A review of using neural network and Kalman filter based on ARIMA for wind speed forecasting

العناوين الأخرى

دراسة مراجعة في استخدام الشبكات العصبية و مرشح كالمن بالاعتماد على ARIMA للتكهن بسرعة الرياح

المؤلفون المشاركون

al-Hannun, Usamah Bashir Shukr
Lee, Muhammad Hisyam

المصدر

المجلة العراقية للعلوم الإحصائية

الناشر

جامعة الموصل كلية علوم الحاسبات و الرياضيات

تاريخ النشر

2019-12-31

دولة النشر

العراق

عدد الصفحات

14

التخصصات الرئيسية

الرياضيات

الموضوعات

• الرياح
• الشبكات العصبية(الحاسبات الإلكترونية)
• السرعة

الملخص الإنجليزي

The wind speed forecasting is important to observe the wind behaviour and control the harms caused by extreme speeds.

A linear ARIMA model is unable to identify the nonlinear pattern of wind speed data.

ARIMA modelling process causes the stochastic uncertainty as a second reason of inaccurate forecasting results.

In this study, a review of an ARIMA-artificial neural network (ANN) and ARIMA-Kalman filter (KF) methods is presented.

ANN and KF based on ARIMA are used to improve wind speed forecasting by handling the nonlinearity and the uncertainty respectively.

This hybrid method will improve the accuracy of wind speed forecasting.

Usually, the forecasting results of the hybrid methods will be in better forecasting than other compared methods.

In conclusion, the proposed hybrid methods especially the hybrid ANN-KF based on ARIMA method can be used to forecast wind speed data with nonlinearity and uncertainty characteristics more accurately

نوع البيانات

أوراق مؤتمرات

رقم السجل

BIM-900004

• APA
• MLA
• AMA