Under the background of intelligent water conservancy, the accurate monitoring and prediction of lake water level has become a key factor to ensure the hydrological balance of the basin, maintain the health of the ecosystem and promote social stability. This paper takes the monthly water level data of Dongping Lake from 2010 to 2024 as the research object, and deeply explores the dynamic prediction method of lake water level in complex changing environment. The seasonal autoregressive integrated moving average model (ARIMA (2,1,1)) was introduced to establish and screen out the model with the best prediction performance for the water level of Dongping Lake. Then, the ARIMA (2,1,1) model was used to predict the water level data of Dongping Lake from January to June 2024. The results showed that the predicted value was basically similar to the measured value, which fully verified the accuracy and practicability of the model in the complex and changeable environment. The research results can provide reference for the construction of the digital twin model of Dongping Lake, and also provide reference for the water level prediction of similar lakes.