API
A static API: every endpoint is a file under /api/ on this site, generated by scripts/build_api.py from the cleaned data and served by GitHub Pages. Plain HTTP GET, no keys, CORS-open (Access-Control-Allow-Origin: *).
- Base URL, always the newest release:
https://aij.eddy3d.com/api - One release, frozen:
https://aij.eddy3d.com/api/<release>, for example.../api/3.2.1. The files and paths are the same; see Versions and pinning. - Fallback, the same files raw from the repository:
https://raw.githubusercontent.com/Eddy3D-Dev/AIJ-Data-Validation/main/docs/api - Machine-readable:
/api/openapi.json(OpenAPI 3.1), a JSON Schema for every document under/api/schemas/, and the agent entry point/llms.txt.
Versions and pinning
Every release of the data and API is kept as a frozen copy of the whole API under /api/<release>/. The unversioned /api/ paths serve the newest release, and versions.json lists them all:
| Address | What it serves | When it changes |
|---|---|---|
/api/versions.json | every release, newest first: version, date, status, schema_version, data_version, base_url, tag, changelog | with each release |
/api/<release>/..., for example /api/3.2.1/index.json | exactly the files of that release | never, once the release is tagged |
/api/..., for example /api/index.json | the newest release (the latest alias) | with each release |
A release’s copy and the alias hold the same files at the same relative paths. They differ only in the address a document gives its own files. Inside a release’s copy every URL to a file of the API points back into the same copy: index.json says "tree": "release" and its base_url ends in the release number, and the $schema URL of every document, the ?v= download URLs, the servers of openapi.json and the endpoint URLs in llms.txt all start with that base_url. The one file outside it is versions.json (versions_url in index.json), which has a single home at the unversioned path because it changes with every release. A client that starts in a copy never leaves it for data. The alias says "tree": "latest" and names the copy it currently equals in release_base_url.
To pin a script to a release, use that release’s base URL and change nothing else.
LIVE=https://aij.eddy3d.com/api
BASE=$LIVE/3.2.1 # pinned to release 3.2.1; $LIVE alone follows the newest release
curl -s -A "aij-client/1.0" $BASE/index.json | jq '.versions' # release 3.2.1, whatever is newer
curl -s -A "aij-client/1.0" $BASE/cases/B.json | jq '.points.csv.url' # .../api/3.2.1/cases/B/points.csv?v=...
Cite the API with cite.api in index.json or the About page.
To find the releases, and to learn the base URL of one, GET /api/versions.json:
curl -s -A "aij-client/1.0" $LIVE/versions.json | jq -r '.versions[] | "\(.version) \(.status) \(.base_url)"'
Each entry of versions has:
| Field | Meaning |
|---|---|
version, date | the release (the repository’s VERSION, tag vX.Y.Z) and its date in CHANGELOG.md |
status | latest for the newest release, superseded for every older one |
schema_version, data_version | the JSON contract and the data fingerprint of that release |
base_url, index_url | where that release’s files are frozen, and its index.json. Fetch base_url + a path from index.json |
fallback_base_url | the same files raw from the repository at the release tag, or null |
self_contained | true if every API URL inside that release’s documents points back into its base_url; if false, build URLs from base_url and a path in its index.json instead of following the links |
tag, tag_url, changelog_url | the git tag, the GitHub release page and the changelog at that tag |
An address that names a base is one of four, depending on whether it is pinned and where it is served:
| On this site | Raw from the repository | |
|---|---|---|
| Pinned (never changes) | release_base_url; base_url inside a release’s copy | release.base_url; fallback_base_url inside a release’s copy |
| Latest | base_url in the alias | fallback_base_url in the alias |
Conventions
Every JSON document starts with the same four keys.
| Key | Meaning |
|---|---|
$schema | the URL of the document’s JSON Schema (draft 2020-12). The build validates every document against it. |
schema_version | the version of this JSON contract, semantic versioning (see Versions and changes). The release of the data is versions.release in index.json. |
data_version | a SHA-256 over every served measurement, geometry and score file. It changes when, and only when, the data changes. Quote it with any number you take from here. |
license | CC-BY-4.0, the SPDX identifier. Citation: About and cite. |
Downloads are objects, not bare URLs. Wherever a document points at a CSV, STL or ZIP file, it gives:
{"url": ".../api/cases/B/points.csv?v=e8214b704821", "path": "cases/B/points.csv",
"media_type": "text/csv", "bytes": 19396, "sha256": "e8214b7048218de9..."}
The ?v= query is the first 12 hex digits of the file’s SHA-256. The server ignores it, but a changed file gets a new URL, so no cache can serve a stale copy. GitHub Pages caches CSV and ZIP files for 186 days. Use the URLs the documents give, and check what you download against sha256. index.json lists every file in the API this way.
