Research on a multiparameter water quality prediction method based on a hybrid model

Abstract

Watershed water quality monitoring is of great significance in the protection and management of water environments. Because existing water quality prediction algorithms cannot achieve high-precision multiparameter analysis and usually require a large amount of data, this paper proposes the VARLST hybrid water quality prediction model. The proposed model combines the traditional statistical vector autoregressive moving average model and the bidirectional long short-term memory neural network to achieve multiparameter water quality data prediction on small data samples, and the model data processing is simple and highly efficient. This model is used to analyze the characteristics of water quality parameters in the Inner Mongolian section of the Yellow River Basin in northern China. The average error and fitting accuracy of the prediction results are 0.0015 and 99.869%, respectively; hence, the model achieves high-accuracy prediction of multiparameter indicators using less data, outperforming single models in terms of feasibility and accuracy.