A Study on Photovoltaic Prediction Algorithm and Profitability for Machine Learning based Prediction Market Participation

Sang-jin Kim; Jae-Hyeok Yu; Byung-Hoon Jang; Sung-Min Woo

doi:10.7836/kses.2022.42.6.173

All Issue

2022 Vol.42, Issue 6 Preview Page

Research Article

A Study on Photovoltaic Prediction Algorithm and Profitability for Machine Learning based Prediction Market Participation 머신러닝 기반의 예측 시장 참여를 위한 태양광 발전량 예측 알고리즘 및 수익성에 관한 연구

30 December 2022. pp. 173-183

PDF XML

Abstract

With the introduction of the power generation prediction system, research is being conducted to predict hourly solar power generation through various algorithms and reduce prediction errors. However, increasing settlement revenue when participating in the prediction market is more important than improving prediction accuracy. In this study, we propose a method for predicting solar power generation using forecast and predicted weather data. In addition, the clustering algorithm was used based on solar radiation forecast data, and the causes of low prediction accuracy and profitability were analyzed for each cluster. Through this study, participation in the renewable energy generation prediction market is expected to be activated and opportunities for various business models will be provided.

Keywords

Photovoltaic

Weather forecast

Weather prediction

Machine Learning

Clustering algorithm

키워드

태양광 발전량

일기예보

기상예측

기계학습

군집화 알고리즘

References

KPX, 20222. https://new.kpx.or.kr/board.es?mid=a10109010700&bid=0082&act=view&list_no=68145. last accessed on the 14^th December 2022.

Ministry of Commerce, Industry and Energy, Introduction of Renewable Energy Generation Prediction System, 2020.

Lee, J., Park, W., Lee, I., and Kim, S., Comparison of Solar Power Prediction Model Based on Statistical and Artificial Intelligence Model and Analysis of Revenue for Forecasting Policy, Journal of IKEEE, Vol. 26, No. 3, pp. 355-363, 2022.

Sohn, H., Jung, S., and Kim, S., The Prediction and Valuation of Gas Consumption in Building using Artificial Neural Networks Based on Clustering Method, The Korean Journal of Applied Statistics, Vol. 29, No. 1, pp. 193-203, 2016. 10.5351/KJAS.2016.29.1.193

Choi, D., Lee, Y., and Ko, M., A Study on Electricity Demand Forecasting Based on Time Series Clustering in Smart Grid, Korea Institute of Ecological Architecture and Environment, Vol. 18, No. 5, pp. 69-74, 2018.

Ke, G, Meng, Q., Finely, T., Wang, T., Chen, W., and Ma, W., Lightgbm: A Highly Efficient Gradient Boosting Decision Tree, Advances in Neural Information Processing System, Vol. 30, pp. 3146-3154, 2017.

Information

Publisher :Korean Solar Energy Society
Publisher(Ko) :한국태양에너지학회
Journal Title :Journal of the Korean Solar Energy Society
Journal Title(Ko) :한국태양에너지학회 논문집
Volume : 42
No :6
Pages :173-183
Received Date : 2022-11-07
Revised Date : 2022-12-16
Accepted Date : 2022-12-21
DOI :https://doi.org/10.7836/kses.2022.42.6.173

Journal of the Korean Solar Energy Society ISSN:1598-6411(Print) 2508-3562(Online) 한국태양에너지학회 논문집

All Issue