研究生: |
沈孟璇 meng-hsuan shen |
---|---|
論文名稱: |
居家用電雲端監測系統暨便捷人工智慧預測技術之行動應用程式開發 Development of a Mobile Application for a Home Energy Cloud Monitoring System with Prediction Technology Using Artificial Intelligence |
指導教授: |
周瑞生
Jui-Sheng Chou |
口試委員: |
蔡志豐
Chih-Fong Tsai 歐昱辰 Yu-Chen Ou 許丁友 Ting-Yu Hsu |
學位類別: |
碩士 Master |
系所名稱: |
工程學院 - 營建工程系 Department of Civil and Construction Engineering |
論文出版年: | 2019 |
畢業學年度: | 107 |
語文別: | 中文 |
論文頁數: | 65 |
中文關鍵詞: | 居家用電 、雲端監控 、人工智慧預測 、行動應用程式 |
外文關鍵詞: | household energy consumption, cloud monitoring, artificial intelligence prediction, mobile application |
相關次數: | 點閱:287 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
地緣狹礙且四面環海的台灣,雖居家用電長年供給穩定,減少能源生產過程造成的環境負擔仍是政府持續努力的目標。我國首在民105年10月推出有別以往低壓用戶收費方式的「住商型簡易時間電價」,隨後在民107年4月1日起以「總體電價調漲3%」,期望在電力低壓用戶中,能透過「居家單位」的響應節能措施,藉以降低居家電量的使用。
目前多數台灣居家空間配置的電錶設備為僅能得知總用電量的傳統式電錶,若無自行加裝監測器材如智慧電錶等,難以深入了解自身及家庭用電情況,甚而探究不同時段下用電量之差異性。隨著政府在節能措施補助政策下,正逐年大量換裝智慧電錶,並調漲電費希以價制量。台灣電力公司也公告新型態電力計價方式如住商型二段式時間電價、住商型三段式時間電價,讓居家用戶對於用電計價方式有多重選擇。然,新型計價方式是否對消費者更具吸引力,難以從傳統隔月電費單中量化評估。
有鑒於此,本研究希冀建構一便捷即時監控用電系統,可搭配使用於各式感應裝置蒐集用電量的設備,擴及自行改裝家中電錶儀器、或向台灣電力股份有限公司申請換裝電子式電錶等。之後,使用者的用電資料將自動匯入預先設置好連網系統資料庫中,即時上傳至雲端伺服器並藉由應用程式顯示於智慧行動裝置上。此外,若有額外蒐集之環境特徵資訊,如室內外溫溼度等可能影響居家使用者的能耗參數,可運用本文研提之便捷人工智慧技術,進而於智慧裝置顯示預測的用電量,輔助使用者監控居家用電情況、即時評估用電量異常與否。
本研究開發的監控應用程式將居家用電量透過雲端即時顯示於行動裝置中,同時使用人工智慧語言進行便捷決策樹分析預測居家單位的用電量,期望研發雛型可協助使用者視覺化得知能耗隨時間變化之狀況,得以即時掌握用電趨勢,先期預測之耗電量亦可提供居家用戶作為電價自行試算的參考,作為評估更換時間電價選項之依據。
關鍵字:居家用電、雲端監控、人工智慧預測、行動應用程式
Because Taiwan contains a relatively small and narrow terrain with its four sides surrounding by the ocean, reducing the environmental burdens incurred by power generation is a continual endeavor of the government despite the stable supply of home energy. In October 2016, the Taiwanese government first implemented the Residential- and Commercial-Based Simplified Time-of-Use Program, which differs from previous programs for low-tension users. On April 1, 2018, the government increased the overall energy cost by 3%, in the hope that low-tension users can support household unit-based energy saving measures and reduce household energy consumption in Taiwan.
Currently, most households in Taiwan install the conventional watt-hour meter that only displays the total energy consumption. If users do not install monitoring devices (such as smart meters) on their own, they cannot fully understand their energy consumption conditions at home or compare the differences in energy consumption at different periods. Under the energy saving subsidy policies implemented by the government, the number of users in Taiwan installing smart meters has progressively increased each year. In addition, the government increases the energy cost to discourage energy consumption. The Taiwan Power Company has also announced new rate schedules such as the Residential- and Commercial-Based Two-Stage and Three-Stage Time-of-Use rates, providing household users with a variety of energy plans from which they may choose. However, whether the new rates are more attractive to users is difficult to assess using the quantified data of bimonthly energy bills currently used by the Taiwan Power Company.
