of AC power networks. By analyzing the spectrum of the controllability Gramian with the addition of DC lines, we demonstrate that maximizing the trace of the controllability Gramian, indicative of average controllability, occurs when DC lines connect node pairs with largest effective reactance in the AC network.

• 
  16h20 - 16h45

  Online Interior-point Methods for Time-Varying Equality-constrained Optimization

  • Jean-Luc Lupien, prés., UC Berkeley
  • Antoine Lesage-Landry, Polytechnique Montreal
  • Iman Shames, Australian National University

  An important challenge in the online convex optimization (OCO) setting is to incorporate generalized inequalities and time-varying constraints. The inclusion of constraints in OCO widens the applicability of such algorithms to dynamic but safety-critical settings such as the online optimal power flow (OPF) problem. In this work, we propose the first projection-free OCO algorithm admitting time-varying linear constraints and convex generalized inequalities: the online interior-point method for time-varying equality constraints (OIPM-TEC). We derive simultaneous sublinear dynamic regret and constraint violation bounds for OIPM-TEC under standard assumptions. For applications where a given tolerance around optima is accepted, we propose a new OCO performance metric – the epsilon-regret – and a more computationally efficient algorithm, the epsilon OIPM-TEC, that possesses sublinear bounds under this metric. Finally, we showcase the performance of these two algorithms on an online OPF problem and compare them to another OCO algorithm from the literature.