Simulation of fractional order mathematical model of robots for detection of coronavirus using Levenberg–Marquardt backpropagation neural network

This research motive is to present more realistic results through the fractional order derivatives of the Robots mathematical system (FORMS), which is used to detect the coronavirus-positive cases. This nonlinear FORMS is useful to prevent individuals from the coronavirus and its spread. The classification of mathematical system is categorized into two dynamics, infected and Robots. The designed FORMS has never been solved before through the stochastic Levenberg–Marquardt backpropagation (LBMBP) neural networks (NNs), i.e., LBMBP-NNs. The solution of three cases of the FORMS is presented along with the statistics of 76% training, 12% authorization and 12% testing. To observe the exactness of LBMBP-NNs, a reference dataset is constructed using the Adams scheme. For the validation and capability of LBMBP-NNs, the illustrations are drawn based on the state transition values, regression measures, correlation performances and error histograms.