arXiv subjects

Boxun Huang

Publications and source records attributed to Boxun Huang.

2 recordsLinked to original sources

arXiv2026

Harnessing GPU Acceleration in Large-Scale Process Optimization

This paper presents a proof-of-concept workflow for equation-oriented process optimization that runs entirely on a GPU. Process optimization models often incorporate complex interconnected unit operations, dynamics, and uncertainties, resulting in large nonlinear programs that can be computationally demanding for conventional CPU-based solvers. Although emerging GPU-based solvers offer substantial computational benefits, their application to process optimization has been limited by the lack of GPU-compatible process modeling tools. We address this gap by prototyping the GPU-compatible process optimization models using an existing GPU-capable optimization software stack, including ExaModels (algebraic modeling system), MadNLP (optimization solver), and cuDSS (linear solver). ExaModels formulates the process optimization problem in a GPU-compatible way by exposing its repeated algebraic structure, while MadNLP and cuDSS solve the resulting nonlinear program on the GPU. This workflow is demonstrated on a CO2 absorber design problem under feed uncertainty, in which a shared column diameter is minimized subject to equilibrium and hydraulic constraints in all scenarios. For the largest case with 5,000 scenarios and 1.5 million variables, the GPU workflow achieves a speedup of approximately 21\times over a single-threaded CPU baseline using JuMP, Ipopt, and MA57.

math.OC↗

arXiv2026

Design of Carbon Capture Processes Under Part-load Operating Conditions

Solvent-based carbon capture can reduce CO2 emissions resulting from a continued reliance on fossil power plants for firm power. These capture processes remove CO2 from flue gases via a solvent. Careful design via process systems optimization can limit the overall cost of carbon capture, which is both capital- and energy intensive. As dispatchable power plants operate to meet varying load demand, the design process needs to account for varying operating points. However, optimizing the design over multiple operating points yields high computational complexity, which is why designs are often based on a single operating point in practice. Here, we identify optimal carbon capture process designs via stochastic optimization, reducing computational complexity through a data-driven approach-to-equilibrium model of the absorption and desorption processes. We represent variable flue gas conditions based on part-load operation data of a representative coal power plant. Accounting for this variability in the design substantially reduces equipment size and total plant cost by 6-9 % at the expense higher operating costs, yielding a reduction in total cost of carbon capture by 0.7-1.7 %. Given the capital intensity of carbon capture, variability of flue gas conditions therefore should be considered at the design stage, particularly if capture is deployed on plants subject to load following.

math.OC↗