Paper introduces uncertainty-guided surrogate for microfluidic droplet generator inverse design
Original title: Microdroplet Generator Surrogate Enhancement and Inverse Design Through an Uncertainty-Guided Framework
Neural ensemble surrogates with targeted enrichment and PSO achieve nearly doubled accuracy for optimizing microdroplet devices.
The work presents a unified methodology coupling adaptive surrogate enhancement with simulation-driven inverse design for microfluidic devices. High-fidelity Volume-of-Fluid simulations generate ground-truth data while neural network ensembles predict flow regimes and droplet diameters. Uncertainty quantification identifies epistemic uncertainty at regime boundaries, guiding physics-informed enrichment via additional simulations. Particle Swarm Optimization enables constrained inverse design.
- Uncertainty concentrates near regime transition boundaries
- Fraction of predictions within 0-3% error nearly doubles
- High-error cases above 10% drop substantially
- Framework applicable to broad range of microfluidic systems
From the original article
1 min read at arxiv.orgMicrofluidic droplet generation is critical for many applications, yet efficient device design in high-dimensional parameter spaces remains computationally prohibitive due to complex multiphase flow dynamics. This work presents a unified methodology that couples adaptive surrogate enhancement with simulation-driven inverse design.
By David Gabriel Harispe · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.