Water Quality Prediction of Liuxi River based on ICEEMDAN-WOA-LSTM Model

Water quality prediction is a crucial task for ensuring the safety and sustainability of water sources. It supports water resource managers in making informed decisions and taking appropriate actions to prevent water pollution and maintain water quality standards. Considering the non-linearity, latency, and complexity of water quality data, this study proposes a combined model for water quality prediction based on Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) and Long Short-Term Memory (LSTM) networks. The model decomposes and reconstructs variables from time series data to achieve water quality predictions. Comparative experiments with other network models demonstrate superior performance of the proposed model.