Short Term Electric Load Forecasting Based on Data Transformation and Statistical Machine Learning
Source abstract
Abstract
The continuous penetration of renewable energy resources (RES) into the energy mix and the transition of the traditional electric grid towards a more intelligent, flexible and interactive system, has brought electrical load forecasting to the foreground of smart grid planning and operation. Predicting the electric load is a challenging task due to its high volatility and uncertainty, either when it refers to the distribution system or to a single household. In this paper, a novel methodology is introduced which leverages the advantages of the state-of-the-art deep learning algorithms and specifically the Convolution Neural Nets (CNN). The main feature of the proposed methodology is the exploitation of the statistical properties of each time series dataset, so as to optimize the hyper-parameters of the neural network and in addition transform the given dataset into a form that allows maximum exploitation of the CNN algorithm’s advantages. The proposed algorithm is compared with the LSTM (Long Short Term Memory) technique which is the state of the art solution for electric load forecasting. The evaluation of the algorithms was conducted by employing three open-source, publicly available datasets. The experimental results show strong evidence of the effectiveness of the proposed methodology.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Energy Load and Power Forecasting · Image and Signal Denoising Methods · Solar Radiation and Photovoltaics
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electrical and Computer Engineering, University of Patras, 265 00 Patras, Greece
Research, Technology & Development Department, Independent Power Transmission Operator (IPTO) S.A., 89 Dyrrachiou & Kifisou Str. Gr, 104 43 Athens, Greece
Ecco International Inc., San Francisco, CA 94104, USA
Funding
H2020 European Research Council
References (65)
Show source-provided references
1. Time series forecasting using a hybrid ARIMA and neural network model10.1016/S0925-2312(01)00702-0 ↗
2. A regression-based approach to short-term system load forecasting10.1109/59.99410 ↗
3. An implementation of a neural network based load forecasting model for the EMS10.1109/59.331456 ↗
4. Papalexopoulos, A.D., Hao, S., and Peng, T.M. (1993, January 5–8). Application of neural network technology to short-term system load forecasting. Proceedings of the Joint International Power Conference Athens Power Tech, Athens, Greece.
5. Load forecasting using support vector machines: A study on EUNITE competition 200110.1109/TPWRS.2004.835679 ↗
6. Energy load forecasting using empirical mode decomposition and support vector regression10.1109/TSG.2012.2235089 ↗
7. Long short-term memory10.1162/neco.1997.9.8.1735 ↗
8. Zheng, J., Xu, C., Zhang, Z., and Li, X. (2017, January 22–24). Electric load forecasting in smart grids using long-short-term-memory based recurrent neural network. Proceedings of the 2017 51st Annual Conference on Information Sciences and Systems (CISS), Baltimore, MD, USA.
9. A genetic algorithm approach for modelling low voltage network demands10.1016/j.apenergy.2017.06.057 ↗
10. Autoregressive with exogenous variables and neural network short-term load forecast models for residential low voltage distribution networks10.3390/en7052938 ↗
11. Forecasting low voltage distribution network demand profiles using a pattern recognition based expert system10.1016/j.energy.2014.01.032 ↗
12. Short-term load forecasting with seasonal decomposition using evolution for parameter tuning10.1109/TSG.2015.2395822 ↗
13. Short-term load forecasting based on a semi-parametric additive model10.1109/TPWRS.2011.2162082 ↗
14. Local short and middle term electricity load forecasting with semi-parametric additive models10.1109/TSG.2013.2278425 ↗
15. Next-day MV/LV substation load forecaster using time series method10.1016/j.epsr.2014.10.003 ↗
16. Neural network-based model design for short-term load forecast in distribution systems10.1109/TPWRS.2015.2390132 ↗
17. Combined operations of renewable energy systems and responsive demand in a smart grid10.1109/TSTE.2011.2161624 ↗
18. Continuous dependence for a thermal convection model with temperature-dependent solubility10.1016/j.amc.2017.03.004 ↗
19. The real-time optimisation of DNO owned storage devices on the LV network for peak reduction10.3390/en7063537 ↗
20. A peak reduction scheduling algorithm for storage devices on the low voltage network10.1109/TSG.2014.2323115 ↗
21. Short-term smart learning electrical load prediction algorithm for home energy management systems10.1016/j.apenergy.2015.01.122 ↗
22. A fast technique for smart home management: ADP with temporal difference learning10.1109/TSG.2016.2629470 ↗
23. Transactive home energy management systems: The impact of their proliferation on the electric grid10.1109/MELE.2016.2614188 ↗
24. Using peer-to-peer energy-trading platforms to incentivize prosumers to form federated power plants10.1038/s41560-017-0075-y ↗
25. Global energy forecasting competition 2017: Hierarchical probabilistic load forecasting10.1016/j.ijforecast.2019.02.006 ↗
26. A new error measure for forecasts of household-level, high resolution electrical energy consumption10.1016/j.ijforecast.2013.08.002 ↗
27. Sousa, J.M., Neves, L.M., and Jorge, H.M. (2009, January 18–20). Short-term load forecasting using information obtained from low voltage load profiles. Proceedings of the 2009 International Conference on Power Engineering, Energy and Electrical Drives, Lisbon, Portugal.10.1109/POWERENG.2009.4915229 ↗
28. Veit, A., Goebel, C., Tidke, R., Doblander, C., and Jacobsen, H.A. (2014, January 11–13). Household electricity demand forecasting: Benchmarking state-of-the-art methods. Proceedings of the 5th International Conference on Future Energy Systems, Cambridge, UK.10.1145/2602044.2602082 ↗
