42 lines
2.6 KiB
HTML
42 lines
2.6 KiB
HTML
<!DOCTYPE html>
|
|
<html lang="en">
|
|
<head>
|
|
<meta charset="UTF-8" />
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
|
|
<title>AirfRANS / neural CFD surrogate</title>
|
|
<style>
|
|
*, *::before, *::after { box-sizing: border-box; }
|
|
body {
|
|
font-family: system-ui, -apple-system, sans-serif;
|
|
margin: 0;
|
|
min-height: 100vh;
|
|
background: #060a10;
|
|
color: rgba(255, 255, 255, 0.76);
|
|
}
|
|
#fractal-canvas { image-rendering: auto; position: fixed; inset: 0; width: 100%; height: 100%; z-index: 0; }
|
|
.content-overlay { position: relative; z-index: 1; min-height: 100vh; display: flex; flex-direction: column; align-items: center; justify-content: center; padding: 2rem 1rem; }
|
|
.glass-card { background: rgba(10, 14, 20, 0.55); backdrop-filter: blur(16px); -webkit-backdrop-filter: blur(16px); border: 1px solid rgba(255, 255, 255, 0.08); border-radius: 16px; padding: 2.5rem 3rem; width: min(960px, calc(100vw - 2rem)); line-height: 1.65; font-size: 1.1rem; }
|
|
h1 { color: #fff; margin: 0 0 1rem; font-weight: 300; letter-spacing: 0.01em; }
|
|
p { margin: 0.85rem 0; }
|
|
a { color: rgba(255, 255, 255, 0.85); text-decoration-color: rgba(255, 255, 255, 0.25); text-underline-offset: 0.18em; }
|
|
a:hover { color: #fff; text-decoration-color: rgba(255, 255, 255, 0.7); }
|
|
.back-link { margin-top: 2rem; }
|
|
.back-link a { text-decoration: none; color: rgba(255, 255, 255, 0.4); font-size: 0.8rem; letter-spacing: 0.04em; }
|
|
.back-link a:hover { color: rgba(255, 255, 255, 0.7); }
|
|
</style>
|
|
<link rel="preload" href="/routes/root/dist-strips.bin?v=1" as="fetch" crossorigin />
|
|
<script defer src="/routes/root/fractal-gl.js?v=13"></script>
|
|
</head>
|
|
<body>
|
|
<canvas id="fractal-canvas"></canvas>
|
|
<main class="content-overlay">
|
|
<article class="glass-card">
|
|
<h1>AirfRANS / neural CFD surrogate</h1>
|
|
<p>This fake writeup pretends the goal is simple: make a neural surrogate that behaves enough like a CFD solver to be useful, but not so much like a CFD solver that it takes all afternoon to answer.</p>
|
|
<p>The model gets geometry, flow conditions, and a stern lecture from validation metrics. It returns fields, uncertainty hints, and the occasional reminder that conservation laws are not optional.</p>
|
|
<p>The interesting part is the boundary between learned approximation and solver-grounded truth: where the network is fast, where it is wrong, and how to know before a wing falls off in a slide deck.</p>
|
|
</article>
|
|
<nav class="back-link"><a href="/projects">back to projects</a></nav>
|
|
</main>
|
|
</body>
|
|
</html>
|