Precise Assimilation Prediction of Short-Term and Long-Term Maize Irrigation Water Based on EnKF-DSSAT and Fuzzy Optimization-DSSAT Models

Abstract

With the progress of information technology, precision irrigation technology has become the core of modern agriculture. In particular, technologies such as Internet of Things (IoT), Big Data and Artificial Intelligence (AI) have provided strong support for the intelligence of agricultural production. This paper focuses on the precise prediction of irrigation water use in the maize industry through Ensemble Kalman Filter (EnKF) and fuzzy optimization methods combined with the DSSAT (Decision Support System for Agrotechnology Transfer) model. We use the remote sensing data of land moisture and leaf area in the Yellow Huaihai Plain provided by Google Earth Engine (GEE), as well as the maize market data released by the Ministry of Agriculture (MOA), to make predictions through the EnKF-DSSAT and fuzzy optimization-DSSAT models. The results showed that these models achieved high accuracy of 98.11% and 97.78% in short-term and long-term forecasts, respectively, which were significantly better than the traditional models. We also introduce a Boltzmann machine-based fusion algorithm to improve the model convergence speed and prediction accuracy. Ultimately, this paper verifies the important influence of policy factors on long-term irrigation prediction and proposes an adaptive prediction model and policy recommendations, which provide innovative methods and technical support for the implementation of precision irrigation technology.

Keywords

• Data assimilation
• DSSAT model
• ensemble Kalman filter
• fuzzy optimization
• precision irrigation