臺灣博碩士論文加值系統

月電力需求預測所提供的資訊乃產電企業發展供電系統的重要基礎，因此本研究之目的是為發電企業建立電力預測系統，以預估未來之用電需求，方便企業作更有效的能源管理。然而，由於影響月電力需求的氣候因子，例如：溫度、濕度等會因為「月」的時間較長而稀釋其效果，因此本研究以過去用電歷史資料為一「整合因子」來做未來的電力預測。本研究應用趨勢萃取法將原來的電力需求序列分為趨勢序列與波動序列，其中趨勢序列可描述電力需求序列的趨勢；波動序列則描述電力需求序列的波動特性。有了此二序列後，本研究應用基因演算法與類神經網路來預測趨勢序列；然而由於波動序列有其週期特性，因此本研究運用傅立葉級數分析週期特性的優勢來預測未來之波動序列；因此，結合整合上述特性，本研究提出GANN-Fourier series預測模式。最後，本研究以美國之用電資料根據預測準確度與計算效率兩個指標，比較過去的類神經預測方法與本研究所提出的預測系統，以驗證所提方法之價值。

The information from electricity demand forecasting helps energy generation enterprises develop an electricity supply system. This study aims to develop a monthly electricity forecasting model to predict the electricity demand for energy management. Given that the influence of weather factors, such as temperature and humidity, is diluted in the overall monthly electricity demand, the forecasting model uses historical electricity consumption data as an integrated factor to obtain future prediction. The proposed approach is applied to a monthly electricity demand time series forecasting model that includes trend and fluctuation series, of which the former describes the trend of the electricity demand series and the latter describes the periodic fluctuation imbedded in the trend. An integrated genetic algorithm and neural network model (GANN) is then trained to forecast the trend series. Given that the fluctuation series demonstrates an oscillatory behavior, we apply Fourier series to fit the fluctuation series. The complete demand model is named GANN–Fourier series. U.S. electricity demand data are used to evaluate the proposed model and to compare the results of applying this model with those of using conventional neural networks.

ABSTRACT I
中文摘要 II
ACKNOWLEDGEMENT III
FIGURE & TABLE CAPTIONS VI
LIST OF NOTATIONS IX
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 LITERATURE REVIEW 5
2.1 NEURAL NETWORKS FOR TIME SERIES FORECASTING 5
2.2 GENETIC ALGORITHM 7
2.2.1 ENCODE OPERATOR 8
2.2.2 SELECTION OPERATOR 10
2.2.3 CROSSOVER OPERATOR 11
2.2.4 MUTATION OPERATOR 13
2.3 FOURIER SERIES 14
2.4 DISCUSSION AND CONCLUSION 15
CHAPTER 3 THE PROPOSED SYSTEM 16
3.1 TREND EXTRACTION SYSTEM 17
3.2 ELECTRICITY DEMAND FORECASTING SYSTEM -GANN-FOURIER SERIES 20
3.2.1 TREND FORECASTING WITH GANN 20
3.2.2 FLUCTUATION FORECASTING WITH FOURIER SERIES 24
3.3 DISCUSSION AND CONCLUSION 25
CHAPTER 4 NUMERICAL ILLUSTRATION 27
4.1 STEP-BY-STEP ILLUSTRATION OF THE PROPOSED FORECASTING MODEL 28
4.2 EVALUATION AND DISCUSSION 32
CHAPTER 5 CONCLUSION & FUTURE WORK 35
5.1 SUMMARY AND CONCLUSION 35
5.2 FUTURE WORK 36
REFERENCES 37

