AI & ML
Control-Oriented Modeling of Heat Exchangers with Improved Low-Flow Dynamics
arXiv eess.SY0 views
arXiv:2610.41364v1 Announce Type: new Abstract: Existing low-order models of heat exchangers can lose dynamic accuracy at low flow rates, when fluid residence times become significant. This paper proposes two control-oriented models that improve the low-flow performance. The first extends a finite-volume formulation with the first spatial temperature moment, whereas the second uses Chebyshev--Lobatto collocation. Both capture the coupled heat-transfer and transport dynamics and improve the performance when the system behavior is dominated by the inlet-temperature dynamics. At equal state dimension, comparisons with a high-order reference show more accurate delayed outlet-temperature responses while retaining steady-state accuracy.
Source:https://arxiv.org/abs/2610.41364PreviousMulti-Timescale Deliverability of Shared Hybrid Storage for AI Data Center CampusesNextRiemannian Optimization on Unknown Manifolds via Stochastic Proximal Refinement
Related posts
Kalshi court losses pile up as judges refuse to block state gambling laws
0Ars Technica

Tracking Vietnamese news trends with embeddings, HDBSCAN and an LLM
0dev.to
How to Organize and Improve Your AI Prompts Over Time
0dev.to
The Hackintosh Reborn
0Hacker News

Formally verifying agent-written full-stack web apps with PDD Kit
0Microsoft Research
I built an Android app with AI. Does that count as vibe coding?
0dev.to Discuss