New Hybrid Weighted Predictive Data Analysis Approach: Application to Temporal Power Transformer Maintenance Data
New Hybrid Weighted Predictive Data Analysis Approach: Application to Temporal Power Transformer Maintenance Data |
||
 |
 |
|
| © 2024 by IJETT Journal | ||
| Volume-72 Issue-8 |
||
| Year of Publication : 2024 | ||
| Author : Moïse Manyol, Samuel Eke, Georges Olong, Ndoumbè Matéké Max, Alain Biboum, Ruben Mouangue |
||
| DOI : 10.14445/22315381/IJETT-V72I8P129 | ||
How to Cite?
Moïse Manyol, Samuel Eke, Georges Olong, Ndoumbè Matéké Max, Alain Biboum, Ruben Mouangue, "New Hybrid Weighted Predictive Data Analysis Approach: Application to Temporal Power Transformer Maintenance Data," International Journal of Engineering Trends and Technology, vol. 72, no. 8, pp. 312-324, 2024. Crossref, https://doi.org/10.14445/22315381/IJETT-V72I8P129
Abstract
This paper proposes a new methodological approach for extracting typical data profiles from time series of power transformer maintenance databases with the aim of preventing future faults. The proposed approach operates in two stages. First, the model formalizes the date variable as a time series and then analyzes the data to identify interdependencies between them. Rio Tinto Alcan's dissolved gas data from transformer T0001 show that H2, CH4, C2H4, C2H6, CO, CO2, and N2 are dependent, while C2H2 and O2 are independent. In the second step, to capture short- and long-term trends, seasonalities, and dependencies in the data on the one hand and to extract non-linear trends and seasonalities for variable forecasting on the other, an ARIMA (Autoregressive Integrated Moving Average) + GES (Generalized Exponential Smoothing) model in combination is applied to the data series. The hybrid ARIMA (2, 1, 1) + GES (0.1, 0.1, 1) model, with a weighting of 0.5, produced errors of 0.27 and 4.5%, respectively, in terms of Mean Absolute Scaled Error (MASE) and Mean Symmetric Absolute Percentage Error (SMAPE). The ARIMA model taken individually gave MASE equals 0.55 and SMAPE equals 8.9%. Similarly, the proposed model is better than the naive model because the MASE is less than 1 (0.27%). The data series was subjected to other forecasting models, and it was found that the model proposed in this article is more accurate given the error results obtained since the smaller the error, the more accurate the forecasting model.
Keywords
Smoothing, Autoregression, Seasonality, Trend, Forecasting.
References
[1] Transformer Explosion Knocks Out Power In Northern N.J., Cbsnews, 2013. [Online]. Available: https://www.cbsnews.com/newyork/news/transformer-failure-knocks-out-power-in-northern-n-j/
[2] Daniel Kosiorowski et al., “Generalized Exponential Smoothing in Prediction of Hierarchical Time Series,” Statistics in Transition New Series, vol. 19, no. 2, pp. 331-350, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Samaher Al-Janabi et al., “Design and Evaluation of a Hybrid System for Detection and Prediction of Faults in Electrical Transformers,” International Journal of Electrical Power & Energy Systems, vol. 67, pp. 324-335, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Aman Samson Mogos, Xiaodong Liang, and Chi Yung Chung, “Distribution Transformer Failure Prediction for Predictive Maintenance Using Hybrid One-Class Deep SVDD Classification and Lightning Strike Failures Data,” IEEE Transactions on Power Delivery, vol. 38, no. 5, pp. 3250-3261, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Oussama Laayati et al., “Smart Energy Management System: Oil Immersed Power Transformer Failure Prediction and Classification Techniques Based on DGA Data,” 2022 2nd International Conference on Innovative Research in Applied Science, Engineering and Technology, Meknes, Morocco, pp. 1-6, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[6] J.R. Jimenez-Octavio et al., “New Transformer Sizing Criteria Based on Moving Average Techniques for Railway Systems,” ASME/IEEE 2007 Joint Rail Conference and Internal Combustion Engine Division Spring Technical Conference, Pueblo, Colorado, USA, pp. 327-332, 2009.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Abdolrahman Peimankar et al., “Multi-Objective Ensemble Forecasting with an Application to Power Transformers,” Applied Soft Computing, vol. 68, pp. 233-248, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Juan L. Velasquez-Contreras, Miguel A. Sanz-Bobi, and Samuel Galceran Arellano, “General Asset Management Model in the Context of an Electric Utility: Application to Power Transformers,” Electric Power Systems Research, vol. 81, no. 11, pp. 2015-2037, 2011.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Allou Samé et al., “Decomposition and Classification of Functional Data for Water Consumption Analysis,” International Francophone Conference on Knowledge Extraction and Management, Clustering and Co-clustering Workshop, Reims, France, pp. 1-11, 2016.
