کنفرانس بین المللی کنترل ، ابزار دقیق و اتوماسیون

صفحه اصلی / نهمین کنفرانس بین المللی کنترل ، ابزار دقیق و اتوماسیون

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

نویسندگان :

Mahasta Pahlawan¹ Roohollah Barzamini² Seyyed Abolfazl Yasini³

1- دانشگاه آزاد تهران مرکز 2- دانشگاه آزاد تهران مرکز 3- دانشگاه خواجه نصیرالدین طوسی

کلمات کلیدی :

Medium Term Load Forecasting (MTLF)،convolutional neural network،deep neural network،linear regression،long short-term memory neural network (LSTM)

چکیده :

This paper presents a high-accuracy prediction method for Medium Term Load Forecasting (MTLF) of a manufacturing plant, specifically Iran Khodro Company, using a convolutional neural network (CNN) and a long short-term memory neural network (LSTM). The performance of this method is compared with classical regression techniques such as linear regression, ridge regression, and lasso. The results demonstrate a coefficient of determination (r2_score) of 0.95 for the test data using the deep neural network algorithm, while the classical methods achieve an r2_score of 0.81. This significant difference highlights the superior capability of the proposed method. The model utilizes historical data based on past time of electric charge as input to train the deep learning-based neural network and implement the proposed algorithm. The monthly energy consumption data spanning 9 years from 2011 to 2019 for Iran Khodro Company is employed in this research.