Units and frames. SI throughout: m, m/s, m²/s². Each case has one frame, and its points.json and STL files share it. A–D and H–L are model-scale metres with \(+x\) along the wind at \(\mathrm{WD} = 0\). E, F, G and M are full-scale metres: E and F with \(+x\) east, \(+y\) north and meteorological wind directions, G with \(+x\) downwind, M with \(+x\) north and \(+y\) west. Read frame, units and known_issues in cases/{id}.json before using any number.
Templated endpoints. Paths such as cases/{id}.json and schemas/{name}.schema.json are templates, not URLs to fetch. In index.json such an endpoint says templated: true and lists every concrete file under urls, for example .../api/schemas/case.schema.json. A single-file endpoint carries one url.
Errors. A path that does not exist returns HTTP 404 with GitHub Pages’ HTML page, not JSON. Case ids are listed in cases.json. The other path parameters are enumerated in openapi.json.
Clients. The custom domain is served through Cloudflare, which rejects some clients with HTTP 403. These include Python’s urllib with its default User-Agent and some AI crawlers; curl, wget, requests, httpx and browsers are accepted. Send a descriptive User-Agent header, as the examples below do, or use the fallback base URL: it serves byte-identical files (the same sha256) straight from the repository.
STL content type. GitHub Pages labels .stl responses application/vnd.ms-pki.stl, which is a Windows certificate type. The body is ASCII STL regardless; read it as such.
Endpoints
| Endpoint | What it returns |
|---|---|
GET /api/index.json | every endpoint and every file with its size and SHA-256; versions, compatibility policy, changelog, licence, citation |
GET /api/versions.json | every release, newest first, with the base_url of its frozen copy (served only at this path; see Versions and pinning) |
GET /api/cases.json | every case: id, title, frame, units, counts, engine status, links |
GET /api/cases/{id}.json | one case in full: configurations (Case C’s 9, Case E’s before/after), frame, units, inflow fit and approach-flow profile, point-table summary, geometry files, validation status, known_issues, attribution |
GET /api/cases/{id}/points.json | the cleaned measurement table: SI units, one frame, null where not measured; columns and column_meaning define each field |
GET /api/cases/{id}/points.csv | the same table as CSV (an empty cell is null) |
GET /api/cases/{id}/approach_flow.csv | the measured approach-flow profile: z (m), U (m/s), and k (m²/s²) or u_rms (m/s) where published |
GET /api/cases/{id}/geometry.json | the case’s STL files with statistics (triangles, bounding box, edge-manifold census), renders and captions, and the box list where the geometry is boxes (A–D and G–L) |
GET /api/cases/{id}/geometry.stl | cleaned ASCII STL of the default configuration: metres, case frame, outward normals |
GET /api/cases/{id}/geometry_{configuration}.stl | the other configurations: C/geometry_0D.stl and C/geometry_1D.stl (geometry.stl is 2D); E/geometry_bc.stl, before construction (geometry.stl is after construction) |
GET /api/cases/{id}/validation.json | Eddy3D’s scores per engine and run. Each result carries every score, the headline scores to quote, captioned figures, the reproducing fixture, and pairs |
GET /api/cases/{id}/pairs/{engine}.csv | the measured and modelled value at every scored point of one engine’s runs, so any score can be recomputed; each result’s pairs_columns defines the columns |
GET /api/validation.json | every case and engine in one table: status, runs, and each run’s headline R, FAC2, FB and hit rate |
