A control process based on data often contains large amounts of sample data. The accuracy of system is always influenced by the randomness of the initial data when the RBF neural network is used to predict model. However, the grey accumulated generating operation (AGO) can reduce the effect of randomness of the initial data, which is able to make data more regular. Based on the two points above, a new kind of method is proposed, which is called grey RBF neural network. This method not only can reduce the randomness, speed up the network convergence, but also improve the modeling accuracy. The grey RBF neural network can be proved to be feasible and effective by applying the grey RBF neural network to the synthetic ammonia decarburization process, and comparing the simulation results with the results which was only using RBF network.
Adaptive Robust of RBF Neural Network Control Based on Model Local Approximation Method for Upper Limb Rehabilitation Robotic Arm
