TemporalFlowViz: Parameter-Aware Visual Analytics for Interpreting Scramjet Combustion Evolution

2025-09-05

AI Summary

Hypersonic scramjet combustion exhibits highly unsteady, multiscale dynamics, rendering large-scale temporal flow-field data challenging to interpret visually and compare across experimental cases. To address this, we propose a parameter-aware visual analytics framework: (1) pretrained Vision Transformers extract high-dimensional features from flow-field images; (2) dimensionality reduction, density-based clustering, and temporal trajectory modeling jointly enable automatic discovery of combustion patterns; and (3) a semantics-guided vision-language model generates interpretable natural-language descriptions, bridging latent-space representations with expert domain knowledge. The framework supports combustion-mode identification, operational-condition comparison, and physics-informed hypothesis generation. Evaluated on real hypersonic experimental data, it demonstrates effectiveness in explainable knowledge discovery—yielding actionable insights into underlying combustion mechanisms. This work establishes a novel paradigm for systematic, interpretable mechanistic analysis of complex propulsion phenomena.

Technology Category

• Computer Vision: Multi-modal Vision
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Understanding the complex combustion dynamics within scramjet engines is critical for advancing high-speed propulsion technologies. However, the large scale and high dimensionality of simulation-generated temporal flow field data present significant challenges for visual interpretation, feature differentiation, and cross-case comparison. In this paper, we present TemporalFlowViz, a parameter-aware visual analytics workflow and system designed to support expert-driven clustering, visualization, and interpretation of temporal flow fields from scramjet combustion simulations. Our approach leverages hundreds of simulated combustion cases with varying initial conditions, each producing time-sequenced flow field images. We use pretrained Vision Transformers to extract high-dimensional embeddings from these frames, apply dimensionality reduction and density-based clustering to uncover latent combustion modes, and construct temporal trajectories in the embedding space to track the evolution of each simulation over time. To bridge the gap between latent representations and expert reasoning, domain specialists annotate representative cluster centroids with descriptive labels. These annotations are used as contextual prompts for a vision-language model, which generates natural-language summaries for individual frames and full simulation cases. The system also supports parameter-based filtering, similarity-based case retrieval, and coordinated multi-view exploration to facilitate in-depth analysis. We demonstrate the effectiveness of TemporalFlowViz through two expert-informed case studies and expert feedback, showing TemporalFlowViz enhances hypothesis generation, supports interpretable pattern discovery, and enhances knowledge discovery in large-scale scramjet combustion analysis.

Problem

• Analyzing large-scale temporal flow field data from scramjet simulations
• Overcoming challenges in visual interpretation and cross-case comparison
• Bridging latent representations with expert reasoning for combustion dynamics

Innovation

• Pretrained Vision Transformers extract high-dimensional embeddings
• Dimensionality reduction and clustering uncover latent combustion modes
• Vision-language model generates natural-language summaries from annotations