A Hybrid Approach using ARIMA, Kalman Filter and LSTM for Accurate Wind Speed Forecasting

TL;DR

This paper proposes a hybrid model combining ARIMA, Kalman filter, and LSTM to improve wind speed forecasting accuracy, addressing the intermittency challenge of wind energy for better renewable energy utilization.

Contribution

A novel hybrid approach integrating ARIMA, Kalman filter, and LSTM for more accurate wind speed prediction compared to existing methods.

Findings

The hybrid model outperforms existing wind speed forecasting methods.

Simulation results show improved accuracy of the proposed approach.

The method effectively captures wind speed variability for renewable energy planning.

Abstract

Present energy demand and modernization are leading to greater fossil fuel consumption, which has increased environmental pollution and led to climate change. Hence to decrease dependency on conventional energy sources, renewable energy sources are considered. Wind energy is a long-term renewable energy resource but its intermittent nature makes it difficult in harnessing it. Since wind speed prediction is vital there are different methodologies for wind speed estimation available in the literature. In this work, a new hybrid model is proposed by combining auto-regressive integrated moving average (ARIMA), Kalman filter and long short-term memory (LSTM) for estimating wind speed which works more accurately than the existing methods proposed in the literature. From simulations, it is observed that the proposed method works with better accuracy when compared to the existing methods.