Optimization study on spatial distribution of rice based on a virtual plant approach

Abstract

How to increase crop yield is the most important issue in agricultural production. Many studies have been devoted to optimizing spatial distribution of crops, to improve light interception and increase photosynthetic assimilation. However, finding an optimal solution based on field experiments is almost impossible since the large number of combinations of factors that are related, and the cost in terms of finances and time are prohibitive. A new optimization strategy was proposed in this study, integrating a Functional-Structural Model of rice with a workflow based on a Mixed Particle Swarm Optimization (MPSO) algorithm. The 3D modelling platform GroIMP was used to implement the model and optimization workflow. MPSO is a new Particle Swarm Optimization-based algorithm with multistage disturbances, which has improved abilities to get rid of local optima and to explore solution space. Spacing between plants was used as optimization target in the first example. An optimal plant spacing was obtained within the model framework of current environmental settings together with the functional and structural modules. Simulation results indicate that the optimized plant spacing could increase rice yield, and that the optimization results remain stable.

Abstract from the paper’s OpenAlex record; open copy licensed cc-by. Read at the source ↗

The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.