نوع مقاله : پژوهشی

نویسندگان

1 دانشکده مهندسی برق- واحد نجف آباد، دانشگاه آزاد اسلامی، نجف آباد، ایران

2 مرکز تحقیقات ریز شبکه های هوشمند- واحد نجف آباد، دانشگاه آزاد اسلامی، نجف آباد، ایران

چکیده

پیش‌بینی میان-‌مدت بار الکتریکی اغلب برای برنامه‌ریزی عملیات نیروگاه‌های حرارتی و آبی، زمان‌بندی بهینه برای بازرسی و تعمیرات و نگهداری نیروگاه‌ها و شبکه برق استفاده می‌شود. در این مقاله یک روش ترکیبی با استفاده از تبدیل موجک و ماشین یادگیری شدید مقاوم به داده‌های خارج از محدوده، برای پیش‌بینی بلند‌مدت بار ارائه ‌شده است. داده‌های بار و دمای ساعتی، از پایگاه داده GEFCOM 2014 استخراج‌ شده و به دو دسته آموزش و آزمایش تقسیم شده است. از تبدیل موجک یک سطحی برای تجزیه داده‌ها به‌منظور استخراج ویژگی‌ها و کاهش ابعاد ماتریس داده‌ها استفاده می‌شود. دو دسته مقادیر مؤلفه‌های فرکانس پایین (تقریب) و مقادیر مؤلفه‌های فرکانس بالا (جزئیات) حاصل از تجزیه جهت آموزش و پیش‌بینی به مدل‌ وارد شده و خروجی‏ مقادیر پایین با خروجی مقادیر بالای مدل‌ جمع می‏شود تا پیش‏بینی نهایی را تشکیل دهد. جهت سنجش و مقایسه دقت و کارایی روش پیشنهادی، اعمال تبدیل موجک روی داده‌ها، برای سه مدل‌ دیگر ماشین یادگیری شدید انجام گردیده است. همچنین داده‌ها بدون اعمال تبدیل موجک به چهار مدل پیش­بینی دیگر نیز وارد شده و نتایج پیش‌بینی حاصل با روش پیشنهادی مورد مقایسه قرار گرفته است. نتایج ارزیابی فوق نشان می‌دهد که تبدیل موجک و ماشین یادگیری شدید مقاوم به داده‌های خارج محدوده باعث بهبود دقت پیش‌بینی می‌گردد و مقدار میانگین درصد خطای مطلق به عدد ۰۹۶۶/۳ کاهش یافته است. مقدار خطای کلی محاسبه شده روش پیشنهادی بهترین نتیجه در بین سایر مدل‌های ماشین یادگیری شدید و روش‌های بدون پیش‌پردازش بوده است. خطای فوق بر مبنای مقدار میانگین درصد خطای مطلق به­ترتیب ۴۲۰۸/۰ نسبت به مدل ماشین یادگیری شدید اصلی، ۱۱۹۴/۰ نسبت به مدل تنظیم‌شده و ۱۳۵۳/۰ نسبت به مدل تنظیم‌شده و وزن‌دار، کاهش یافته است.

چکیده تصویری

یک روش پیش‌بینی بلندمدت بار الکتریکی مبتنی بر استخراج ویژگی برای کاهش اثر داده های خارج از محدوده

تازه های تحقیق

- یک روش ترکیبی مقاوم به داده‌های خارج از محدوده برای پیش‌بینی بلند‌مدت بار ارائه شده است.

- از تبدیل موجک و ماشین یادگیری شدید در روش پیشنهادی استفاده شده است.

- از تبدیل موجک یک سطحی برای تجزیه داده‌ها به‌منظور استخراج ویژگی‌ها و کاهش ابعاد ماتریس داده‌ها استفاده می‌شود.

