Discrete Fractional Order Modeling of Plant Capture Carbon Dioxide Dynamical Analysis with Neural Networking

Abstract

A nonlinear mathematical model is suggested to examine the role of varying abilities of plants in absorbing atmospheric Carbon Dioxide (CO₂) and its impact on the ecosystem. Various plant species show distinct abilities to absorb and capture CO₂, which can affect the overall reduction in atmospheric CO₂ quantities. The study objectives to consider how differences in plant capacities for CO₂ absorption influence atmospheric CO₂ levels. The focus is on identification the impact of plant growth rates and reaping rates on CO₂ concentration. The model is formulated using a difference operator to facilitate numerical exploration. It employs the Discrete Numerical Iterative Method (DNIM) joint with neural networks, particularly the Levenberg-Marquardt (LM) algorithm, known as DNIM-LM. The model's performance, training status, error distribution, regression, and suitability are calculated using artificial intelligence procedures. The dataset is split into 70% for training, 15% for authentication, and 15% for testing. The analysis shows that plants with higher CO₂ absorption capacities attain faster reductions in atmospheric CO₂ quantities as their growth rate increases. On the other hand, an increase in the harvesting rate coefficient is related to an increase in CO₂ concentration. The study determines that variations in plant absorption capacities expressively influence the dynamic contrast of atmospheric CO₂ reduction in the environment. This finding highlights the importance of plant growth rates and harvesting performs in managing CO₂ levels, present insights into ecosystem management and carbon sequestration policies.