Daily average relative humidity forecasting via two LSTM-attention methods

• DOI: 10.23919/ccc55666.2022.9902384
• Publication Date: 2022-10-11T19:33:35Z
• Authors: Han Wu; Yan Liang; Pan-hai Zheng
• Keywords: 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

The daily average relative humidity is significant for both agriculture and industry. Due to high stochastic, intermittent non-linear characteristics by nature, the accurate forecasting of a very challenging task. For improving performance, two LSTM-attention methods with attention mechanism added after input before output are developed in this paper. First, meteorological data during 1 January 1999 31 December 2017 from station Shaanxi, China, were analyzed, where rainfall mean transformed Log operations reduce fluctuations make their distributions close normal distribution. Then, designed forecast humidity, LSTM used extract time-varying features automatically mine internally causal relationships, mechanisms applied improve accuracy. Experimental results suggest that method gains better performance than baseline MSE, RMSE MAE.