Elman neural network for predicting aero optical imaging deviation based on improved slime mould algorithm

Abstract:

This research suggests a methodology to optimize Elman neural network based on improved slime mould algorithm (ISMA) to anticipate the aero optical imaging deviation. The improved Tent chaotic sequence is added to the SMA to initialize the population to accelerate the algorithm's speed of convergence. Additionally, an improved random opposition-based learning was added to further enhance the algorithm's performance in addressing problems that the SMA has such as weak convergence ability in the late iteration and an easy tendency to fall into local optimization in the optimization process when solving the optimization problem. Finally, the algorithm model is compared to the Elman neural network and the SMA optimization Elman neural network model. The three models are assessed using four evaluation indicators, and the findings demonstrate that the ISMA optimization model can anticipate the aero optical imaging deviation in an accurate way.