- داده‌های بار و دمای ساعتی، از پایگاه داده GEFCOM 2014 استفاده شده است.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

A Feature Extraction Based Long-Term Electricity load forecasting Framework to Reduce the Outliers Data Effects

نویسندگان [English]

  • Mohammad Davoud Saeidi 1
  • Majid Moazzami 2

1 Department of Electrical Engineering- Najafabad Branch, Islamic Azad University, Najafabad, Iran

2 Smart Microgrid Research Center- Najafabad Branch, Islamic Azad University, Najafabad, Iran

چکیده [English]

Electrical load forecasting is the prediction of future demands based on various data and factors containing different consumptions on weekdays, electricity prices and weather conditions that are different for societies and places. Generally, medium-term electrical load forecasting is often used for the operation of thermal and hydropower plants, optimal time planning for maintenance of power plants and the power grids. However, long-term electrical load forecasting is used to manage on-time future demands and generation, transmission and distribution expansion planning. In this paper, a hybrid long-term load forecasting approach using wavelet transform and an outlier robust extreme learning machine is proposed. Hourly load and temperature data were extracted from the GEFCOM 2014 database and divided into two classes of training and test. The one-level wavelet transform is used to decompose data to extract properties and reduce the dimensions of the data matrix. Decomposed low-frequency component (approximations) and high-frequency component values (details) from wavelet analysis are entered into the model for training and forecasting. For comparison accuracy of the proposed method, wavelet transform is applied to the data for the other three extreme learning machines. Also data without wavelet transform entered into four other forecasting models and the load forecasting results are compared with the proposed method. The results of the above mentioned evaluation show that electrical load forecasting by using wavelet transform and outlier robust extreme learning machine improves forecasting accuracy and the MAPE reduces to 3.0966. The overall calculated error by the proposed method was the best result obtained between the three several models of extreme learning machines and without preprocessing model. The MAPE is 0.4208 less than the ELM, 0.944 less than the RELM, and 0.1353 less than the WRELM model, respectively.

کلیدواژه‌ها [English]

  • Extreme learning machine
  • Improve forecast accuracy
  • Preprocessing
  • Long-term load forecasting
  • mean absolute percentage error
  • Wavelet Transform

Citation: M.D. Saeidi, M. Moazzami, "A feature extraction based long-term electricity load forecasting framework to reduce the outliers data effects", Journal of Intelligent Procedures in Electrical Technology, vol. 14, no. 56, pp. 1-20, March 2024 (in Persian).

