Drought Prediction Method Based on an Improved CEEMDAN-QR-BL Model

Yang Liu,Lihu Wang

doi:10.1109/access.2020.3048745

Abstract

In this study, a new broad learning (BL) model based on an improved complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) is proposed to resolve the low accuracy, poor robustness, and long delay problems that are present in current drought assessments. First, the extreme delay method was applied to improve the CEEMDAN end effect. The improved CEEMDAN method was then used to decompose a series of non-steady-state signals from drought monitoring into multiple steady-state components. A BL model based on orthogonal trigonometry (QR) was then used to predict these multiple steady-state components, and the predicted components were further reorganised to obtain a high-precision drought sequence. On this basis, CEEMDAN was introduced into the orthogonal triangular broad learning (QR-BL), and a drought prediction model (CEEMDAN-QR-BL) combining CEEMDAN and QR-BL was proposed. Finally, the De Martonne aridity index was used to calculate the drought sequence results and determine the drought grades. To meet the real-time requirements of drought prediction, parallel computing was introduced into the CEEMDAN-QR-BL model, and a drought prediction method based on parallel CEEMDAN-QR-BL was constructed. The experimental results show that, when compared with a support vector regression model combined with an empirical mode decomposition, the reliability and accuracy of the CEEMDAN-QR-BL increases by 29.57% and 11.84%, respectively. In addition, when compared with only BL, the prediction efficiency of QR-BL improved by 62.29%.

Highlights

Drought is a complex interdisciplinary issue involving meteorology, hydrology, geology, ecology, agriculture, the social economy, and other multi-disciplinary and multi-sectoral subjects
Combining the ensemble empirical mode decomposition (EEMD) with white noise and an artificial neural network, and subsequently applying it to drought prediction can effectively avoid the modal aliasing problem of the Empirical Mode Decomposition (EMD) method and further improve the accuracy of drought prediction [16]
DAN end effect with extreme extension was solved, the improved complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) with the improved width learning model was integrated, and this was combined with the parallel thinking to build a ‘decomposition-synthesis’ strategy based on the parallel CEEMDAN-QR-broad learning (BL) model; its basic realisation is as follows: 1) First, the two ends of the original signal were extended by the extreme value extension method, and the extended signal with CEEMDAN was decomposed to obtain N different proof modal components to complete the transition from the unsteady-state timing signal to the steady-state timing signal

Summary

Introduction

Drought is a complex interdisciplinary issue involving meteorology, hydrology, geology, ecology, agriculture, the social economy, and other multi-disciplinary and multi-sectoral subjects. The intermittency, uncertainty, and randomness of weather signals has brought great challenges to weather and drought forecasting. According to different prediction principles, drought prediction can be divided into mechanism-driven and data-driven models. Mechanism-driven models are often realised by modelling the relationship between drought indicators and potential drought factors [1]. The Multivariate Ensemble Streamflow Prediction model predicts future drought states by analysing the relationship between precipitation, soil moisture, and drought [2].

Methods

Results

Conclusion