Research on hydrological survey and forecasting model based on rolling sampling Markov chain

Abstract

Abstract Hydrological survey is of great significance in flood disaster warning, hydropower engineering construction, and efficient utilization of watershed resources. Due to the seasonal and random fluctuations of hydraulic resources, it is difficult to accurately predict waterway runoff. Based on this, this article proposes a hydrological survey and prediction method based on rolling sampling Markov chain. Firstly, based on historical hydrological and meteorological data, establish a watershed runoff model and summarize and analyze the characteristics of the patterns; Then, based on the first-order Markov chain model, the relationship between adjacent days is considered. According to seasonal and meteorological factors, the historical data is rolling sampled in a 10 day sampling interval, and a multi state transition probability matrix is established to construct a year-round time-series runoff forecasting model; Finally, taking the Jinsha River Basin as an example, the annual runoff situation of the waterway was simulated. The calculation results have verified the effectiveness of the proposed method, indicating that it can simulate runoff well and provide certain guidance for watershed management and construction.