Thursday , June 21 2018

Adaptive Neuro – Fuzzy Inference Systems –
An Alternative Forecasting Tool for Prosumers

Otilia DRAGOMIR1, Florin DRAGOMIR1, Veronica STEFAN2, Eugenia MINCA1
1 Automation, Computer Science and Electrical Engineering Department,
Valahia University of Târgovişte,
2, Carol I Blvd., Târgoviște, 130024, Romania
drg_otilia@yahoo.com, drg_florin@yahoo.com
2 Faculty of Economic Sciences,
Valahia University of Târgovişte,
2 Carol I Bd., Târgovi te, 130024, Romania
veronica.stefan@ats.com.ro, minca.eugenia@gmail.com

Abstract: The goal of this paper is to propose a forecasting tool to producers/ consumers (prosumers) of renewable energy sources, based on artificial intelligence techniques, trying to obtain optimal predictions. The exploration and the assessment of the criteria used for choosing the adequate forecasting tool are made in the artificial intelligence context. In this respect, firstly, the criteria used for choosing the best forecasting technology, in relation to each step of the modelling process are presented. Secondly, the identified criteria are tested on two Adaptive Neuro- Fuzzy Inference System (ANFIS) models, in order to underline the effects of these users’ decisions over the forecasting performances.

Keywords: Forecasting, Neural Network, Adaptive Neuro – Fuzzy Inference Systems, Prosumers.

>>Full text
CITE THIS PAPER AS:
Otilia DRAGOMIR, Florin DRAGOMIR, Veronica STEFAN, Eugenia MINCA, Adaptive Neuro – Fuzzy Inference Systems – An Alternative Forecasting Tool for Prosumers, Studies in Informatics and Control, ISSN 1220-1766, vol. 24 (3), pp. 351-360, 2015.

  1. Introduction

The successful entry on energy market, recently liberalized in Romania, depends on the capacity of each producer to predict through different methods the future evolution of the quantity of renewable energy sources. Thus, the integration of the predictive ability is a very important function in the command and control of a power system, especially at the micro-grid level, heavily dependent on the consumer and with reduced availability for storage. European context. The current energy policy of the European Union (EU) considers the security of supply, competitiveness and sustainability as central targets. In order to achieve these targets, a series of constraints (“20-20-20 objective”) is imposed through European strategies [1]: 20% reduction in emissions of greenhouse gases compared to 1990, providing 20% of entire EU energy consumption by Renewable Energy Sources (RES) and a 20% reduction in energy use in comparison with a similar scenario in which no action regarding sustainability has been taken. In order to achieve these objectives and generate a “sustainable growth”, a policy of encouraging distributed generation from RES, such as solar power must be followed. Intense concerns at European level regarding the Distributed Power from RES (DP- RES) were materialized by setting up a giant cluster of projects called Integration of Renewable Energy Sources and Distributed Generation into the European Electricity Grid, IRED cluster [2].

The studies that were conducted by this consortium highlighted the need for an energy management system at micro to macro level, the existing control strategies not being always successfully applied. National context. The percentage of RES in the electricity production in Romania is currently of 17.8%. EU has set a 24% target for Romania for energy generation from RES by 2020, but needs for investments and large operating costs as main barriers for the successful implementation of an increased generating capacity have also been identified. The compatibility with the EU objectives in the field of clean energy and national levels is achieved through a regional policy [3]. The European policies have had a national resonance since 2003, when the draft for the Strategy project for the use of renewable energy [4] was proposed. Two new important trends on the national energy market are to be noticed: firstly, the consumers’ involvement in the complex process of the energy efficient management from RES, and secondly, the increased attention paid to both the technical plan and the organizational and economical plan for the energy production from RES. The integration of RES in the national energy system structure has created a favorable context for our researches. Unfortunately, Romania follows a centralized approach of the regional policy and although the country is covered adequately with electricity networks and the potential development for RES is high, its aging infrastructure (30% of it was built in the 1960s) causes significant losses along the energy supply chain. Moreover, for a large number of energy resources, the current energy systems are hardly scalable. The European Commission believes that the current energy infrastructure is inadequate to connect and serve the entire Europe and recognizes the challenges [1].