GET /api/geometry.json | every STL of every case in one list: case, configuration, units, frame, bounding box, size, SHA-256 |
GET /api/geometry.zip | every STL in one archive. Each file keeps its own case’s frame and units. Overlaid, the files would stack at the origin; E, F, G and M are full scale and the others model scale; C’s three centre blocks and E’s two states are alternatives |
GET /api/cases/{id}/probes.csv | the published points to probe for a comparison of your own solver: the key columns a scored file carries, \(x\) \(y\) \(z\) in the case frame, and scored_as (Validate your own solver) |
GET /api/submissions.json | the community results: every submitted solver with its headline scores per case, scored by the build from the submitted values (Community results) |
GET /api/submissions/{slug}.json | one submission in full: its submission.yaml, and every score, coverage and group per case; submissions/{slug}/{id}.csv is the values it gave |
GET /api/schemas/{name}.schema.json | the JSON Schema of each document: index, versions, cases, case, points, geometry, validation, validation-summary, geometry-index, submissions, submission |
GET /api/openapi.json | OpenAPI 3.1 description of every endpoint, with the schemas embedded |
GET /api/llms.txt | plain-text entry point for agents (also at /llms.txt) |
Nomenclature: scores
Scores
- \(n\) — number of pairs of a measured value \(o_i\) and a modelled value \(p_i\) at one point
- \(R\) — Pearson correlation coefficient of the pairs
- \(R^2\) — square of \(R\): share of the measured variance a linear fit to the model explains
- \(\mathrm{FAC2}\) — fraction of pairs with \(0.5 \le p_i/o_i \le 2\)
- \(\mathrm{FB}\) — fractional bias \(2(\overline{o}-\overline{p})/(\overline{o}+\overline{p})\), positive when the model is low
- \(\mathrm{HR}\) (hit rate) — fraction of pairs with \(\lvert p_i-o_i\rvert/\lvert o_i\rvert \le \delta_{\mathrm{rel}}\) or \(\lvert p_i-o_i\rvert \le \delta_{\mathrm{abs}}\), where \(\delta_{\mathrm{rel}} = 0.25\) and \(\delta_{\mathrm{abs}} = 0.05\thinspace U_H\) for a velocity component
Quantities scored
- \(U\), \(V\) and \(W\) (mean velocity components) over \(U_H\)
- \(k\) (turbulent kinetic energy) over \(U_H^2\)
- the speed \(\lvert\mathbf{U}\rvert\) and \(\sqrt{\lvert\mathbf{U}\rvert^2+2k}\), against a thermistor’s \(V_{\mathrm{s}}\)
- \(c^\ast = C/c_0\) (dimensionless concentration) at turbulent Schmidt numbers \(\mathrm{Sc_t}\) of 0.3, 0.5 and 0.7
- \(\theta\) (floor-referenced temperature)
An engine’s status for a case is scored (a validation run exists and its scores are in cases/{id}/validation.json), running (the solve is in progress; no scores yet), pending (not run yet) or not comparable (the engine cannot produce the measured quantity at the measured location: OpenLB at 2 m on Cases C to E, where its empty-domain control misses the bar; the measurement is on the Validation page). An engine that is not scored has no result for that case.
The engine ids are openfoam12-realizableKE, openfoam12-kEpsilon and openfoam12-kOmegaSST (the last two for Case I only), umf-realizableKE and umf-buoyantKEpsilon (UMF, urbanMicroclimateFoam, an OpenFOAM 12 solver and Eddy3D’s engine for the non-isothermal Cases J and L; umf-buoyantKEpsilon for J only), openlb, fluidx3d and ml-surrogate. An engine is listed for a case only where it applies. Runs of one engine on one case that differ in a configuration (Case C’s nine, Case F’s two vertical frames, Case G’s five arms, Case M’s three Schmidt numbers) share the engine’s pairs file. Each result has notes, the caveats that come with the run.
