EVF

Description

The MATLAB package accompanying this manuscript is organized into two main directories: Main Code Contains the primary MATLAB scripts used to implement the Entropy Variability Factor (EVF) framework, including rolling Shannon entropy computation, EVF and SEVF calculation, Entropy Variability Index (EVI) classification, statistical testing, visualization generation, and real-world case-study analysis. Sensitivity Analysis Contains supplementary MATLAB scripts used to evaluate the robustness of the EVF framework with respect to rolling-window length, histogram-bin selection, null-zone calibration, and EVI threshold stability. The Main Code directory is further divided into the following subfolders: • Sinusoidal Time Series Includes synthetic benchmark simulations used to demonstrate the monotonic response of rolling Shannon entropy to progressively increasing disorder levels. • Null Zone Calibration Contains Monte Carlo simulations used to characterize EVF behavior under stable conditions and to empirically derive the ±2% EVF null zone. • EVI Thresholds Includes Monte Carlo analyses used to establish empirical EVI severity thresholds corresponding to negligible, small, moderate, and large structural variability changes. • Texas Case Study Contains scripts used for the February 2021 Texas power crisis analysis, including entropy computation, EVF/EVI generation, and comparison against the February 2020 baseline. • California Case Study Contains scripts used for the August 2020 California heatwave analysis, including regional temperature screening, CAISO load analysis, and EVF-based variability assessment relative to August 2019. All MATLAB scripts were developed and tested using MATLAB R2024a. The code is organized to allow reproducibility of all figures, tables, and analyses presented in the manuscript.

Steps to reproduce

Run MATALB scripts on MATLAB 2024a version or higher.

Institutions

• United Arab Emirates University

  Abu Dhabi, Al Ain

Categories

Licence