A Consumer Appliance-Level Framework for Optimized Load Shedding

---

Overview

An implementation of the paper Consumer Appliance-level Load Shedding Optimization for Real-time Application (by Marven E. Jabian, Ryohei Funaki, and Junichi Murata) which is yet to appear in a peer-reviewed journal.

This is part of the curricular requirements of the graduate course EE 257 Optimization in Power Systems offered by Mindanao State University - Iligan Institute of Technology and handled by Engr. Abdul Aziz G. Mabaning.

---

Directory

Paper

The primary documentation for this project is the paper entitled A Consumer Appliance-Level Framework for Optimized Load Shedding. Everything related to the paper is in Paper/. Some of the key files:

• paper_ieee_journ.tex contains the LaTeX code. As of now, the paper is only available in the IEEE journal article format.
• paper_ieee_journ.pdf is the compiled PDF file of the paper.
• references.bib contains the bibliographic information.
• figures.pptx is basically a naive "figure manager" where most of the figures in the paper are created and prepared.

Note: As this project is currently an on-going effort, the paper is not yet final and will continually be updated throughout the duration of the course.

Datasets

Data files and regarding the datasets are in Dataset/. Aside from the paper, documentation pertaining to the datasets are also avaialable in the following files:

• Preparation.ipynb contains codes used in building the datasets.
• Quantitative Summaries.ipynb contains codes used in performing quantitative analysis on the datasets.
• Benchmarks.ipynb contains codes used in benchmark tests on the datasets.

Experiments

Experiment files and codes are in Experiments/.

Others

Miscellaneous/ contains some supporting files that are not directly used in the paper, in the datasets, or in the experiments.

Much of the work is performed in a conda environment with the following key packages:

• Python 3.8.0
• Matplotlib 3.3.2
• NumPy 1.19.2
• pandas 1.1.2

To replicate the environment, use calols.yml; e.g., via

(base) PS C:\WINDOWS\system32> conda env create --file calols.yml

We use MathType to typeset equations and mathematical symbols in figures.pptx. Preferences (e.g., fonts) are stored in Default MathType Preferences.eqp, which can then be loaded in MathType.

---

Updates

This project is under active development.

22 October 2020

• Datasets
  • quantitative summaries of the datasets
• Updated paper draft
  • added quantitative summaries of the datasets
• Others
  • tidied up Paper/ directory
    • figures used in the paper are in Paper/figures/
    • files related to the paper in IEEE journal article format are in Paper/ieee_journ/

17 October 2020

• Public release
• Datasets release
  • preparation procedure
• Paper (draft)
  • modified IEEE journal article format