Dynamic Error-Correlation Adaptive Weighting for Hybrid Energy Consumption Forecasts

Daniel-Costin Ebanca, Irina Valentina Tudor, Cristina Popirlan

Abstract


Accurate energy consumption forecasting is essential for grid stability, operational planning, and resource optimization. This study investigates the effectiveness of hybrid forecasting strategies by combining four widely used time series models ARIMA, ETS, THETAM, and NNETAR and proposes a Dynamic Error-Correlation Adaptive Weighting (DECAW) mechanism for ensemble construction. The proposed framework integrates recency-weighted validation errors and explicit penalization of forecast error correlation to enhance diversification and robustness. Using a dataset of electricity consumption collected from a medium-sized Romanian city, the performance of individual models and hybrid approaches was assessed and compared. Results reflect the dominance of linear seasonal dynamics in the dataset. Hybrid models produced competitive results, with DECAW yielding the lowest hybrid MAPE and MAE and marginally improving upon conventional cross-validated weighting. DECAW con- sistently redistributed weights toward structurally complementary components and reduced the influence of correlated or unstable predictors. The findings demonstrate that correlation aware adaptive weighting enhances ensemble stability and robustness, particularly in moderately complex seasonal systems.


Full Text:

PDF

References


N. Roustaei, Application and interpretation of linear-regression analysis, Medical hypothesis discovery and innovation in ophthalmology 13 (2024), no. 3, 151–159.

V. Arumugam, V. Natarajan, Time series modeling and forecasting using Autoregressive Integrated Moving Average and Seasonal Autoregressive Integrated Moving Average models,

Instrumentation Mesure Metrologie 22 (2023), no. 4, 161–168.

J.S. Chou, D.S. Tran, Forecasting energy consumption time series using machine learning techniques based on usage patterns of residential householders, Energy 165 (2018), no. B, 709– 726.

H. Hamdoun, A. Sagheer, H. Youness, Energy time series forecasting-analytical and empirical assessment of conventional and machine learning models, Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology 40 (2021), no. 6, 12477–12502.

H.R. Maier, S. Galelli, S. Razavi, A. Castelletti, A. Rizzoli, I.N. Athanasiadis, M. Sanchez-Marre, M. Acutis, W. Wu, G.B. Humphrey, Exploding the myths: An introduction to artificial neural networks for prediction and forecasting, Environmental Modelling & Software

(2023), 105776.

S. Zhang, J. Liu, J. Wang, High-Resolution Load Forecasting on Multiple Time Scales Using Long Short-Term Memory and Support Vector Machine, Energies 16 (2023), no. 4, 1806.

S. Ali, S. Bogarra, M.N. Riaz, P.P. Phyo, D. Flynn, A. Taha, From Time-Series to Hybrid Models: Advancements in Short-Term Load Forecasting Embracing Smart Grid Paradigm, Applied Sciences 14 (2024), no. 11, 4442.

S. Baratsas, F. Iseri, E.N. Pistikopoulos, A hybrid statistical and machine learning based forecasting framework for the energy sector, Computers & Chemical Engineering 188 (2024), 108740.

E.G.A. Antonini, A. Di Bella, I. Savelli, L. Drouet, M. Tavoni, Weather- and climate-driven power supply and demand time series for power and energy system analyses, Scientific Data 11 (2024), 1324.

X. Wen, J. Liao, Q. Niu, N. Shen, Y. Bao, Deep learning-driven hybrid model for short-term load forecasting and smart grid information management, Scientific Reports 14 (2024), 13720.

J.P. Mohanty, S. Dash, Forecasting Energy Consumption Using Hybrid CNN and LSTM Auto-Encoder Network with Hyperband Optimization, International Journal for Research in Applied Science & Engineering Technology (IJRASET) 10 (2022), no. X, 1041–1055.

I.H. Ou Ali, A. Agga, M. Ouassaid, M. Maaroufi, A. Elrashidi, H. Kotb, Predicting short-term energy usage in a smart home using hybrid deep learning models, Frontiers in Energy Research 12 (2024), 1323357.