What each case serves today
| Case | Points | Geometry | Validation |
|---|---|---|---|
| A | 186 rows: \(U\), \(V\), \(W\), \(\overline{u'u'}\), \(\overline{v'v'}\), \(\overline{w'w'}\), \(k\) (variances; metres) | 1 box, 12 triangles: 0.08 × 0.08 × 0.16 m | OpenFOAM 12 realizable k-ε; OpenLB at \(\mathrm{d}x = 1.6\thinspace\mathrm{m}\); FluidX3D at \(H/64\); ML surrogate |
| B | 224 rows: \(U\), \(V\), \(W\), \(\overline{u'u'}\), \(\overline{v'v'}\), \(\overline{w'w'}\), \(k\) (converted from mm; \(V\) sign-unified to \(+y\)) | 1 box, 12 triangles: 0.05 × 0.20 × 0.20 m | OpenFOAM 12 realizable k-ε; OpenLB at \(\mathrm{d}x = 1.6\thinspace\mathrm{m}\); FluidX3D at \(H/64\); ML surrogate |
| C | 1,080 rows: \(V_{\mathrm{s}}\) over 9 configurations (config 0D/1D/2D × wind_direction_deg 0/22.5/45) | 3 STLs (0D: 8 cubes; 1D, 2D: 9 boxes) with box lists | OpenFOAM 12 realizable k-ε, each configuration (\(\sqrt{\lvert\mathbf{U}\rvert^2 + 2k}\) and \(\lvert\mathbf{U}\rvert\) at each wind direction); ML surrogate |
| D | 267 rows: 234 thermistor (78 points × 3 directions, 36 unmeasured) and 33 split-fibre with \(U\), \(V\), \(W\), \(\overline{u'u'}\), \(\overline{v'v'}\), \(\overline{w'w'}\), \(k\) | 83 boxes, 996 triangles, watertight, rebuilt from the shipped STL | OpenFOAM 12 realizable k-ε at WD 0°, 22.5° and 45°; ML surrogate |
| E | 2,560 rows: velocity ratio over config bc/ac × 16 directions × 80 points; full-scale metres | 2 STLs from the 3dm (before: 8,396 triangles, 542 bodies; after: 8,432, 545 bodies), every body closed and outward; 11 edges where two buildings touch | OpenFOAM 12 realizable k-ε, wind from SW, before and after; ML surrogate |
| F | 472 rows: velocity_ratio and velocity_ratio_sd at 13 points: 264 tunnel thermistor rows (planes at 5, 9 and 20 m) and 208 field cup-anemometer rows (campaign); full-scale metres | 1 STL, 56,250 triangles: terrain and buildings, outward, closed with a 1 m base slab below the rim | OpenFOAM 12 realizable k-ε, wind from NNW, in two vertical frames (configuration: literal, agl) |
| G | 28 rows: velocity_ratio and k_ratio (sonic anemometers in the field; no speed in m/s is published) | 1 box, 12 triangles: the tree belt’s 2 × 74 × 7 m envelope with its 1.2 m gap | OpenFOAM 12 realizable k-ε as a drag sweep: one run per arm (configuration: no belt, three crowns, the real belt as a half domain) |
| H | 1,652 rows: 826 split-fibre rows (\(U\), \(V\), \(W\), \(\overline{u'u'}\), \(\overline{v'v'}\), \(\overline{w'w'}\), \(k\)) and 826 hydrocarbon-analyser rows (C, c_rms, concentration_ratio) | 1 box, 12 triangles: 0.10 × 0.10 × 0.20 m | OpenFOAM 12 realizable k-ε (velocity, \(k\) and \(c^\ast\)) |
| I | 630 rows: 91 LDV rows (\(U\), \(V\), \(W\), \(k\)) and 539 hydrocarbon-analyser rows (C, concentration_ratio) | 1 box, 12 triangles: the 0.1 m cube | OpenFOAM 12 under realizable k-ε, standard k-ε and k-ω SST |
| J | 444 rows: \(U\), \(V\), \(W\), variances, \(k\), concentration, temperature and their turbulent fluxes | 1 box, 12 triangles: 0.08 × 0.08 × 0.16 m | UMF under realizable k-ε (umf-realizableKE) and buoyantKEpsilon (umf-buoyantKEpsilon) |
| K | 203 rows: \(U\), \(V\), \(W\), variances, \(k\) and concentration | 81 cubes, 972 triangles, watertight | OpenFOAM 12 realizable k-ε (velocity, \(k\) and \(c^\ast\) at three turbulent Schmidt numbers) |
| L | 344 rows: as J, with the Reynolds shear stresses and concentration fluxes | 126 cubes, 1,512 triangles, watertight | UMF realizable k-ε (umf-realizableKE) |
| M | 330 rows: 15 tunnel split-fibre rows, 45 tunnel hydrocarbon-analyser rows and 270 field sorbent-tube rows (campaign, run); full-scale metres | 1 STL, 43,950 triangles: the campus, outward, closed with a 1 m base slab | OpenFOAM 12 realizable k-ε, WD 180, one run per turbulent Schmidt number (configuration: sct03, sct05, sct07) |
