arXiv subjects

Matthew Brun

Publications and source records attributed to Matthew Brun.

5 recordsLinked to original sources

arXiv2026

Modeling Adversarial Wildfires for Power Grid Disruption

Electric power infrastructure faces increasing risk of damage and disruption due to wildfire. Operators of power grids in wildfire-prone regions must consider the potential impacts of unpredictable fires. However, traditional wildfire models do not effectively describe worst-case, or even high-impact, fire behavior. To address this issue, we propose a mixed-integer conic program to characterize an adversarial wildfire that targets infrastructure while respecting realistic fire spread dynamics. We design a wind-assisted fire spread set based on the Rothermel fire spread model and propose principled convex relaxations of this set, including a new relaxation of the inner product over Euclidean balls. We present test cases derived from the recent Park, Eaton, and Palisades fires in California and solve models to identify the minimum time-to-outage of multiple-element contingencies and the maximum load shed associated with a sequence of element outages caused by a realistic wildfire. We use the minimum time-to-outage values to screen contingencies and construct security-constrained optimal power flow models that promote operational resilience against wildfire.

math.OC↗

arXiv2025

Dispatch-Aware Deep Neural Network for Optimal Transmission Switching

Optimal transmission switching (OTS) improves optimal power flow (OPF) by selectively opening transmission lines, but its mixed-integer formulation increases computational complexity, especially on large grids. To address this, we propose a dispatch-aware deep neural network (DA-DNN) that accelerates DC-OTS without relying on pre-solved labels, eliminating costly OTS label generation that becomes impractical at scale. DA-DNN predicts line states and passes them through an embedded differentiable DC-OPF layer, using the resulting generation cost as the loss function so that physical network constraints are enforced throughout training and inference. To stabilize training, we adopt a customized weight and bias initialization that keeps the embedded DC-OPF feasible from the first epoch. To improve inference robustness, we incorporate a binary regularization term that reduces ambiguity in the relaxed line-status outputs prior to thresholding. Once trained, DA-DNN produces a feasible topology and dispatch pair with highly predictable computation time comparable to a single DC-OPF solve, while conventional MIP solvers can become intractable. Moreover, the embedded OPF layer enables DA-DNN to generalize to untrained system configurations, such as changes in line flow limits, and to support post-contingency corrective operation. As a result, the proposed method captures the economic advantages of OTS while maintaining scalability and generalization ability.

eess.SY↗

arXiv2025

Alternating Methods for Large-Scale AC Optimal Power Flow with Unit Commitment

Security-constrained unit commitment with alternating current optimal power flow (SCUC-ACOPF) is a central problem in power grid operations that optimizes commitment and dispatch of generators under a physically accurate power transmission model while encouraging robustness against component failures. SCUC-ACOPF requires solving large-scale problems that involve multiple time periods and networks with thousands of buses within strict time limits. In this work, we study a detailed SCUC-ACOPF model with a rich set of features of modern power grids, including price-sensitive load, reserve products, transformer controls, and energy-limited devices. We propose a decomposition scheme and a penalty alternating direction method to find high-quality solutions to this model. Our methodology leverages spatial and temporal decomposition, separating the problem into a set of mixed-integer linear programs for each bus and a set of continuous nonlinear programs for each time period. To improve the performance of the algorithm, we introduce a variety of heuristics, including restrictions of temporal linking constraints, a second-order cone relaxation, and a contingency screening algorithm. We quantify the quality of feasible solutions through a dual bound from a convex second-order cone program. To evaluate our algorithm, we use large-scale test cases from Challenge 3 of the U.S. Department of Energy's Grid Optimization Competition that resemble real power grid data under a variety of operating conditions and decision horizons. The experiments yield feasible solutions with an average optimality gap of 1.33%, demonstrating that this approach generates near-optimal solutions within stringent time limits.

math.OC↗

arXiv2024

BattOpt: Optimal Facility Planning for Electric Vehicle Battery Recycling

The electric vehicle (EV) battery supply chain will face challenges in sourcing scarce and expensive minerals required for manufacturing and in disposing of hazardous retired batteries. Integrating recycling technology into the supply chain has the potential to alleviate these issues; however, players in the battery market must design investment plans for recycling facilities. In this paper, we propose a multistage stochastic optimization model for computing minimum cost recycling capacity decisions, in which retired batteries are recycled and recovered materials are used to manufacture new batteries. We transform EverBatt, a leading evaluation framework for battery recycling cost and environmental impact, into a prescriptive decision-making tool for determining optimal investment strategies. Our model is a separable concave minimization subject to linear constraints, a class for which we describe a finitely convergent global optimization algorithm based on piecewise linear approximation that solves up to 14x faster than comparable algorithms. We propose an equivalent reformulation of the model that eliminates variables by introducing integrality constraints, drastically reducing solve times. To produce a set of operational scenarios, we generate time-series projections for new battery demand, retired battery supply, and material costs, leveraging state-of-the-art econometric models. We apply our model to compute optimal investment plans and quantify the impact of optimal decision-making: optimal investment can reduce battery manufacturing costs by 22% and environmental impacts by 26%. We measure the operational benefits of different recycling technologies and demonstrate how high uncertainty motivates greater investment in recycling capacity. Finally, we analyze the impact of policy instruments to reveal key insights on the efficacy of capacity grants and production credits.

math.OC↗

arXiv2022

Leveraging mid-infrared spectroscopic imaging and deep learning for tissue subtype classification in ovarian cancer

Mid-infrared spectroscopic imaging (MIRSI) is an emerging class of label-free techniques being leveraged for digital histopathology. Modern histopathologic identification of ovarian cancer involves tissue staining followed by morphological pattern recognition. This process is time-consuming, subjective, and requires extensive expertise. This paper presents the first label-free, quantitative, and automated histological recognition of ovarian tissue subtypes using a new MIRSI technique. This technique, called optical photothermal infrared (O-PTIR) imaging, provides a 10X enhancement in spatial resolution relative to prior instruments. It enables sub-cellular spectroscopic investigation of tissue at biochemically important fingerprint wavelengths. We demonstrate that enhanced resolution of sub-cellular features, combined with spectroscopic information, enables reliable classification of ovarian cell subtypes achieving a classification accuracy of 0.98. Moreover, we present statistically robust validation from 74 patient samples with over 60 million data points. We show that sub-cellular resolution from five wavenumbers is sufficient to outperform state-of-the-art diffraction-limited techniques from up to 235 wavenumbers. We also propose two quantitative biomarkers based on the relative quantities of epithelium and stroma that exhibits efficacy in early cancer diagnosis. This paper demonstrates that combining deep learning with intrinsic biochemical MIRSI measurements enables quantitative evaluation of cancerous tissue, improving the rigor and reproducibility of histopathology.

q-bio.TO↗