L. Gasmi, S. Kabou, N. Laiche, R. Nichani, Time series forecasting using deep learning hybrid model (ARIMA-LSTM), Studies in engineering and exact sciences 5 (2024), no. 2, e6976.

S.Y. Chou, A. Dewabharata, F.E. Zulvia, M. Fadil, Forecasting Building Energy Consumption Using Ensemble Empirical Mode Decomposition, Wavelet Transformation, and Long Short-Term Memory Algorithms, Energies 15 (2022), no. 3, 1035.

R. Mathumitha, P. Rathika, K. Manimala, Intelligent deep learning techniques for energy consumption forecasting in smart buildings: a review, Artificial Intelligence Review 57 (2024), 35.

A. Salhi, F. Algarni, R. Alshamrani, A. Althbiti, A. Ismail, M. Basma, Hassan Leveraging artificial intelligence to enable sustainable urban development through the creation of smart and environmentally friendly carbon-free cities, Scientific Reports 15 (2025), 35791.

J.O. Ojadi, O.K. Owulade, C.S. Odionu, E.C. Onukwulu, Deep Learning Models for Predicting and Mitigating Environmental Impact of Industrial Processes in Real-Time, International Journal of Scientific Research in Science, Engineering and Technology 12 (2025), no. 2, 119–154.

A.R. Singh, M.S. Sujatha, A.D. Kadu, M. Bajaj, H.K. Addis, K. Sarada, A deep learning and IoT-driven framework for real-time adaptive resource allocation and grid optimization in smart energy systems, Scientific Reports 15 (2025), 19309.

N. Gholizadeh, P. Musilek, Federated learning with hyperparameter-based clustering for electrical load forecasting, Internet of Things 17 (2022), 100470.

V.S.P. Rajulapati, Experimental and Computational Approaches to AI-Driven Load Forecasting and Dynamic Pricing in Smart Grids, International Journal of Computational and Experimental Science and Engineering 11 (2025), no. 3.

S.S. Krishna, R. Krishan, Day Ahead Solar Generation Forecasting in Smart Grids to Minimize Electricity Bill, International Journal for Research in Applied Science & Engineering Technology (IJRASET) 11 (2023), no. IX, 1230–1256.

J. Pascual, J. Barricarte, P. Sanchis, L. Marroyo, Energy management strategy for a renewable-based residential microgrid with generation and demand forecasting, Applied Energy 158 (2015), 12–25.

A. Piacentino, N. Duic, N. Markovska, B. Vad Mathiesen, Z. Guzovic, V. Eveloy, H. Lund, Sustainable and cost-efficient energy supply and utilisation through innovative concepts and technologies at regional, urban and single-user scales, Energy 182 (2019), 254–268.

M.A. Mirjalili, A. Aslani, R. Zahedi, M. Soleimani, A comparative study of machine learning and deep learning methods for energy balance prediction in a hybrid building-renewable energy system, Sustainable energy research 10 (2023), no. 8.

F. Moazzen, M.J. Hossain, Multivariate Deep Learning Long Short-Term Memory-Based Forecasting for Microgrid Energy Management Systems, Energies 17 (2024), no. 17, 4360.

G.E.P. Box, G.M. Jenkins, G.C. Reinsel, G.M. Ljung, Time Series Analysis: Forecasting and Control, Journal of Time Series Analysis 37 (2015), no. 5, 709–711.

S. Varma, R. Simon, Bias in error estimation when using cross-validation for model selection, BMC Bioinformatics 7 (2006), 91.

V. Assimakopoulos, K. Nikolopoulos, The theta model: a decomposition approach to forecasting, International Journal of Forecasting 16 (2000), no. 4, 521–530.

R. Hyndman, A. Koehler, K. Ord, R. Snyder, Forecasting with Exponential Smoothing, Springer Series in Statistics, 2008.

R.J. Hyndman, Y. Khandakar, Automatic Time Series Forecasting: The forecast Package for R, Journal of Statistical Software 27 (2008), no. 3, 1–22.




DOI: https://doi.org/10.52846/ami.v53i1.2341