[Google Scholar] [Publisher Link]
[10] Pere Izquierdo Gómez et al., “Data-Driven Thermal Modelling for Anomaly Detection in Electric Vehicle Charging Stations,” 2022 IEEE Transportation Electrification Conference & Expo, Anaheim, CA, USA, pp. 1005-1010, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Yunus Biçen, Faruk Aras, and Hulya Kirkici, “Lifetime Estimation and Monitoring of Power Transformer Considering Annual Load Factors,” IEEE Transactions on Dielectrics and Electrical Insulation, vol. 21, no. 3, pp. 1360-1367, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[12] C.C. Thompson, A.D. Stringer, and C.I. Barriga, “An Evaluation of Transformer Historical Failure Data for Facility Resiliency and Reliability,” 2019 IEEE Power & Energy Society General Meeting, Atlanta, GA, USA, pp. 1-5, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Belkacem Balah, “Frontiers of Use of Monthly Rainfall Forecasting with the Arima Model in Agriculture,” 1st International Seminar Agriculture 4.0: Rural Engineering at the Service of the Environment, Algiers, Algeria, pp. 1-6, 2018.
[Google Scholar] [Publisher Link]
[14] Dominique Desbois, “An Introduction to Box and Jenkins Methodology: Using ARIMA Models with SPSS,” MODULAD Review, pp. 1-25, 2005.
[Google Scholar] [Publisher Link]
[15] Régis Bourbonnais, and Virginie Terraza, Time Series Analysis - 5th ed. Course and Corrected Exercises - Applications to Economics and Management, Dunod, pp. 1-368, 2022.
[Google Scholar] [Publisher Link]
[16] Jean-Marie Dufour, “Trend Extraction and Seasonal Adjustment Using Moving Average Method,” McGill University, pp. 1-36, 2003.
[Google Scholar] [Publisher Link]
[17] Jacques Padet, “Some Analytical Properties of Moving Averages, and their Application to the Detection of Trends in a Signal,” Entropy: Thermodynamics – Energy – Environment – Economy, vol. 5, no. 2, pp. 1-20, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[18] M. Grun Rehomme, and D. Ladiray, “Centered and Non-Centered Moving Averages, Construction and Comparison,” Journal of Applied Statistics, vol. 42, no. 3, pp. 33-61, 1994.
[Google Scholar] [Publisher Link]
[19] Patrícia Ramos, and José Manuel Oliveira, “A Procedure for Identification of Appropriate State Space and ARIMA Models Based on Time-Series Cross-Validation,” Algorithms, vol. 9, no. 4, pp. 1-14, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Didier Delignieres, “Time Series–ARIMA Models, EA Seminar “Sport–Performance–Health,” Mars, pp. 1-19, 2000.
[Google Scholar] [Publisher Link]
[21] Youssef Hmamouche, “Prediction of Wide Time Series,” Doctoral Dissertation, Aix Marseille University, pp. 1-108, 2018.
[Google Scholar] [Publisher Link]
[22] S. Dhamodharavadhani, and R. Rathipriya, “Region-Wise Rainfall Prediction Using MapReduce-Based Exponential Smoothing Techniques,” Advances in Big Data and Cloud Computing, Advances in Intelligent Systems and Computing, vol. 750, pp. 229-239, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Alexei Botchkarev, “Evaluating Performance of Regression Machine Learning Models Using Multiple Error Metrics in Azure Machine Learning Studio,” SSRN, pp. 1-16, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Chao Chen, Jamie Twycross, and Jonathan M. Garibaldi, “A New Accuracy Measure Based on Bounded Relative Error for Time Series Forecasting,” PLoS One, vol. 12, no. 3, pp. 1-23, 2017.
[CrossRef] [Google Scholar] [Publisher Link]