According to the aforementioned assertions, this study was conducted to establish a convenient real-time energy monitoring system that can be integrated with various types of sensors for capturing energy data, the conventional watt-hour meter installed at home, or the electricity meter that can be installed by submitting a request form to the Taiwan Power Company. The energy data captured by the proposed system can be automatically sent to a preset online database, allowing the data to be uploaded to a cloud server and displayed on the user’s smart mobile device. In addition, if the system captures data on environmental factors (e.g., indoor and outdoor temperature and humidity) influential to energy consumption parameters, the artificial intelligence (AI) technology proposed in this study can be easily used to display predicted energy data on the user’s smart device. This enables users to monitor their home energy consumption and identify abnormal consumption in real time.
Through a cloud server, the proposed system displays real-time home energy data on users’ mobile devices. In addition, AI languages were coded to apply decision tree analysis to easily predicting the energy consumption of household units. The prototype system can assist users in visualizing their energy consumption with respect to time and thereby understanding their consumption patterns. The predicted energy data also allow the user to estimate their energy costs and serve as a reference for changing their current time-of-use program.
Keywords: household energy consumption, cloud monitoring, artificial intelligence prediction, mobile application
[1] 前瞻基礎特別條例, (2017).
[2] B. Akhmetov, M. Aitimov, Data Collection and Analysis Using the Mobile
Application for Environmental Monitoring, Procedia Computer Science 56 (2015)
532-537.
[3] Author, title, Journal of the Chinese Institute of Civil and Hydraulic Engineering
(year) pages.
[4] M. Beccali, M. Cellura, V. Lo Brano, A. Marvuglia, Forecasting daily urban
electric load profiles using artificial neural networks, Energy Conversion and
Management 45 (18–19) (2004) 2879-2900.
[5] L. Breiman, Classification and Regression Trees (1984).
[6] P. Carvalho, Smart Metering Deployment in Brazil, Energy Procedia 83 (2015)
360-369.
[7] J.W. Cort, M. Kenji, Advantages of the mean absolute error (MAE) over the root
mean square error (RMSE) in assessing average model performance, Climate
Research 30 (1) (2005) 79-82.
[8] C. Cortes, Support-vector networks, (1993).
[9] U. Fayyad, From Data Mining to Knowledge Discovery in Databases, merican
Association for Artificial Intelligence (1996).
[10] B. Glasgo, C. Hendrickson, I.M.L. Azevedo, Using advanced metering
infrastructure to characterize residential energy use, The Electricity Journal 30 (3)
(2017) 64-70.
[11] J. Han, Data Mining: Concepts and Techniques, (2001).
[12] A. Hoog, Chapter 4 - Android file systems and data structures, Android
Forensics, Syngress, Boston, 2011, pp. 105-157.
[13] R.J. Hyndman, A.B. Koehler, Another look at measures of forecast accuracy,
International Journal of Forecasting 22 (4) (2006) 679-688.
[14] A.K. Jain, Artificial neural networks : A Tutorial, (1996).
[15] R. Khasha, M.M. Sepehri, S.A. Mahdaviani, T. Khatibi, Mobile GIS-based
monitoring asthma attacks based on environmental factors, Journal of Cleaner
Production 179 (2018) 417-428.
[16] P . McCullagh, Generalized linear models, European Journal of Operational
Research 16 (3) (1984) 285-292.
[17] S. Moon, Application of Mobile Devices in Remotely Monitoring Temporary
Structures During Concrete Placement, Procedia Engineering 196 (2017) 128-134.
[18] P.-D. Moroşan, R. Bourdais, D. Dumur, J. Buisson, Building temperature
regulation using a distributed model predictive control, Energy and Buildings 42
37
(9) (2010) 1445-1452.
[19] J.L. Rodgers, W.A. Nicewander, Thirteen Ways to Look at the Correlation
Coefficient, The American Statistician 42 (1) (1988) 59-66.
[20] R. Simpson, Android overtakes Windows for first time, statcounter (2017).
[21] C. Tofallis, A better measure of relative prediction accuracy for model selection
and model estimation, Journal of the Operational Research Society 66 (8) (2015)
1352-1362.
[22] G.K.F. Tso, K.K.W. Yau, Predicting electricity energy consumption: A comparison
of regression analysis, decision tree and neural networks, Energy 32 (9) (2007)
1761-1768.
