Skip to content

Repository files navigation

Enclosure Research

A research toolkit for evaluating low-cost outdoor sensor boxes across sensing, enclosure, power, firmware, calibration, and field reliability.

CI Python 3.11 Sources: 26

Results · Reproduce · Data and figures · Manuscript

Predicted enclosure thermal bias

Analytical temperature and relative-humidity bias under solar loading. See the results for interpretation and figure lineage for inputs, assumptions, and generation commands.

Overview

Outdoor sensor performance depends on the complete deployed system, not only a sensor datasheet. This repository connects four evidence paths:

  • a 26-source literature matrix;
  • reliability analysis for exported deployment logs;
  • a first-order enclosure thermal-bias model; and
  • templates and plans for calibration, CAD, and FEA work.

The current field-log summary is provisional because deployment history and raw exports are maintained outside the repository. The thermal study is an analytical model awaiting laboratory comparison.

flowchart LR
    classDef input    fill:#bbdefb,stroke:#1565c0,stroke-width:2px,color:#1f2933,font-weight:bold;
    classDef process  fill:#b2dfdb,stroke:#00796b,stroke-width:2px,color:#1f2933;
    classDef core     fill:#e1bee7,stroke:#7b1fa2,stroke-width:2px,color:#1f2933,font-weight:bold;
    classDef result   fill:#ffe0b2,stroke:#ef6c00,stroke-width:2px,color:#1f2933;

    L[/Literature records/]:::input --> S[System requirements]:::process
    D[/External deployment logs/]:::input --> R[Reliability analysis]:::process
    A[/Model assumptions/]:::input --> T{{Thermal-bias sweep}}:::core
    S --> C(Design comparison):::result
    R --> C
    T --> C
Loading

Shapes: parallelogram = input · rectangle = process · hexagon = core method · rounded = result.

Reproduce the analysis

Create an environment and regenerate the repository-contained checks and thermal figure:

python -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements.txt
python analysis/check_literature_coverage.py
python analysis/thermal_bias.py

Analyze the two deployment exports separately:

python analysis/analyze_deployment_logs.py \
  --data-dir /path/to/DataEnclosure \
  --out-dir analysis/output

The input directory must contain data1.3_24 - Sheet1.csv and data2_5_29 - Sheet1.csv. Those raw files are not committed. The script writes an audit report, four plots, and filtered subsets; see docs/data-and-figures.md before interpreting them.

Documentation

Document Purpose
docs/results.md Current field-log and model results, kept separate from the repository overview
docs/data-and-figures.md Data sources, filters, equations, plot lineage, and reproduction boundaries
docs/figure-manifest.json Machine-readable generator/input/output map for committed result figures
analysis/thermal_bias_results.md Full thermal-model assumptions and sensitivity results
paper/manuscript_v1.md Working paper
literature/literature_matrix.csv Source-level evidence extraction
docs/cad_fea_plan.md Planned geometry and solver work

Status

  • Done (literature): 26-source literature matrix, per-source pros/cons analyses, and a cross-source synthesis.
  • Done (model): first-order analytical thermal-bias model with a sensitivity sweep; every number is a simulation output, not a measurement.
  • Provisional (external logs): reliability audit of two deployment-log exports kept outside the repository; delivery and continuity metrics only, no accuracy claims.
  • Pending: heat-soak, reference co-location, and CHT/FEA comparisons are planned but not started; no experimental accuracy or calibration results exist yet. See docs/cad_fea_plan.md.

Repository map

analysis/      data checks, reliability analysis, thermal model, and figures
literature/    source matrix and cross-source comparisons
ProConsList/   sensor and enclosure assessment for each bibliography entry
paper/         manuscript source and bibliography
templates/     baseline, deployment, and calibration data templates
deliverables/  rendered reports and summaries
docs/          results, provenance, and plans

Contributing

See CONTRIBUTING.md. Literature changes must update the bibliography, matrix, summary, and corresponding ProConsList/ entry together.

License

No open-source license is included. Contact the repository owner before reuse beyond the permissions provided by copyright law.

About

Research toolkit for evaluating low-cost outdoor sensor boxes: 26-source literature matrix, deployment-log reliability audit, analytical thermal-bias model, and calibration/deployment templates.

Topics

Resources

Contributing

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages