A Medium-Term Load Forecasting of Iran Khodro Company Using Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) Deep Neural Networks

Subject Areas : International Journal of Smart Electrical Engineering

Mahasta Pahlawan ¹ , Roohollah Barzamini ² , Shoorangiz Shams Shamsabad Farahani ³ , Seyyed Abolfazl Yasini Shiadeh ⁴ , Nargess Sadeghzadeh-Nokhodberiz ⁵ , Mohammad Arabian ⁶

1 - Department of Electrical Engineering, Islamic Azad University Tehran Central Branch,
2 - Department of Electrical Engineering, Central Tehran, Islamic Azad University, Tehran, Iran
3 - Department of Electrical Engineering, Islamshahr branch, Islamic Azad University
4 - Iran Khodro Company
5 - Department of electrical and computer Engineering, Qom University of Technology
6 - Department of Electrical Engineering, Central Tehran, Islamic Azad University, Tehran, Iran

Received: 2025-02-17 Accepted : 2025-07-21 Published : 2025-07-24

Keywords: Convolutional neural network, deep neural network, linear regression, long short-term memory neural network (LSTM), medium Term Load Forecasting (MTLF) ,

Abstract :

This paper introduces a high-accuracy prediction framework for Medium Term Load Forecasting (MTLF) in a manufacturing plant at Iran Khodro Company, leveraging the power of deep learning. Specifically, the proposed method integrates a Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) neural network, effectively capturing both spatial and temporal dependencies in the data. The performance of this hybrid deep learning model is rigorously evaluated against traditional regression techniques, including Linear Regression, Ridge Regression, and Lasso Regression. experimental results demonstrate a remarkable coefficient of determination (R² score) of 0.95 on the test dataset when employing the deep neural network model, significantly outperforming classical methods, which achieve an R² score of only 0.81. This substantial improvement underscores the superior predictive accuracy and generalization capability of the proposed approach. the model is trained using historical energy consumption data, where past electricity load values serve as inputs for the deep learning architecture. The dataset consists of nine years of monthly energy consumption records (2011–2019) collected from Iran Khodro Company, providing a robust foundation for medium-term load forecasting. The findings of this study highlight the effectiveness of deep learning in industrial energy demand prediction, offering a reliable and scalable solution for optimizing energy management in manufacturing sectors.

References:

[1] LeCun, Y., Bengio, Y., Hinton, G. ,2015. Deep learning. Nature 521: 436–444
[2] LeCun, Y., Bengio, Y. and Hinton, G., Deep learning, Nature, vol. 521, no. 7553, pp. 436-444, May 2015.Available: https://doi.org/10.1038/nature14539
[3] Binghui, Xu, 2024. Multi-Channel Convolutional Neural Network with Long Short-Term Memory for Power Grid Safety Monitoring, Second International Conference on Data Science and Information System (ICDSIS).

[4] Shams Shamsabad Farahani, S., Nekoui. M.A., Yazdanpanah-Goharrizi. A., 2006. Predictive Control of a Heat exchanger Based on Local Fuzzy Models and Neural Networks, Istanbul university journal of Electrical and Electronics engineering, volume 6, pp.129-138.
[5] Shams Shamsabad Farahani, S., Arefi, M.M., Zaeri, A.H., 2020. Electroencephalography Artifact Removal using Optimized Radial Basis Function Neural Networks, Majlesi Journal of Electrical Engineering, 14(4).
[6] Kumar, K. S , Goyal, A.K., Sain, M. , 2024. Lung Cancer Detection Using Convolutional Neural Network in Machine Learning, 15th International Conference on Computing Communication and Networking Technologies (ICCCNT).

[7] Givechin Koohi, M. Shams Shamsabad Farahani, S., 2019. Identification and Detection of Healthy and Non - Healthy Cells in Breast Cancer using Neural Networks, 5th national conference on Electrical engineering and Mechatronics, Tehran, Iran, (in Persian).
[8] Wan, Z., Zhang, T. , 2021. Skin Lesions Detection via Convolutional Neural Networks, 2nd International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)
[9] Shams Shamsabad Farahani, S., Nekouei M.A., Nikzad, M., 2009. Predictive Control of a Single Link Flexible Joint Robot Based on Neural Network and Feedback Linearization, Australian Journal of Basic and Applied Sciences, 3(3): 2322-2333.
[10] Shams Shamsabad Farahani, S., Pourmahdian, H., 2024. Neural network-based control schemes applied to two-link robots, a review, in the 7th national conference on new Technologies in Electrical, Computer and Mechanical engineering of Iran.

[11] Liu, Wen-Bin, Bui, Thuc-Minh, Quoc, Tien Nguyen, et.al., 2024. Failure Classified Method for Diesel Generators Based Long Short-Term Memory Approach, International Conference on System Science and Engineering (ICSSE), 2024.
[12] Wang, Y., Yang, L. , Jia, X. , Liu, X. , Wang, F. , 2024. Automatic Generation Control of Power System Based on Lexicon Long Short Term Memory, International Conference on Data Science and Network Security (ICDSNS),2024.
[13] Jin, F. , Ran, J. , 2024. Short-Term Wind Power Prediction Based on IAVOA-LSTM Neural Network, 7th International Conference on Computer Information Science and Application Technology (CISAT).
[14] Ehsani, M., Oraee, A., Abdi, B. et al. 2024.Adaptive dynamic sliding mode controller based on extended state observer for brushless doubly fed induction generator. Int. J. Dynam. Control 12, 3719–3732.
[15] Ehsani, M., Oraee, A., Abdi, B., Behnamgol, V., Hakimi, S. M., 2023. Adaptive Dynamic Sliding Mode Algorithm for BDFIG Control. IJEEE Iranian Journal of Electrical and Electronic Engineering, 19 (1).
[16] Kim, K.N.L., Tien Nguyen Quoc;Phuong Nguyen Thanh, 2024. Deep Learning Method For Load Forecast In Smart Solar System Based Long Short-Term Memory, 8th International Artificial Intelligence and Data Processing Symposium (IDAP).
[17] Idil Sülo;Seref Recep Keskin;Gülüstan Dogan;Theodore Brown, 2019. Energy Efficient Smart Buildings: LSTM Neural Networks for Time Series Prediction, International Conference on Deep Learning and Machine Learning in Emerging Applications (Deep-ML).
[18] Li, L., Guo, L., Wang, J., and Peng, H., 2023. Short-Term Load Forecasting Based on Spiking Neural P Systems, Appl. Sci., vol. 13, no. 2.
[19] Vanting, N.B., Ma, Z., and Jørgensen, B.N., 2022. Evaluation of neural networks for residential load forecasting and the impact of systematic feature identification, Energy Informatics, vol. 5.
[20] Barzamini R., F. Hajati, S. Gheisari, and M.B. Motamadine, 2012. "Short term load forecasting using multi-layer perception and fuzzy inference systems for Islamic countries," Journal of Applied Sciences, 12(1), pp. 40-47. 2012.
[21] Barzamini, R., M.B. Menhaj, S. Kamalvand, and A. Tajbakhsh, 2005. Short term load forecasting of Iran national power system using artificial neural network generation two, in Proceedings of the 2005 IEEE Russia Power Tech, Moscow, Russia, pp. 1-6.
[22] Barzamini, R., M.B. Menhaj, A. Khosravi, and S.H. Kamalvand, 2005, Short term load forecasting for Iran national power system and its regions using multi-layer perceptron and fuzzy inference systems, in Proceedings of the 2005 IEEE International Joint Conference on Neural Networks, Montreal, QC, Canada, vol. 1, pp. 495-500.
[23] Abdulameer Alhammadi , Farzaneh Rahmani , Izadi A. , Hajati F. , Shams Shamsabad Farahani S., Jabr A., AL-Salman W., Saneii, S.M., , Barzamini, R., 2024. Prediction of Environmental Conditions of the Greenhouse Using Neural Networks Optimized with the Grasshopper Optimization Algorithm (GOA), Power system Technology, Volume 48 Issue 3, pp.622-635.
[24] Zhang, Y. and LeCun, Y., Text Understanding from Scratch, Available: https://arxiv.org/abs/1502.01710
[25] Chollet, F., 2017. Deep Learning with Python, Manning Publications, Available: https://www.manning.com/books/deep-learning-with-python
[26] Barzamini, H., Rahimi, M., Shahzad, M., and Alhoori, H., 2022. Improving Generalizability of ML-enabled Software through Domain Specification, in IEEE International Conference on Software Maintenance and Evolution (ICSME), pp. 1-12.
[27] Géron, A. 2019. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, O'Reilly Media,
[28] Dudek, G., 2015. Pattern-based local linear regression models for short-term load forecasting, Electric Power Systems Research, vol. 126, pp. 9-17.
[29] Du, D., Xie, J., and Fu, Z., 2018. Short-Term Power Load Forecasting Based on Spark Platform and Improved Parallel Ridge Regression Algorithm, in IEEE Transactions on Power Systems, 33(4), pp. 4004-4012.
[30] Ziel, F. and Liu, B., 2016. Lasso estimation for GEFCom2014 probabilistic electric load forecasting, International Journal of Forecasting, 32(3), pp. 1022-1027,
[31] Lu, S., Xu, Q., Jiang, C., Liu, Y., and Kusiak, A., 2022. Probabilistic load forecasting with a non-crossing sparse-group Lasso-quantile regression deep neural network, Energy, vol. 246.