Forecasting uncertainty of Thailand's electricity consumption compare with using artificial neural network and multiple linear regression methods

Jaisumroum, Nattapon; Teeravaraprug, Jirarat

doi:10.1109/iciea.2017.8282862

Cited by 7 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Electricity consumption in Thailand has increased continually during the past few decades due to the growth of industrialization [26]. Therefore, electricity demand forecasting plays an essential role in utility companies as it provides insights into the requirements for operating power systems and facility planning.…”

Section: Electricity Demand Forecatingmentioning

confidence: 99%

A Reinforcement Learning Approach for Ensemble Machine Learning Models in Peak Electricity Forecasting

Pannakkong,

Vinh,

Tuyen

et al. 2023

Energies

View full text Add to dashboard Cite

Electricity peak load forecasting plays an important role in electricity generation capacity planning to ensure reliable power supplies. To achieve high forecast accuracy, multiple machine learning models have been implemented to forecast the monthly peak load in Thailand over the past few years, yielding promising results. One approach to further improve forecast accuracy is to effectively select the most accurate forecast value for each period from among the forecast values generated by these models. This article presents a novel reinforcement learning approach using the double deep Q-network (Double DQN), which acts as a model selector from a pool of available models. The monthly electricity peak load data of Thailand from 2004 to 2017 are used to demonstrate the effectiveness of the proposed method. A hyperparameter tuning methodology using a fractional factorial design is implemented to significantly reduce the number of required experimental runs. The results indicate that the proposed selection model using Double DQN outperforms all tested individual machine learning models in terms of mean square error.

show abstract

Section: Electricity Demand Forecatingmentioning

confidence: 99%

A Reinforcement Learning Approach for Ensemble Machine Learning Models in Peak Electricity Forecasting

Pannakkong,

Vinh,

Tuyen

et al. 2023

Energies

View full text Add to dashboard Cite

show abstract

“…The results of the work show 3.99% of the MAPE, considered to be under the tolerable limits. In [12], the forecasting of the electricity consumption in Thailand, as a case study, has been presented using artificial neural network and multi linear regression. Using the ANN technique the minimum cost of generation is calculated.…”

Section: Introductionmentioning

confidence: 99%

Solar irradiance forecasting using fuzzy logic and multilinear regression approach: A case study of Punjab, India

Mehta

Basak

2019

IJAAS

View full text Add to dashboard Cite

<p class="MDPI17abstract">The accurate forecasting of solar irradiance depends on various uncertain parameters like time of day, temperature, wind speed, humidity, and atmospheric pressure. All these play an important role in calculating PV power output. In this paper, a novel approach for forecasting of solar irradiance using flexible and accurate fuzzy logic and robust multi-linear regression approach has been proposed considering the above mentioned five variables. Based on the simultaneous consideration of those five variables, the solar irradiance is forecasted using the proposed methodology at a particular location in India, and the results are compared with the real time measured value of solar irradiance at that location on the days for which solar irradiance are forecasted. The proposed method is validated by comparing the results with real time data. The error analysis of the fuzzy logic based proposed system shows the root mean square error of 10.011 and mean absolute percentage error of 1.703%, while compared with real time data measured by instruments pyranometer, anemometer etc. The same results are found better while compared with the results obtained using multilinear regression approach.</p>

show abstract

“…Позже для повышения точности прогнозных оценок, получаемых на базе нормативных методов, т.е. экстраполяцией временных рядов электропотребления, стали применять более современные методы, основанные на учете большого количества факторов в динамике с применением новых достижений математики -нейронных сетей и техноценоза [6,7,8,9,10]. Однако не видно явного результата от применения нейронных сетей.…”

unclassified

Approachestoforecastingpowerconsumptionof Power Systems

Morozova¹

2018

DSMM

View full text Add to dashboard Cite

Аннотация. Рассмотрены вопросы, связанные с прогнозированием годового электропотребления энергосистем. Исследованы экстраполяционные свойства регрессионных и авторегрессионных трендов. Проведено улучшение экстраполяционных свойств регрессионных моделей прогнозирования электропотребления энергосистем применением комбинированных моделей за счет учета случайных отклонений от тренда. Предложена методика, позволяющая сузить ширину доверительных интервалов прогноза электропотребления на основе учета внутригодовой (сезонной) неравномерности режима электропотребления. Дальнейшее повышение прогностических возможностей моделей предполагает проведение факторного анализа показателей, влияющих на электропотребление.

show abstract

Forecasting uncertainty of Thailand's electricity consumption compare with using artificial neural network and multiple linear regression methods

Cited by 7 publications

References 3 publications

A Reinforcement Learning Approach for Ensemble Machine Learning Models in Peak Electricity Forecasting

A Reinforcement Learning Approach for Ensemble Machine Learning Models in Peak Electricity Forecasting

Solar irradiance forecasting using fuzzy logic and multilinear regression approach: A case study of Punjab, India

Approachestoforecastingpowerconsumptionof Power Systems

Contact Info

Product

Resources

About