Digital twin-based winter wheat growth simulation and optimization

Abstract

Context

Wheat is an essential food crop, and precise simulations and feedback mechanisms during its growth are crucial for advancing the intelligence of wheat production systems.

Objective

(1) Building Virtual Real Interaction of Data. (2) Constructing twin simulation of winter wheat growth process. (3) Constructing a feedback control technology for winter wheat growth process based on digital twins.

Methods

This study utilized digital twin technology integrated with crop growth models to optimize monitoring and management processes through sequential experiments simulating wheat growth. By utilizing Internet of Things (IoT) devices and drones, the integration of wheat growth data and the creation of a digital twin environment were achieved. The integration of digital twin technology with crop growth models allowed precise simulation and intelligent management of wheat growth processes.

Results

The wheat growth digital twin model, developed based on the DSSAT framework, can effectively simulate wheat growth. Model calibration and dynamic parameter adjustments resulted in an R² of 0.98 for the simulation accuracy of LAI (leaf area index) and AGB (above-ground biomass). Simulation errors for flowering and maturity stages were 0.6 days and 1.1 days, respectively, while yield simulation errors remained below 0.6 t/hm². Additionally, optimal management strategies were proposed for various winter wheat varieties.

Conclusions

Digital twin technology enables precise simulation of wheat growth, supports effective feedback regulation, and significantly enhances the intelligence of wheat production.