Feng Baolin, Feng Xue,ChuangYao, and Qiao Xing.  Application of Two Fractional Differential Equations Based on Competitive Neural Network Algorithm in Control System.  Dynamic Systems and Applications 29 (2020) No. 3, 569-578

https://doi.org/10.46719/dsa202029313

ABSTRACT.
Fractional calculus is the theory of studying the differential and integral of any order. Fractional calculus is an arbitrary generalization of integer calculus in order. It has strong advantages in many fields such as physics, neural network, medicine, control engineering and has a wide application background. Highly concerned. A large number of studies have confirmed that there is inevitably a time lag in the actual network, and the time lag has an important influence on the dynamics of the fractional system. This paper proposes an improved classification algorithm based on the existing classification algorithm for the problem that the competitive neural network can not accurately classify the training samples without obvious classification features. In the competitive neural network learning process, the algorithm introduces the feature vector of training samples and applies it to the control system. It is found that the algorithm has a good effect in the classification process, which not only reduces the training error, but also has Good classification accuracy, showing good learning efficiency, and comparing the effectiveness and superiority of the algorithm.

Keywords. Two fractional differential equations; competitive neural network algorithm effect