In these contexts, two challenges have to be met: monitoring, diagnosis and forecasting functioning of grids, integrating renewable energy sources in on- grid or off- grid mode and ensuring the optimization of these ones functioning with the integrated, proactive management system based on multi-objective decision-making scheme. To support these interests, we consider that the integration of the predictive ability is a very important function in the command and control of a power system, especially at the micro- grid level, heavily dependent on the consumer and with reduced availability for storage. The challenge and the value added of our paper consist in proposing a forecasting tool to producers/ consumers (prosumers) of renewable energy resources, based on artificial intelligence techniques, trying to obtain optimal forecasts. The paper is organized in two sections: – first of all, there are recessed the most used criteria for choosing an optimal forecasting architecture, in relation with each step of the forecasting tool modeling: data preprocessing, forecasting tool architecture identification, parametrization and implementation; – secondly, the identified criteria are tested on two Adaptive Neuro- Fuzzy Inference System (ANFIS) models, in order to underline the effects of these users’ decisions over the forecasting performances.

REFERENCES

  1. Strategia Europa 2020, , http://ec.europa.eu/europe2020/europe2020-in-a-nutshell/priorities/index_ro.htm
  2. IRED cluster, http://www.ired-cluster.org/
  3. Politica de coeziune 2014-2020, http://ec.europa.eu/regional_policy/activity/ energy/index_ro.cfm
  4. Autoritatea Nationala de Reglementare in domeniul Energiei, http://www.anre.ro/documente.php?id=393
  5. DRAGOMIR, O., F. DRAGOMIR, E. MINCA, An Application Oriented Guideline for Choosing a Prognostic Tool, AIP Conference Proceedings: 2nd Mediterranean Conference on Intelligent Systems and Automation, vol. 1107, 2009, pp. 257-262.
  6. DRAGOMIR, O., F. DRAGOMIR, R. GOURIVEAU, E. MINCA, Medium Term Load Forecasting using ANFIS Predictor, Proceedings of the 18th IEEE Mediterranean Conference on Control and Automation, Marrakech, Morocco, pp. 551556, 2010.
  7. DRAGOMIR, O., DRAGOMIR, F., MINCA, E.,  Forecasting of renewable energy load with radial basis function (RBF) neural networks, Proceedings of the 8th Int. Conference on Informatics in Control, Automation and Robotics, Noordwijkerhout, Olanda, 2011.
  8. JANG, J.-S. R., C.-T. SUN, E. MIZUTAN, Neuro-Fuzzy and Soft Computing, Prentice Hall, 1997.
  9. LACROSE, A., A. TILTI, Fusion and Hierarchy Can Help Fuzzy Logic Controller Designer, Proceedings of IEEE International Conference on Fuzzy Systems, Barcelona, 1997.
  10. HEATON, J., Introduction to Neural Networks with JAVA, Heaton Research, Inc. www. heatonresearch.com
  11. HIPPERT, H. S., Neural Networks for Short-term Load Forecasting: A Review and Evaluation, IEEE Transactions on Power Systems, 2001.
  12. MAQSOOD, I., M. R. KHAN, A., ABRAHAM, Intelligent Weather Monitoring Systems using Connectionist Neural Models, Parallel & Scientific Computations, vol. 10, 2002, pp. 157-178.
  13. ELSHOLBERG, A., S. P. SIMONOVIC, U. S. PANU, Estimation of Missing Streamflow Data using Principles of Chaos Theory, Journal of Hydrology, vol. 255, 2002.
  14. SIFAOUI, A. ABDELKRIM, S. ALOUANE, M. BENREJEB, On New RBF Neural Network Construction Algorithm for Classification, Studies in Informatics and Control, ICI Publishing House, vol. 18, no. 2, pp. 6-17, 2009.
