Development of a hybrid Digital Twin for the dispersion of transient CO2 clouds accompanying accidental pipeline leaks - replication data

This repository contains examples of datasets used to study and validate CO2 dispersion behavior in controlled conditions and in numerical dispersion modeling.

(1) Laboratory measurements (multi-sensor time series).
Controlled dispersion experiments were performed to capture the transient evolution of a CO2 cloud. CO2 concentration was recorded using an array of fixed-position NDIR sensors distributed in 3D space. Each experiment session is stored as a time series in long-table format (one record per sensor sample), including timestamps, sensor identifiers, measured concentration (ppm), and sensor coordinates. Additional session metadata (domain dimensions, source position/orientation, and selected scenario/processing parameters used during post-processing) are embedded in the table to support reproducibility and direct reuse in developed software.

(2) DNV PHAST simulation results.
Dispersion scenarios were simulated in DNV PHAST for different meteorological conditions (wind speed and Pasquill stability class). For each scenario, the cloud boundary corresponding to a concentration threshold of 5000 ppm was exported as a set of points representing the contour in a vertical cross-section (downwind distance vs. cloud height). These contours are intended for comparison against laboratory-derived fields/contours and for benchmarking hybrid modeling or digital-twin approaches.