The fractional neural grey system model and its application

Abstract

Recently, grey system, neural network, and fractional order calculus theory have become popular research areas, and an increasing number of scholars have joined these studies, conducted illuminating research, and produced a number of significant results. Numerous research studies have demonstrated that these three strategies are crucial to solving a wide range of practical problems. In this paper, we present a fractional order neural grey system model with a three-layer structure in which the input of the network is a fractional order cumulative sequence, and the output is a predicted value in order to maximize the benefits of each of the three elements. The purpose of this research is to present a strategy for reducing the number of conditions in order to improve the stability of parameter estimation by using QR decomposition. The order of the models is determined by an intelligent optimization algorithm. Finally, real-world examples are used to validate the model’s validity, and experimental results indicate that the newly presented model is more accurate than previous models.