Processed Data, Thermal Model and Analysis Code for a PWM-Controlled Far-Infrared Emitter System
Description
This package provides processed time-series data, a lumped thermal model, Python analysis code, results tables and figures for a laboratory-scale far-infrared (FIR) emitter system operated under proportional control with pulse-width modulation (PWM). The analysis covers 14 unloaded experimental trials at temperature setpoints of 100–300 °C, comprising 21,521 observations. The package includes data preparation and quality flagging; audits of measurement resolution, conversion and timing; controller-performance assessment; leave-one-trial-out thermal prediction at 10, 30 and 60 s horizons; closed-loop simulation using the experimental control law; comparisons of one-step temperature predictors at approximately 0.402 s; and a measurement uncertainty budget. Peak emission wavelengths are estimated from temperature using Wien’s displacement law. The accompanying Jupyter notebooks reproduce the analyses using the separately published raw experimental dataset (DOI: 10.17632/7pp4rx65mf.1). The raw dataset is not included in this package and must be downloaded separately.
Files
Steps to reproduce
1. Download this package and the raw experimental dataset (DOI: 10.17632/7pp4rx65mf.1). 2. Place the raw dataset in `Package_1/` and this package in `Package_2/`, with both folders sharing the same parent directory. Alternatively, update `PACKAGE1_DIR` in `Code/fir_common.py`. 3. Install the Python dependencies listed in `Package_2/Code/requirements.txt`. 4. Open Jupyter in `Package_2/Code/` and run notebooks 01–06 in numerical order, retaining the supplied parameters and random seed of 42. 5. The notebooks generate processed datasets, results tables, the fitted linear dynamic model and figures in `Processed_Data/`, `Results/` and `Figures/`. Further instructions are provided in `Code/HOW_TO_RUN.txt`.
Institutions
- Rajarata University of Sri LankaNorth Central Province, Anuradhapura