29. Short-term electricity load forecasting of buildings in microgrids10.1016/j.enbuild.2015.04.011 ↗
30. Ghofrani, M., Hassanzadeh, M., Etezadi-Amoli, M., and Fadali, M.S. (2011, January 4–6). Smart meter based short-term load forecasting for residential customers. Proceedings of the 2011 North American Power Symposium, Boston, MA, USA.10.1109/NAPS.2011.6025124 ↗
31. A sparse coding approach to household electricity demand forecasting in smart grids
32. Long-term retail energy forecasting with consideration of residential customer attrition10.1109/TSG.2014.2388078 ↗
33. Predicting consumer load profiles using commercial and open data10.1109/TPWRS.2015.2493083 ↗
34. Relative humidity for load forecasting models10.1109/TSG.2016.2547964 ↗
35. An efficient approach to short-term load forecasting at the distribution level10.1109/TPWRS.2015.2489679 ↗
36. Evaluating combined load forecasting in large power systems and smart grids10.1109/TII.2012.2219063 ↗
37. Marinescu, A., Harris, C., Dusparic, I., Clarke, S., and Cahill, V. (2013, January 18). Residential electrical demand forecasting in very small scale: An evaluation of forecasting methods. Proceedings of the 2013 2nd International Workshop on Software Engineering Challenges for the Smart Grid (SE4SG), San Francisco, CA, USA.10.1109/SE4SG.2013.6596108 ↗
38. Using smart meter data to improve the accuracy of intraday load forecasting considering customer behavior similarities10.1109/TSG.2014.2364233 ↗
39. Short-term load and wind power forecasting using neural network-based prediction intervals10.1109/TNNLS.2013.2276053 ↗
40. Analysis and modeling for short-to medium-term load forecasting using a hybrid manifold learning principal component model and comparison with classical statistical models (SARIMAX, Exponential Smoothing) and artificial intelligence models (ANN, SVM): The case of Greek electricity market10.3390/en9080635 ↗
41. Predicting future hourly residential electrical consumption: A machine learning case study10.1016/j.enbuild.2012.03.010 ↗
42. Deep learning for household load forecasting—A novel pooling deep RNN10.1109/TSG.2017.2686012 ↗
43. Short-term residential electric load forecasting: A compressive spatio-temporal approach10.1016/j.enbuild.2015.11.068 ↗
44. Energy analysis of a building using artificial neural network: A review10.1016/j.enbuild.2013.06.007 ↗
45. Short-term residential load forecasting based on resident behaviour learning10.1109/TPWRS.2017.2688178 ↗
46. Keren, G., and Schuller, B. (2016, January 24–29). Convolutional RNN: An enhanced model for extracting features from sequential data. Proceedings of the 2016 International Joint Conference on Neural Networks (IJCNN), Vancouver, BC, Canada.10.1109/IJCNN.2016.7727636 ↗
47. Acharya, S.K., Wi, Y.M., and Lee, J. (2019). Short-Term Load Forecasting for a Single Household Based on Convolution Neural Networks Using Data Augmentation. Energies, 12.10.3390/en12183560 ↗
48. Canizes, B., Silva, M., Faria, P., Ramos, S., and Vale, Z. (2015, January 10–13). Resource scheduling in residential microgrids considering energy selling to external players. Proceedings of the 2015 Clemson University Power Systems Conference (PSC), Clemson, SC, USA.10.1109/PSC.2015.7101700 ↗
49. Almabdy, S., and Elrefaei, L. (2019). Deep convolutional neural network-based approaches for face recognition. Appl. Sci., 9.10.3390/app9204397 ↗
50. Li, L., Ota, K., and Dong, M. (2017, January 21–23). Everything is image: CNN-based short-term electrical load forecasting for smart grid. Proceedings of the 2017 14th International Symposium on Pervasive Systems, Algorithms and Networks & 2017 11th International Conference on Frontier of Computer Science and Technology & 2017 Third International Symposium of Creative Computing (ISPAN-FCST-ISCC), Exeter, UK.10.1109/ISPAN-FCST-ISCC.2017.78 ↗
51. Zhang, W. (1988, January 24–26). Shift-invariant pattern recognition neural network and its optical architecture. Proceedings of the Annual Conference of the Japan Society of Applied Physics, Keio Plaza Hotel, Tokyo, Japan.
52. Parallel distributed processing model with local space-invariant interconnections and its optical architecture10.1364/AO.29.004790 ↗
53. Saha, S. (2016, December 16). A Comprehensive Guide to Convolutional Neural Networks—the ELI5 Way. 2018. Available online: https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53.
54. Deep learning10.1038/nature14539 ↗
55. LeCun, Y., Kavukcuoglu, K., and Farabet, C. (June, January 30). Convolutional networks and applications in vision. Proceedings of the 2010 IEEE International Symposium on Circuits and Systems, Paris, France.10.1109/ISCAS.2010.5537907 ↗
56. Backpropagation applied to handwritten zip code recognition10.1162/neco.1989.1.4.541 ↗
57. Neocognitron: A neural network model for a mechanism of visual pattern recognition10.1109/TSMC.1983.6313076 ↗
58. Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012). Imagenet classification with deep convolutional neural networks. (NIPS’12), Proceedings of the 25th International Conference on Neural Information Processing Systems—Volume 1, Lake Tahoe, Nevada, 3–8 December 2012, Curran Associates Inc.
59. Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
60. Deep learning: Methods and applications10.1561/2000000039 ↗
61. Mills, T.C., and Mills, T.C. (1991). Time Series Techniques for Economists, Cambridge University Press.
62. Greene, W.H. (1989). LIMDEP. New York: Econometric Software. _. 1990. Econometric Analysis, MacMillan.
63. A multi-source dataset of urban life in the city of Milan and the Province of Trentino10.1038/sdata.2015.55 ↗
64. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance10.3354/cr030079 ↗
65. Mean absolute percentage error for regression models10.1016/j.neucom.2015.12.114 ↗