本文主旨在探究負載預測及契約容量最佳化。首先，蒐集相關
文獻，比較傳統線性回歸及灰色理論之負載預測方法，再提出資料
修正，比較修正前後負載預測之結果。經結果得之，資料濾除後可
以降低誤差，並續之，利用粒子群演算法計算最佳契約容量。求得
最佳解後，與前一年度比較可節省大量電費。本研究利用商用套裝
軟體 Matlab 建構完成負載預測模型及契約容量最佳化模型，並續
而進行分析、比較與探討。最後，配合台電最新需量競價措施，針
對用戶未來負載，選定實施日期、時間及抑低容量，評估是否適用
並提供各界參考，有助於用戶電費節省及對於電力負載能更有效控
管。

This thesis aims to study the load forecasting and optimization of
contract capacity. To begin with, collect the relevant literature. Then,
compare the traditional linear regression and grey theory of the forecast
load method. Furthermore, put forward revised data. Finally, compare
the load forecasting results of pre- and post- prediction. Filtered data
can reduce the error. After that, using of particle swarm algorithm
(PSO) does the best contract capacity calculation. Company use by
calculating the best value can save more electricity than previous years.
In this study, the auther uses the commercial suite software MATLAB
to construct the load forecasting model and the contract capacity
optimization model, and then the auther proceeds with the analysis,
comparison and discussion. In summary, the study with the latest
measures of Taipower, demand bidding, for the user's future load
selected implementation date, time and capacity to reduce the
assessment can evaluate wether the method is applicable or not, and
provide reference. The study can help users to save electricity and
more effective control the power load.

[1] D. C. Park, M. A. El-Sharkawi, R. J. Marks, L. E. Atlas and M. J.
Damborg, "Electric Load Forecasting Using an Artificial Neural
Network, " IEEE Transactions on Power Systems, vol. 6, no. 2, pp.
442-449, May 1991.
[2] 林群絜，神經網路於電力負載預測影響因子之分析，義守大學
電機工程學系碩士班論文，2013。
[3] 孫智超，微型電網負載預測之研究，國立臺灣科技大學電機工
程學系碩士論文，2012。
[4] A. O. Kusakci and B. Ayvaz, "Electrical Energy Consumption
Forecasting for Turkey Using Grey Forecasting Technics with
Rolling Mechanism, " 2015 2nd International Conference on
Knowledge-Based Engineering and Innovation (KBEI), Tehran,
pp. 8-13, 2015.
[5] Shuang Liu and Li-Xia Tian, "The Study of Long-Term Electricity
Load Forecasting Based on Improved Grey Prediction Model",
2013 International Conference on Machine Learning and
Cybernetics, Tianjin, pp. 653-656, 2013.
[6] 台灣電力公司，季節電價，http://www.taipower.com.tw/UpFile/
PowerSavFile/main_6_2_2.pdf，2015/4/12。
[7] 台灣電力公司，營業規章，http://www.taipower.com.tw/content/
rules_item/rules_item01.aspx?CUType=42016，2017/6/22。
[8] H. Patel, M. Pandya and M. Aware, "Short Term Load Forecasting
of Indian System Using Linear Regression and Artificial Neural
Network," 2015 5th Nirma University International Conference
on Engineering (NUiCONE), Ahmedabad, pp. 1-5, 2015.
[9] 鄧聚龍、郭洪，灰預測原理與應用，全華圖書出版公司，1996。
[10] 溫坤禮、趙忠賢、張宏志、陳曉瑩、溫惠筑，灰色理論，五南
圖書出版公司，2013。
[11] 鄧聚龍，灰色預測與決策，華中理工大學出版社。
[12] 台灣世曦工程顧問股份有限公司、台灣港務股份有限公司高雄
分公司，高雄港及安平港之用電需量統計表，2016。
[13] 葉鎮愷，灰色馬可夫模型於電力負載之研究暨 Matlab 工具箱
之研發，建國科技大學電機工程研究所碩士學位論文，2007。
[14] 溫坤禮、張簡士琨、葉鎮愷、王建文、林慧珊，MATLAB 在
灰色理論的應用，全華圖書出版公司，2007。
[15] X. Tong, Q. Chen, J. Fan, Q. Yan and C. Kang, "Adaptability
Verification and Application of The T-distribution in Short-term
Load Forecasting Error Analysis," 2014 International Conference
on Power System Technology, Chengdu, pp. 145-150, 2014.
[16] 臺灣國際商港區供電供水最佳化與節能燈具運用之評估與規
劃委託技術服務期中報告書，台灣世曦工程顧問股份有限公司，
2016。
[17] 陳映君，契約用電最適規劃之研究技術報告，朝陽科技大學，
2008。
[18] R. Eberhart and J. Kennedy, "A New Optimizer Using Particle
Swarm Theory", Micro Machine and Human Science, 1995 MHS
'95, Proceedings of the Sixth International Symposium on Nagoya,
pp. 39-43, 1995.
[19] J Kennedy.and R Eberhart., “Particle Swarm Optimization”,
International Conference on Neural Networks, vol.4,
pp.1942-1948, 1995.
[20] 黃大鎔，改良式模糊粒子群演算法及其在螺旋電感最佳化設計
之應用，國立中央大學電機工程學系碩士論文，2016。
[21] Yuhui Shi and Russell Eberhart, “A Modified Particle Swarm
Optimizer”, Department of Electrical Engineering Indiana
University Purdue University Indianapolis hdianqolis, 1998.
[22] 楊智翔，結合負載預測於配電網重購之研究，國立臺灣科技大
學電機工程學系碩士論文，2016。
[23] 台灣電力公司，需量競價，http://www.taipower.com.tw/UpFile/
PowerSavFile/3.2%E9%9C%80%E9%87%8F%E7%AB%B6%E
5%83%B9%E6%8E%AA%E6%96%BDDM_106.01.10.pdf，
2017/1/10。