自我相關函數於計算電力用戶工作天數之應用

The Application of Autocorrelation Function to Calculate the Work Days of Power Customers

黃郁文(Yu-Wen Huang)

指導教授 ： 黃怡詔

國立屏東科技大學/管理學院/工業管理系所/碩士(2015年)

https://doi.org/10.6346/NPUST.2015.00063

摘要

隨著國內近年來產業快速成長，電量需求成了目前臺灣工業發展上重要的課題之一，因此電力公司必須對各產業用電情況相當瞭解，來滿足各用戶用電需求，並避免造成用電上的浪費。由於臺灣的用電戶眾多，使得電力公司管理各用戶用電量型態不易，造成無法建制好的配電方案。為了瞭解用戶用電型態，本研究透過自我相關函數（Autocorrelation Function, ACF）大量且快速的自動判斷出各用戶的用電週期性，再以電力公司用電量資料進行正規化來降低單位影響。最後透過迴歸化分析（Regression Analysis）來預測出用電量的增減趨勢，藉此自動判斷出各用戶於每循環的工作天數。未來研究可搭配用電性質或用電量等屬性，做為後續用戶的分群之參考。

關鍵字

自我相關函數 ； 電力管理 ； 迴歸化分析

並列摘要

Due to the rapid growth of domestic industry in recent years, electricity demand has become one of the important issues for the industrial development of Taiwan. The power company has to quite understand the electricity consumption of all industry to meet the demand and to avoid the waste of electricity. Owing to the large number of users in Taiwan, the power company failed to manage the electricity consumption pattern of individual users. This caused difficulty in establishing a good distribution plan. In order to realize the electricity consumption of all users, this study applied autocorrelation function (ACF) and determined the periodic data of large amount of users rapidly and automatically. Furthermore, it normalized the power consumption of individual user and reduced unit influence. Finally, through the application of regression analysis, the trend of changes in electricity consumption can be predicted. The actual working days in a cycle of each individual user can also be automatically diagnosed. In the future, the study result can be combined with studies of the character and amount of electricity consumption to group users.

並列關鍵字

Autocorrelation function ； Electricity management ； Regression analysis

參考文獻

1. A. Gupta and S. Venkataraman (2013), “Reducing Price Volatility of Electricity Consumption for a Firm's Energy Risk Management,” The Electricity Journal, vol. 26, no. 3, pp. 89-105.

2. A. Ghanbari, S.F. Ghaderi, M.A. Azadeh (2010), “Adaptive Neuro-Fuzzy Inference System vs. Regression based approaches for annual electricity load forecasting,” IEEE International Conference on Computer and Automation Engineering, vol. 5, pp. 26-30.

3. D.C. Li, W.-Li. Dai, W.-T. Tseng (2011), “A two-stage clustering method to analyze customer characteristics to build discriminative customer management: A case of textile manufacturing business,” Expert Systems with Applications, vol. 38, no. 6, pp. 7186-7191.

4. D.F. Rogers, G.G. Polak (2013), “Optimal Clustering of Time Periods for Electricity Demand-Side Management,” IEEE Transactions on Power Systems, vol. 28, no. 4, pp. 3842-3851.

5. E. Carpaneto, G. Chicco, R. Napoli, M. Scutariu (2006), “Electricity customer classification using frequency–domain load pattern data,” International Journal of Electrical Power & Energy Systems, vol. 28, no. 1, pp. 13-20.