  15. BABUSKA, R., JAGER, R., VERBRUGGEN, H. B., Interpolation Issues in Sugeno-Takagi Reasoning, 3rd IEEE International Conference on Fuzzy Systems, pp. 859-863, Orlando, FL, 1994.
  16. FILDES, R., The Evaluation of Extrapolative Forecasting Methods, International Journal of Forecasting, vol. 8, 1992, pp. 81-98.
  17. De GOOIJER, J. G., HYNDMAN, R.J., 25 Years of Time Series Forecasting, International Journal of Forecasting, vol. 22, 2006, pp 443-473.
  18. ALFUHAID, A. S., EL-SAYED, M. A., MAHMOUD, M. S., Cascaded Artificial Neural Networks for Short-term Load Forecasting, IEEE Transactions on Power Systems, 12: 524–1529, 1997.
  19. CHOW, T. W. S., LEUNG, C. T., Neural Network based Short-term Load Forecasting using Weather Compensation, IEEE Transactions Power Systems, vol. 11, 1996, pp. 1736-1742.
  20. HOBBS, B. F., JITPRAPAIKULSARN, S., KONDA, S., CHANKONG, V., LOPARO, K. A., MARATUKULAM, D. J., Analysis of the Value for Unit Commitment of Improved Load Forecasting, IEEE Transactions on Power Systems, vol. 14, 1999, pp. 1342-1348.
  21. KARADY, G. G., FARNER, G. R., Economic Impact Analysis of Load Forecasting, IEEE Transactions on Power Systems, vol. 12, 1997, pp 1388-1392.
  22. ARMSTRON, J. S., COLLOPY, F., Error Measures for Generalizing about Forecasting Methods: Empirical Comparisons, International Journal of Forecasting, vol. 8, 1992, pp. 69–80.
  23. ARMSTRONG, J. S., FILDES, R., Correspondence on the Selection of Error Measures for Comparisons Among Forecasting Methods, Journal of Forecasting, vol. 14, 1995, pp. 67-71.
  24. MOHAMMED, O., PARK, D., MERCHANT, R., DINH, T., TONG, C., AZEEM, A., FARAH, J., DRAKE, C., Practical Experiences with an Adaptive Neural Network Short-term Load Forecasting System, IEEE Transactional Power Systems, vol. 10, 1995, pp. 254-265.
  25. PAPALEXOPOULOS, A. D., HAO, S., PENG, T. M., An Implementation of a Neural Network based Load Forecasting Model for the EMS, IEEE Transactional Power Systems, vol. 9, 1994, pp 1956-1962.
  26. PENG, T. M., HUBELE, N. F., KARADY, G. G., Advancement in the Application of Neural Networks for Short-term Load Forecasting, IEEE Transactional Power Systems, vol. 7, 1992, pp. 250-257.
  27. JANG, J. S. R., SUNI, C.T.,  MIZUTANI, E., Neuro-fuzzy and Soft Computing: A Computational Approach to Learning and Machine Intelligence, Prentice Hall, New York, 1997.
  28. NOROUZI, A., HAMEDI, M., ADINEH, V. R.,  Strength Modeling and Optimizing Ultrasonic Welded Parts of ABS-PMMA using Artificial Intelligence Methods, International Journal of Advanced Manufacturing Technologies, vol. 61, issue 1-4, 2012, pp. 135-147.
  29. JANG, J. S. R., ANFIS: Adaptive Network based Fuzzy Inference Systems, IEEE Transactions on Systems, Manufacture, Cybernetics, vol. 23(3), 1993, pp. 665-685.
  30. WANG, J. S., An Efficient Recurrent Neuro-fuzzy System for Identification and Control of Dynamic Systems, IEEE International Conference, Systems, Manufacture and Cybernetics, 2003.
  31. CHIANG, L.H., E. RUSSEL, R. BRAATZ, Fault Detection and Diagnosis in Industrial Systems, Springer, UK, 2001.
  32. TEODORESCU, H.-N., M.-D. ZBANCIOC, L. PISTOL, Parallelizing Neuro-fuzzy Economic Models in a GRID Environment, Studies in Informatics and Control, ICI Publishing House, vol. 17, no. 1, 2008, pp. 5-16.
  33. GOURIEROU, C., ARCH Models and Financial Applications, Springer, New York, 1997.

https://doi.org/10.24846/v24i3y201512