The Study of Short-Term Load Forecasting Using Deep Neural Networks
碩士 === 國立臺北科技大學 === 自動化科技研究所 === 106 === In this age of smart grids, power load predictions are incredibly important for the electric power industry. If consumers are able to accurately foresee their future power usage, or if power monitoring systems have warning mechanisms, we could potentially rai...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2018
|
Online Access: | http://ndltd.ncl.edu.tw/handle/f9r8mz |
id |
ndltd-TW-106TIT05146017 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-106TIT051460172019-07-04T06:00:00Z http://ndltd.ncl.edu.tw/handle/f9r8mz The Study of Short-Term Load Forecasting Using Deep Neural Networks 應用深度學習於短期負載預測之研究 Yu-Ching Chen 陳昱斳 碩士 國立臺北科技大學 自動化科技研究所 106 In this age of smart grids, power load predictions are incredibly important for the electric power industry. If consumers are able to accurately foresee their future power usage, or if power monitoring systems have warning mechanisms, we could potentially raise awareness on saving electricity, thereby reducing needless power waste. This research is based on a transformer station’s power usage between 2012 and 2018. By analysing data connected to power load and exploring the load prediction machine algorithm in this research, this thesis aims to assist power monitoring systems in improving mechanisms in saving power. This thesis analyses past power loads and other relevant factors in relations to each season. Information is divided into three categories, yearly, summer and winter; utilising these datasets, we create four prediction models of neural networks, including artificial neural networks, long short-term memory, convolutional neural networks, and convolutional long short-term memory. Through repeated training and testing can the model minimise the difference between its prediction and the actual power usage. Experiments results show that there is indeed correlation between the actual usage and the predicted usage; of which, using the model of convolutional long short-term memory to predict by summer dataset, mean absolute percentage error yields the lowest difference rate of 4.67%. Using the prediction results, we can help the electric power industry foresee future power usage and provide dispatch unit with useful information for them. Wen-Hui Chen 陳文輝 2018 學位論文 ; thesis 65 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺北科技大學 === 自動化科技研究所 === 106 === In this age of smart grids, power load predictions are incredibly important for the electric power industry. If consumers are able to accurately foresee their future power usage, or if power monitoring systems have warning mechanisms, we could potentially raise awareness on saving electricity, thereby reducing needless power waste. This research is based on a transformer station’s power usage between 2012 and 2018. By analysing data connected to power load and exploring the load prediction machine algorithm in this research, this thesis aims to assist power monitoring systems in improving mechanisms in saving power. This thesis analyses past power loads and other relevant factors in relations to each season. Information is divided into three categories, yearly, summer and winter; utilising these datasets, we create four prediction models of neural networks, including artificial neural networks, long short-term memory, convolutional neural networks, and convolutional long short-term memory. Through repeated training and testing can the model minimise the difference between its prediction and the actual power usage. Experiments results show that there is indeed correlation between the actual usage and the predicted usage; of which, using the model of convolutional long short-term memory to predict by summer dataset, mean absolute percentage error yields the lowest difference rate of 4.67%. Using the prediction results, we can help the electric power industry foresee future power usage and provide dispatch unit with useful information for them.
|
author2 |
Wen-Hui Chen |
author_facet |
Wen-Hui Chen Yu-Ching Chen 陳昱斳 |
author |
Yu-Ching Chen 陳昱斳 |
spellingShingle |
Yu-Ching Chen 陳昱斳 The Study of Short-Term Load Forecasting Using Deep Neural Networks |
author_sort |
Yu-Ching Chen |
title |
The Study of Short-Term Load Forecasting Using Deep Neural Networks |
title_short |
The Study of Short-Term Load Forecasting Using Deep Neural Networks |
title_full |
The Study of Short-Term Load Forecasting Using Deep Neural Networks |
title_fullStr |
The Study of Short-Term Load Forecasting Using Deep Neural Networks |
title_full_unstemmed |
The Study of Short-Term Load Forecasting Using Deep Neural Networks |
title_sort |
study of short-term load forecasting using deep neural networks |
publishDate |
2018 |
url |
http://ndltd.ncl.edu.tw/handle/f9r8mz |
work_keys_str_mv |
AT yuchingchen thestudyofshorttermloadforecastingusingdeepneuralnetworks AT chényùqín thestudyofshorttermloadforecastingusingdeepneuralnetworks AT yuchingchen yīngyòngshēndùxuéxíyúduǎnqīfùzàiyùcèzhīyánjiū AT chényùqín yīngyòngshēndùxuéxíyúduǎnqīfùzàiyùcèzhīyánjiū AT yuchingchen studyofshorttermloadforecastingusingdeepneuralnetworks AT chényùqín studyofshorttermloadforecastingusingdeepneuralnetworks |
_version_ |
1719220714906583040 |