کنفرانس بین المللی فناوری و مدیریت انرژی

صفحه اصلی / نهمین کنفرانس بین المللی فناوری و مدیریت انرژی

A Hybrid CNN-LSTM Approach Utilizing SCADA Data Enhancing Wind Power Predictions: A Dataset Case Study of Turkey

نویسندگان :

Hamed Kheirandish Gharehbagh¹ Ashkan Safari² Kazem Zare³ Amir Aminzadeh Ghavifekr⁴

1- دانشگاه تبریز 2- دانشگاه تبریز 3- دانشگاه تبریز 4- دانشگاه تبریز

کلمات کلیدی :

Wind Turbine،Power Prediction،SCADA Data،Renewable Energy،Forecasting،Artificial Intelligence

چکیده :

Abstract— In this study, a hybrid deep learning model combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks is developed to predict wind turbine output power using data obtained from the Supervisory Control and Data Acquisition (SCADA) system. The model was trained and evaluated on a dataset derived from real-world wind turbine operations in Turkey. The results demonstrate the effectiveness of the proposed approach, with a Mean Squared Error (MSE) of 0.0043, indicating a high level of accuracy in predicting power output. Additionally, the Mean Absolute Percentage Error (MAPE) of 3.02% highlights the model's ability to make precise predictions, while the Root Mean Squared Percentage Error (RMSPE) of 0.1631 further underlines its accuracy. The high R-squared (R2) value of 0.9682 indicates an excellent fit of the model to the observed data, explaining approximately 96.82% of the variance.