A Hybrid Surrogate Approach for Reducing Photosynthesis Runtime in a Mechanistic Plant Model

Keywords:
Mechanistic Plant Models, Surrogate Modelling, Plant Digital Twins, Hybrid Modelling

Abstract:
This paper examines the use of surrogate models to accelerate a mechanistic plant simulator in the context of plant digital twins. The research aims to reduce the computational cost of the simulator while preserving the biologically meaningful growth behaviour of the underlying model. To this end, the native C implementation was first profiled, which identified the iterative Farquhar– Jarvis photosynthesis routine as the dominant runtime bottleneck. Based on this finding, a simulation-generated dataset was constructed, a compact multilayer perceptron surrogate was trained for the high-irradiance regime, and the resulting model was integrated directly into the existing C simulation pipeline. The results showed that the surrogate achieved near-perfect agreement with the original net assimilation target in the selected regime, while substantially reducing the computational dominance of the original photosynthesis routine. At the same time, the surrogate-enabled simulator preserved relative growth rate and final leaf biomass with only minor deviations across the evaluated photoperiod scenarios. The main contribution of the paper lies in demonstrating that a narrow, regime-aware surrogate can significantly improve runtime efficiency without disrupting the mechanistic structure and system-level behaviour of the original plant model.

CITATION:

IEEE format

Copy to clipboard

A. Joksimović, D. Kostić, T. Naumović, P. Lukovac, Z. Bogdanović, “A Hybrid Surrogate Approach for Reducing Photosynthesis Runtime in a Mechanistic Plant Model,” in Sinteza 2026 - International Scientific Conference on Information Technology, Computer Science, and Data Science, Belgrade, Singidunum University, Serbia, 2026, pp. 260-266. doi:10.15308/Sinteza-2026-260-266

APA format

Copy to clipboard

Joksimović, A., Kostić, D., Naumović, T., Lukovac, P., Bogdanović, Z. (2026). A Hybrid Surrogate Approach for Reducing Photosynthesis Runtime in a Mechanistic Plant Model. Paper presented at Sinteza 2026 - International Scientific Conference on Information Technology, Computer Science, and Data Science. doi:10.15308/Sinteza-2026-260-266

BibTeX format
Download

RefWorks Tagged format
Download