Prediction of the Ammonia Nitrogen Content with Improved Grey Model by Markov Chain

Abstract

Water pollution prediction plays a crucial role in environmental protection and sustainable development. This study proposes an innovative approach to enhance the accuracy of water pollution prediction by combining the grey prediction model (GM) with Markov chain analysis. This research focuses on predicting the concentration of ammonia nitrogen (NH3-N) in Dongting Lake, a significant water body. Grey prediction models (GM) are utilized to forecast NH3-N content, addressing the challenge posed by incomplete or insufficient data. However, due to the dynamic nature of water quality indicators, GM models may have limitations in terms of accuracy. To overcome this issue, this study introduces the concept of Markov chains, incorporating historical state transitions into prediction models to achieve more precise forecasts. The research demonstrates a novel method for water pollution prediction that integrates GM models with Markov chain analysis, resulting in improved accuracy when predicting NH3-N concentrations. A comparison with traditional GM predictions highlights the effectiveness of this approach. The model's performance was evaluated using actual data from the China Automated Water Quality Monitoring Report. Combining grey prediction models with Markov chains outperforms traditional methods when it comes to predicting water pollution levels. The result contributes to advancing the field of water pollution forecasting by enhancing forecasting accuracy and providing informed decision support for environmental protection and management purposes.