[23] Z. Wang, R.S. Srinivasan, A review of artificial intelligence based building energy
use prediction: Contrasting the capabilities of single and ensemble prediction
models, Renewable and Sustainable Energy Reviews 75 (2017) 796-808.
[24] 台灣電力股份有限公司, 住商型簡易時間電價, (2016).
[25] 經濟部, 電價費率審議會議決平均電價自 4 月 1 日起調整 3%, 經濟部能源
局.
[26] 經濟部能源局, 新能源政策, (2017).
[27] 經濟部能源局, 智慧節電計畫, (2015). [1] 前瞻基礎特別條例, (2017).
[2] B. Akhmetov, M. Aitimov, Data Collection and Analysis Using the Mobile
Application for Environmental Monitoring, Procedia Computer Science 56 (2015)
532-537.
[3] Author, title, Journal of the Chinese Institute of Civil and Hydraulic Engineering
(year) pages.
[4] M. Beccali, M. Cellura, V. Lo Brano, A. Marvuglia, Forecasting daily urban
electric load profiles using artificial neural networks, Energy Conversion and
Management 45 (18–19) (2004) 2879-2900.
[5] L. Breiman, Classification and Regression Trees (1984).
[6] P. Carvalho, Smart Metering Deployment in Brazil, Energy Procedia 83 (2015)
360-369.
[7] J.W. Cort, M. Kenji, Advantages of the mean absolute error (MAE) over the root
mean square error (RMSE) in assessing average model performance, Climate
Research 30 (1) (2005) 79-82.
[8] C. Cortes, Support-vector networks, (1993).
[9] U. Fayyad, From Data Mining to Knowledge Discovery in Databases, merican
Association for Artificial Intelligence (1996).
[10] B. Glasgo, C. Hendrickson, I.M.L. Azevedo, Using advanced metering
infrastructure to characterize residential energy use, The Electricity Journal 30 (3)
(2017) 64-70.
[11] J. Han, Data Mining: Concepts and Techniques, (2001).
[12] A. Hoog, Chapter 4 - Android file systems and data structures, Android
Forensics, Syngress, Boston, 2011, pp. 105-157.
[13] R.J. Hyndman, A.B. Koehler, Another look at measures of forecast accuracy,
International Journal of Forecasting 22 (4) (2006) 679-688.
[14] A.K. Jain, Artificial neural networks : A Tutorial, (1996).
[15] R. Khasha, M.M. Sepehri, S.A. Mahdaviani, T. Khatibi, Mobile GIS-based
monitoring asthma attacks based on environmental factors, Journal of Cleaner
Production 179 (2018) 417-428.
[16] P . McCullagh, Generalized linear models, European Journal of Operational
Research 16 (3) (1984) 285-292.
[17] S. Moon, Application of Mobile Devices in Remotely Monitoring Temporary
Structures During Concrete Placement, Procedia Engineering 196 (2017) 128-134.
[18] P.-D. Moroşan, R. Bourdais, D. Dumur, J. Buisson, Building temperature
regulation using a distributed model predictive control, Energy and Buildings 42
37
(9) (2010) 1445-1452.
[19] J.L. Rodgers, W.A. Nicewander, Thirteen Ways to Look at the Correlation
Coefficient, The American Statistician 42 (1) (1988) 59-66.
[20] R. Simpson, Android overtakes Windows for first time, statcounter (2017).
[21] C. Tofallis, A better measure of relative prediction accuracy for model selection
and model estimation, Journal of the Operational Research Society 66 (8) (2015)
1352-1362.
[22] G.K.F. Tso, K.K.W. Yau, Predicting electricity energy consumption: A comparison
of regression analysis, decision tree and neural networks, Energy 32 (9) (2007)
1761-1768.
[23] Z. Wang, R.S. Srinivasan, A review of artificial intelligence based building energy
use prediction: Contrasting the capabilities of single and ensemble prediction
models, Renewable and Sustainable Energy Reviews 75 (2017) 796-808.
[24] 台灣電力股份有限公司, 住商型簡易時間電價, (2016).
[25] 經濟部, 電價費率審議會議決平均電價自 4 月 1 日起調整 3%, 經濟部能源
局.
[26] 經濟部能源局, 新能源政策, (2017).
[27] 經濟部能源局, 智慧節電計畫, (2015).