2026/07/17 by Kanchane Gunawardena · 1 voice
paper · doi:10.5281/zenodo.21419641
openalex publication_date 2026/07/17 · openalex created_date 2026/07/18 · openalex updated_date 2026/07/18
This document describes the complete methodology implemented in the MATLAB function SmoothWindSignal (V2.1), for smoothing wind speed (WS) and wind direction (WD) inputs. The workflow reads EnergyPlus/IWEC weather files (EPW), or EnviMET CSV formatted weather input files, applies robust smoothing to the wind signals within a user-defined calendar window, and exports an Excel workbook, diagnostic plots and a revised EPW or EnviMET CSV formatted weather input file, with only the in-window WS and/or WD values modified. This method includes a configurable wind‑speed (WS) moderating/clamping step that occurs strictly within the defined smoothing window and prior to applying smoothing. When enabled via the Excel input file, WS values inside the selected window that fall below a specified minimum are uplifted to that minimum, and values above a specified maximum are capped to that maximum. The moderated series then feeds the smoothing step, while outside the window the data remains unaltered. Results and logs record whether moderation was applied and how many data points were uplifted/capped. This revised approach facilitates simulation stability by limiting meteorological forcing WS signal extremes, while preserving diurnal variability and mitigating discontinuities at the 0°/360° signal boundary by smoothing the underlying vector components. The workflow/tool was developed to mitigate meteorological forcing errors in EnviMET or similar CFD approaches, where flow change (WS and WD) limits are needed to maintain simulation stability. Python implementation (v2.2) The new Python implementation converts the MATLAB-based wind smoothing workflow into a standalone, cross-platform tool while preserving the main processing logic. The script reads the Excel input workbook and follows the same MATLAB logic to generate the same outputs.