Nomenclature: symbols, columns and scores
Geometry, reference scales and release
- \(H\), \(b\), \(D\) (m) — height of the building, cube, tower or tree belt; width of the building; edge of the cubes of Case C: the lengths the cases are drawn and normalised by
- \(U_H\) (m/s) — reference speed: the mean approach-flow speed at the height \(H\) (each Data page gives the value used)
- \(U_R\), \(H_R\), \(Z_R\) (m/s, m) — reference speed and height of a workbook: Case E, \(U_R\) at \(Z_R\) = 250 m full scale (1 m in the model); Case M, \(U_R\) and \(H_R\) at the reference point of the tunnel and of each field run
- \(U_{\mathrm{ref}}\), \(L_{\mathrm{ref}}\) (m/s, m) — reference speed and length of a case’s normalisation: \(U_H\) and \(H\) (\(U_R\) and \(H_R\) in Case M); for the velocity ratios the approach speed at 15.9 m (Case E) or at 500 m (Case F)
- \(U_b\) (m/s) — Case G: mean approach-flow speed at 9 m, the normalisation of its approach-flow table
- \(q\), \(c_{\mathrm{gas}}\) (m³/s, ppm) — tracer release of the experiment: volume flow of the released gas, and its tracer content (\(10^6\) ppm when pure); \(Q\) is the source strength given to a solve
- \(R_b\) (dimensionless) — bulk Richardson number of the heated-floor cases J and L (negative: unstable stratification)
- \(\mathrm{Re_H}\), \(\mathrm{Re_D}\) (dimensionless) — Reynolds numbers \(U_H H/\nu\) and \(U_H D/\nu\) (\(\nu\): kinematic viscosity of air)
- \(u_\ast\), \(z_0\), \(\kappa\) (m/s, mm, dimensionless) — friction velocity and roughness length of the log law fitted to the approach flow, and the von Kármán constant (0.41)
- \(\mathrm{WD}\) (column
wind_direction_deg; °) — wind direction (each frame says how it is counted); \(\mathrm{WD_R}\) is the mean direction of a field run at its reference point (Case M) - \(x\), \(y\), \(z\) (columns
x,y,z; m) — coordinates in the case’s frame (\(+x\), \(+y\) and \(+z\) are the positive directions; each Data page says which way \(+x\) points), \(z\) up from the floor of the tunnel or from the ground - \(\mathrm{d}x\) (m) — lattice spacing of the OpenLB and FluidX3D runs, in metres or as a fraction of \(H\) (\(H/64\))
- \(L_r\) (m) — reattachment length behind the building, read from a wake line at ground level; reported as \(L_r/H\)
Velocity and turbulence
- \(U\), \(V\), \(W\) (columns
U,V,W; m/s) — mean velocity along \(+x\), \(+y\), \(+z\); \(V\) is positive toward \(+y\) in every case - \(\lvert\mathbf{U}\rvert\) (m/s) — magnitude of the mean velocity vector, \(\sqrt{U^2+V^2+W^2}\)
- \(V_{\mathrm{s}}\) (column
Vs; m/s) — mean omnidirectional scalar speed of a thermistor (never a vector magnitude) - \(\overline{u'u'}\), \(\overline{v'v'}\), \(\overline{w'w'}\) (columns
uu,vv,ww; m²/s²) — velocity variances (overbar: time average, prime: fluctuation about it) - \(\overline{u'v'}\), \(\overline{u'w'}\), \(\overline{v'w'}\) (columns
uv,uw,vw; m²/s²) — Reynolds shear stresses - \(k = \tfrac{1}{2}\left(\overline{u'u'}+\overline{v'v'}+\overline{w'w'}\right)\) (column
k; m²/s²) — turbulent kinetic energy - \(U/U_{\mathrm{ref}}\) (column
velocity_ratio; dimensionless) — mean speed over the case’s reference speed (\(V_{\mathrm{s}}/U_{\mathrm{ref}}\) for a thermistor) - \(\sigma_{U/U_{\mathrm{ref}}}\) (column
velocity_ratio_sd; dimensionless) — standard deviation ofvelocity_ratioover the observations behind a field value (Case F) - \(k/U_{\mathrm{ref}}^2\) (column
k_ratio; dimensionless) — turbulent kinetic energy over the square of the reference speed, where only normalised values are published (Case G)
Concentration and temperature
- \(C\) (column
C; ppm) — mean tracer-gas concentration as published (volume fraction \(\times 10^6\) of the released gas diluted in air) - \(c_{\mathrm{rms}}\) (column