1. Abdel-Aal, R. E., and A. Z. AlGarni, “Forecasting monthly electric energy consumption in eastern Saudi Arabia using univariate time-series analysis,” Energy, 22(11), 1059-1069(1997).
2. Aggarwal, S., R. Garg and P. Goswami, “A review paper on different encoding schemes used in genetic algorithms,” International Journal of Advanced Research in Computer Science and Software Engineering, 4(1), 596-600(2014).
3. Azadeh, A., S. F. Ghaderi and S. Sohrabkhani, “Forecasting electrical consumption by integration of neural network, time series and ANOVA,” Applied Mathematics and Computation, 186(2), 1753-1761(2007).
4. Bakirtzis, A. G.,V. Petridis, S. J. Kiartzis and M.C. Alexiadis, “A neural network short term load forecasting model for the Greek power system,” Power Systems, 11(2), 858-863 (1996).
5. Barakat, E. H, “Modeling of non-stationary time-series data. Part II. Dynamic periodic trends,” International Journal of Electrical Power & Energy Systems, 23(1), 63-68(2001).
6. Deep, K. and M. Thakur, “A new crossover operator for real coded genetic algorithms,” Applied Mathematics and Computation, 188(1), 895-911(2007).
7. Deep, K. and M. Thakur, “A new mutation operator for real coded genetic algorithms,” Applied mathematics and Computation, 193(1), 211-230(2007).
8. Deep, K., K. P. Singh, M. L. Kansal and C. Mohan, “A real coded genetic algorithm for solving integer and mixed integer optimization problems,” Applied Mathematics and Computation, 212(2), 505-518(2009).
9. Ghiassi, M. D., K. Z. David, K. Zimbra and H. Saidane, “Medium term system load forecasting with a dynamic artificial neural network model,” Electric Power Systems Research, 76(5), 302-316(2006).
10. Goldberg, D.E. and K. Deb, “A comparison of selection schemes used in genetic algorithms,” Foundations of Genetic Algorithms, 1(1), 69–93(1991).
11. González-Romera, E., M. A. Jaramillo-Moran and D. Carmona-Fernandez, “Monthly electric energy demand forecasting based on trend extraction,” Power Systems, 21(4), 1946-1953(2006).
12. González-Romera, E., M. A. Jaramillo-Morán, and D. Carmona-Fernández, “Monthly electric energy demand forecasting with neural networks and Fourier series,” Energy Conversion and Management, 49(11), 3135-3142(2008).
13. Gross, G. and F. D. Galiana, “Short-term load forecasting,” Proceedings of the IEEE, 75(12), 1558-1573(1987).
14. Hippert, H. S., C. E. Pedreira and R. C. Souza, “Neural networks for short-term load forecasting: A review and evaluation,” Power Systems, 16(1), 44-55(2001).
15. Holland, J. H. and J. S. Reitman, “Cognitive systems based on adaptive algorithms,” ACM SIGART Bulletin, 63, 49-49(1977).
16. Hor, C.-L., Simon J. Watson and S. Majithia, “Analyzing the impact of weather variables on monthly electricity demand,” Power Systems, 20(4), 2078-2085(2005).
17. Infield, D. G. and D. C. Hill, “Optimal smoothing for trend removal in short term electricity demand forecasting,” Power Systems, 13(3), 1115-1120 (1998).
18. Lee, K. Y., Y. T. Cha and J. H. Park, “Short-term load forecasting using an artificial neural network,” Power Systems, 7(1), 124-132(1992).
19. Leshno, M., V. Y. Lin, A. Pinkus and S. Schocken, “Multilayer feedforward networks with a nonpolynomial activation function can approximate any function,” Neural networks, 6(6), 861-867(1993).
20. Metaxiotis, K., A. Kagiannas, D. Askounis and J. Psarras, “Artificial intelligence in short term electric load forecasting: a state-of-the-art survey for the researcher,” Energy Conversion and Management, 44(9), 1525-1534(2003).
21. Mirasgedis, S., Y. Sarafidis, E. Georgopoulou, D.P. Lalas, M. Moschovits, F. Karagiannis and D. Papa-konstantinou, “Models for mid-term electricity demand forecasting incorporating weather influences,” Energy, 31(2), 208-227(2006).
22. Moghram, I. S. and S. Rahman, “Analysis and evaluation of five short-term load forecasting techniques,” Power Systems, 4(4), 1484-1491(1989).
23. Moré, J. J. “The Levenberg-Marquardt algorithm: implementation and theory,” Numerical analysis, 105-116(1978).
24. Papalexopoulos, A. D. and T. C. Hesterberg, “A regression-based approach to short-term system load forecasting,” Power Systems, 5(4), 1535-1547(1990).
25. Rahman, S. and R. Bhatnagar, “An expert system based algorithm for short term load forecast,” Power Systems, 3(2), 392-399(1988).
26. Rumelhart, D. E., G. E. Hinton, and R. J. Williams, “Learning internal representations by error propagation,” CALIFORNIA UNIV SAN DIEGO LA JOLLA INST FOR COGNITIVE SCIENCE, No. ICS-8506, (1985).
27. Szkuta, B. R., L. A. Sanabria and T. S. Dillon, “Electricity price short-term forecasting using artificial neural networks,” Power Systems, 14(3), 851-857(1999).
28. White, D. and P. Ligomenides, “GANNet: A genetic algorithm for optimizing topology and weights in neural network design,” New Trends in Neural Computation, 322-327(1993).
29. Zhao, S. and G. W. Wei, “Jump process for the trend estimation of time series,” Computational Statistics & Data Analysis, 42(1), 219-241(2003).