[1] M. Shahidehpour, H. Yamin, Z. Li, "Market operations in electric power systems: forecasting, scheduling, and risk management", Wiley-IEEE Press, vol. 40, April 2002 (doi: 10.1002/047122412X).
[2] L. Han, Y. Peng, Y. Li, B. Yong, Q. Zhou, L. Shu, "Enhanced deep networks for short-term and medium-term load forecasting", IEEE Access, vol. 7, pp. 4045–4055, 2019 (doi: 10.1109/ACCESS.2018.2888978).
[3] A.A. Mamun, M. Sohel, N. Mohammad, M.S.H. Sunny, D.R. Dipta, E. Hossain, "A comprehensive review of the load Forecasting techniques using single and hybrid predictive models", IEEE Access, vol. 8, pp. 134911–134939, 2020 (doi: 10.1109/ACCESS.2020.3010702).
[4] J.C. Lopez, M.J. Rider, Q. Wu, "Parsimonious short-term load forecasting for optimal operation planning of electrical distribution systems", IEEE Trans. on Power Systems, vol. 34, no. 2, pp. 1427–1437,  March 2019 (doi: 10.1109/TPWRS.2018.2872388).
[5] G. Zhang, J. Guo, "A novel method for hourly electricity demand forecasting", IEEE on Trans. Power Systems, vol. 35, no. 2, pp. 1351–1363, March 2020 (doi: 10.1109/TPWRS.2019.2941277).
[6] S.M. Baek, "Mid-term load pattern forecasting with recurrent artificial neural network", IEEE Access, vol. 7, pp. 172830–172838, 2019 (doi: 10.1109/ACCESS.2019.2957072).
[7] Z. Deng, B. Wang, Y. Xu, T. Xu, C. Liu, Z. Zhu, "Multi-scale convolutional neural network with time-cognition for multi-step short-term load forecasting", IEEE Access, vol. 7, pp. 88058–88071, 2019 (doi: 10.1109/AC­CE­SS.2­019.2926137).
[8] K. Yan, W. Li, Z. Ji, M. Qi, Y. Du, "A hybrid lSTM neural network for energy consumption forecasting of individual households", IEEE Access, vol. 7, pp. 157633–157642, 2019 (doi: 10.1109/ACCE­SS.20­19.2­94­90­6­5).
[9] W. Li, Q. Shi, M. Sibtain, D. Li, D.E. Mbanze, "A hybrid forecasting model for short-term power load based on sample entropy, two-phase decomposition and whale algorithm optimized support vector regression", IEEE Access, vol. 8, pp. 166907–166921, 2020  (doi: 10.1109/ACCESS.2020.3023143).
[10] H. Bo, Y. Nie, J. Wang, "Electric load forecasting use a novelty hybrid model on the basic of data prepro­ces­si­ng technique and multi-objective optimization algorithm", IEEE Access, vol. 8, pp. 13858–13874, 2020 (doi: 10.1109/ACCESS.2020.2966641).
[11] Z. Wen, L. Xie, Q. Fan, H. Feng, "Long term electric load forecasting based on TS-type recurrent fuzzy neural network model", Electric Power Systems Research, vol. 179, Article Number: 106106, Feb. 2020 (Doi: 10.1016/j.epsr.2019.106106).
[12] M. El-Hendawi, Z. Wang, "An ensemble method of full wavelet packet transform and neural network for short term electrical load forecasting", Electric Power Systems Research, vol. 182, Article Number: 106265, May 2020 (doi: 10.1016/j.epsr.2020.106265).
[13] H.H.H. Aly, "A proposed intelligent short-term load forecasting hybrid models of ANN, WNN and KF based on clustering techniques for smart grid", Electric Power Systems Research, vol. 182, Article Number: 106191, May 2020 (doi: 10.1016/j.epsr.2019.106191).
[14] Z. Liao, H. Pan, X. Fan, Y. Zhang, L. Kuang, "Multiple wavelet convolutional neural network for short-term load forecasting", IEEE Internet of Things Journal, vol. 8, no. 12, pp. 9730-9739, June 2021 (doi: 10.11­09/JIOT.2­020.3026733).
[15] M. Zhang, Z. Yu, Z. Xu, "Short-term load forecasting using recurrent neural networks with input attention mechanism and hidden connection mechanism", IEEE Access, vol. 8, pp. 186514-186529, 2020 (doi: 10.110­9/ACCE­SS.2020.3029224).
[16] G. Hafeez, K.S. Alimgeer, I. Khan, "Electric load forecasting based on deep learning and optimized by heu­ri­stic algorithm in smart grid", Applied Energy, vol. 269, Article Number: 114915, July 2020 (doi: 10.1016/j.apenergy.2020.114915).
[17] S. Abbas, M.A. Khan, L.E. Falcon-Morales, A. Rehman, Y. Saeed, M. Zareei, A. Zeb, E.M. Mohamed, "Modeling, simulation and optimization of power plant energy sustainability for IoT enabled smart cities empowered with deep extreme learning machine", IEEE Access, vol. 8, pp. 39982-39997, 2020 (doi: 10.1109/ACCESS.2020.2976452).
[18] L. Yin, Z. Sun, F. Gao, H. Liu, "Deep forest regression for short-term load forecasting of power systems", IEEE Access, vol. 8, pp. 49090-49099, 2020 (doi: 10.1109/ACCESS.2020.2979686).
