Simulated and measured greenhouse microclimate series for a multilayer-canopy- and control-coupled EnergyPlus co-simulation (278-day paprika cycle)

Description

This dataset contains the simulated and measured time series required to reproduce the validation results of a multilayer-canopy- and control-coupled EnergyPlus co-simulation of the microclimate in a naturally ventilated commercial paprika greenhouse (Inje, Gangwon, Republic of Korea). The record spans the full 278-day cultivation cycle from 14 February to 18 November 2023. Two CSV files are provided. HTVT_validation_10min.csv holds the native 10-min coupling output (40,032 records) paired with the on-site climate-controller log and the in-zone sensor record; HTVT_validation_hourly.csv holds the hourly means (6,672 records) used for the microclimate validation. Simulated columns (prefixed sim_) comprise Cropzone air temperature and relative humidity, the four control outputs (heating setpoint, ventilation setpoint, leeward vent opening, thermal- and shade-screen position), and the baseboard heating supply rate. Measured columns (prefixed obs_) comprise the corresponding controller-log and sensor series and the on-site irradiance record. Unprefixed columns are the public Korea Meteorological Administration ASOS forcing (irradiance from Wonju station 114; temperature, humidity, and wind from Inje station 211). verify.py recomputes the reported metrics from the two CSV files: Cropzone air temperature CV(RMSE) 7.48% and relative humidity CV(RMSE) 10.22% on 6,656 gap-excluded hourly pairs, the control-output consistency metrics on 39,962 gap-excluded 10-min records, the screen-state confusion matrices, the per-stage decomposition over six phenological stages, the out-of-sample metrics on the 6,195 hours outside the infiltration-weighting selection window, and the simulated cycle heating demand of 137.89 kWh per square metre of Cropzone floor area. README.md documents every column, the timestamp and gap conventions, the two grower-imposed ventilation-override regimes, the selection window, and the mapping from each published table to the columns and records that produce it.

Steps to reproduce

Place HTVT_validation_10min.csv, HTVT_validation_hourly.csv, and verify.py in one directory and run "python verify.py" (Python 3 with pandas and numpy). The script reads only the two CSV files, performs no network access, writes no files, and prints each reproduced metric next to its published value. All conventions the script relies on (end-of-interval timestamps in the 10-min file, start-of-hour timestamps in the hourly file, pairwise exclusion of the five logging gaps, Nash-Sutcliffe form of R2, screen binarization thresholds, and the ready-made ventilation-regime column) are stated in README.md.

Institutions

• Kangwon National University

  Gangwon-do, Chuncheon

Categories

Funders

• National Research Foundation of Korea

  Ministry of Science and ICT

  Daejeon, Daejeon

  Grant ID: RS-2021-NR060130
• Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries

  Gyeonggi-do, Anyang-si

  Grant ID: RS-2025-02221279

Licence