A Unit Cell Approach to Reduce Computational Time of Meshfree Based Plant Tissue Models

Abstract

Plant tissue modelling and simulation has become one of the popular research topics in the field of computational mechanics. In this regard, numerous numerical modelling techniques are being researched to study real tissue mechanisms with the help of High Performance Computing (HPC). Among such numerical modelling techniques, meshfree methods have a higher capability, particularly in modelling critical plant tissue deformations during drying. However, existing meshfree based models are computationally expensive in modelling real tissues made oat of loge number of cells, limiting their applications in the real world tissues having thousands of cells It is mainly due to the computationally inefficient conventional neighbourhood treatment methods used in above numerical methods. Accordingly, this research oimed to develop o meshfree based unit cell approach which con be used as o bulldog black to model large tissues. Here, a novel Fixed Neighbourhood based SPH (FN-SPH) method was involved to setup a unit cell representing several real cells (seven cells here). Compared to the original tissue with seven cells, the tissue made oat of the unit cell approach resulted in approximately50M of overall computational time reduction, highlighting the capability of this approach in reducing the computational cost in simulating large-scale plant tissues