[19] M.V. Selvi, S. Mishra, "Investigation of performance of electric load power forecasting in multiple time horizons with new architecture realized in multivariate linear regression and feed-forward neural network techniques", IEEE Trans. on Industry Applications, vol. 56, no. 5, pp. 5603-5612, Sept.-Oct. 2020 (doi: 10.1109/TIA.2020.3009313).
[20] C.S. Lai, Y. Yang, K. Pan, J. Zhang, H.L. Yuan HL, W.WY. Ng, Y. Gao, Z. Zhao, T. Wang, M. Shahidehpour, L.L. Lai, "Multi-view neural network ensemble for short and mid-term load forecasting", IEEE Trans. on Power Systems, vol. 36, no. 4, pp. 2992-3003, July 2021 (doi: 10.1109/TPWRS.2020.3042389).
[21] W. Guo, L. Che, M. Shahidehpour, X. Wan, "Machine-learning based methods in short-term load forecasting", The Electricity Journal, vol. 34, no. 1, Article Number: 106884, Jan./Feb. 2021 (doi: 10.10­16/j.t­ej.2­020.106884).
[22] X. Shao, C. Pu, Y. Zhang, C.S. Kim, "Domain fusion CNN-LSTM for short-term power consumption forecasting", IEEE Access, vol. 8, pp. 188352-188362, 2020 (doi: 10.1109/ACCESS.2020.3031958).
[23] M.H. Pham, M.N. Nguyen, Y.K. Wu, "A novel short-term load forecasting method by combining the deep learning with singular spectrum analysis", IEEE Access, vol. 9, pp. 73736-73746, 2021 (doi: 10.1109/AC­CES­S.2021.3078900).
[24] T.A. Farrag, E.E. Elattar, "Optimized deep stacked long short-term memory network for long-term load forecasting", IEEE Access, vol. 9, pp. 68511-68522, 2021 (doi: 10.1109/ACCESS.2021.3077275).
[25] I.M. Mehedi, H. Bassi, M.J. Rawa, M. Ajour, A. Abusorrah, M.T. Vellingiri, Z. Salam, M.P. Bin, "Intelligent machine learning with evolutionary algorithm based short term load forecasting in power systems", IEEE Access, vol. 9, pp. 100113-100124, 2021 (doi: 10.1109/ACCESS.2021.3096918).
[26] J. Luo, T. Hong, S.C. Fang, "Robust regression models for load forecasting", IEEE Trans. on Smart Grid, vol. 10, no. 5, pp. 5397-5404, Sept. 2019 (doi: 10.1109/TSG.2018.2881562).
[27] P. Wang, B. Liu, T. Hong, "Electric load forecasting with recency effect: A big data approach", Int Journal of Forecasting, vol. 32, no, 3, pp. 585-597, July/Sept. 2016 (doi: 10.1016/j.ijforecast.2015.09.006).
[28] A.J.R. Reis, A.P.A. Silva, "Feature extraction via multiresolution analysis for short-term load forecasting", IEEE Trans. on Power Systems, vol. 20, no. 1, pp. 189-198, Feb. 2005 (doi: 10.1109/TPWRS.2004.840380).
[29] K. Zhang, M. Luo, "Outlier-robust extreme learning machine for regression problems", Neurocomputing, vol.  151, no. 3, pp. 1519-1527, March 2015 (doi: 10.1016/j.neucom.2014.09.022).
[30] J. Yang, Y. Zhang, "Alternating direction algorithms for ℓ1-problems in compressive sensing", SIAM Journal of Scientific Computing, vol. 33, no. 1, pp. 250-278, 2011 (doi: 10.1137/090777761).
[31] Z. Lin, M. Chen, Y. Ma, "The augmented lagrange multiplier method for exact Recovery of corrupted low-rank matrices", Mathematics, Optimization and Control pp. 2-28, Sept. 2010 (doi: 10.1016/j.jsb.2012.10.010).
[32] W. Zuo, D. Meng, L. Zhang, X. Feng, D. Zhang, "A generalized iterated shrinkage algorithm for non-convex sparse coding", Proceeding of the IEEE/ICCV, pp. 217–224, Sydney, NSW, Australia, Dec. 2013 (doi: 10.11­09/IC­CV.2013.34).
[33] D.L. Donoho, Y. Tsaig, "Fast solution of ℓ1-Norm minimization problems when the solution may be sparse", IEEE Trans. on Information Theory, vol. 54, no. 11, pp. 4789-4812, Nov. 2008 (doi: 10.1109/TI­T.200­8.92­99­58).
[34] M.A.T. Figueiredo, R.D. Nowak, S.J. Wright, "Gradient projection for sparse reconstruction: Application to compressed sensing and other inverse problems", IEEE Journal of Selected Topics in Signal Processing, vol. 1, no. 4, pp. 586-597, Dec. 2007 (doi: 10.1109/JSTSP.2007.910281).
[35] T. Hong, P. Pinson, S. Fan, H. Zareipour, A. Troccoli, R.J. Hyndman, "Probabilistic energy forecasting: Global energy forecasting competition 2014 and beyond", International Journal of Forecasting, vol. 32, no. 3, pp. 896–913, July/Sept. 2016 (doi: 10.1016/j.ijforecast.2016.02.001).
[36] Aasim, S.N. Singh, A. Mohapatra, "Data driven day-ahead electrical load forecasting through repeated wavelet transform assisted SVM model", Applied Soft Computing, vol. 111, Article Number: 107730, Nov. 2021 (doi: 10.1016/j.asoc.2021.107730).