c_rms; ppm) — root mean square of the concentration fluctuation - \(c_0\) (ppm) — reference concentration, defined for each case on its Data page: \(c_0 = \dfrac{c_{\mathrm{gas}}\thinspace q}{U_H H^2}\) (\(U_R H_R^2\) in the denominator in Case M)
- \(c^\ast\) (column
concentration_ratio; dimensionless) — dimensionless concentration \(c^\ast = C/c_0\); the value of a solve is \(c\thinspace U_H H^2/Q\) for its mean concentration \(c\) and its source strength \(Q\) - \(\mathrm{Sc_t}\) (dimensionless) — turbulent Schmidt number of a solve (0.3, 0.5 and 0.7)
- \(T\), \(T_{\mathrm{rms}}\) (columns
T,T_rms; °C, K) — mean air temperature as published (Cases J and L) and the root mean square of its fluctuation - \(\overline{u'c'}\), \(\overline{v'c'}\), \(\overline{w'c'}\) (columns
uc,vc,wc; ppm·m/s) — turbulent concentration fluxes - \(\overline{u'\theta'}\), \(\overline{v'\theta'}\), \(\overline{w'\theta'}\) (columns
uT,vT,wT; K·m/s) — turbulent heat fluxes (\(\theta'\): the temperature fluctuation) - \(\theta = \dfrac{T_f - T}{T_f - T_H}\) (dimensionless) — floor-referenced temperature the solves of Cases J and L are scored on: 0 at the floor, 1 in the approach air at \(H\); \(T_f\) is the floor temperature and \(T_H\) the temperature of the approach flow at \(H\)
Scores
- \(n\) — number of pairs of a measured value \(o_i\) and a modelled value \(p_i\) at one point
- \(R\) — Pearson correlation coefficient of the pairs
- \(R^2\) — square of \(R\): share of the measured variance a linear fit to the model explains
- \(\mathrm{FAC2}\) — fraction of pairs with \(0.5 \le p_i/o_i \le 2\)
- \(\mathrm{FB}\) — fractional bias \(2(\overline{o}-\overline{p})/(\overline{o}+\overline{p})\), positive when the model is low
- \(\mathrm{HR}\) (hit rate) — fraction of pairs with \(\lvert p_i-o_i\rvert/\lvert o_i\rvert \le \delta_{\mathrm{rel}}\) or \(\lvert p_i-o_i\rvert \le \delta_{\mathrm{abs}}\), where \(\delta_{\mathrm{rel}} = 0.25\) and \(\delta_{\mathrm{abs}} = 0.05\thinspace U_H\) for a velocity component
- \(\mathrm{NMSE} = \dfrac{\overline{(o-p)^2}}{\overline{o}\thinspace \overline{p}}\) — normalised mean square error
- \(\mathrm{NAD} = \dfrac{\overline{\lvert o-p\rvert}}{\overline{o}+\overline{p}}\) — normalised absolute difference
- \(\mathrm{MG} = \exp\overline{(\ln o-\ln p)}\) — geometric mean bias
- \(\mathrm{VG} = \exp\overline{(\ln o-\ln p)^2}\) — geometric variance
- The ratio of \(\mathrm{FAC2}\) keeps its sign, so a model of the other sign than the measurement is a miss; \(\mathrm{FB}\), \(\mathrm{NMSE}\) and \(\mathrm{NAD}\) divide by the means, so they are undefined (shown as —) for a signed component unless the measured and the modelled mean are both positive
- VDI 3783-9 calls the hit rate \(q\) and the tolerances \(D\) and \(W\); on this site those letters keep their other meanings (\(q\) the tracer release rate, \(D\) the block size of Case C, \(W\) the vertical velocity)
- The detection limit of a concentration is one value per case, \(10^{-3}\) of the largest measured value. It sets \(\delta_{\mathrm{abs}}\) of the hit rate and the both-below rule of \(\mathrm{FAC2}\), and floors the values in \(\mathrm{MG}\) and \(\mathrm{VG}\).
- The scores \(\mathrm{FB}\), \(\mathrm{FAC2}\), \(\mathrm{NMSE}\), \(\mathrm{NAD}\), \(\mathrm{MG}\) and \(\mathrm{VG}\) follow Chang & Hanna (2004); the hit rate follows VDI 3783-9.
Quantities scored
- \(U\), \(V\) and \(W\) (mean velocity components) over \(U_H\)
- \(k\) (turbulent kinetic energy) over \(U_H^2\)
- the speed \(\lvert\mathbf{U}\rvert\) and \(\sqrt{\lvert\mathbf{U}\rvert^2+2k}\), against a thermistor’s \(V_{\mathrm{s}}\)
- \(c^\ast = C/c_0\) (dimensionless concentration) at turbulent Schmidt numbers \(\mathrm{Sc_t}\) of 0.3, 0.5 and 0.7
- \(\theta\) (floor-referenced temperature)
Examples in curl and Python are on the Examples page; the versions, the changelog and how the files are regenerated are under Versions and changes.