Abstract
The methods for predictor design based on grey model have been widely studied in various industrial manufacturing processes to enhance their control performances. However, the prediction capability of grey model is limited by some irrational problems. In this paper, an improved grey model is proposed to acquire high-control system performance. First, the cubic Hermite spline function is integrated into the grey model to enhance its prediction capability of grey model. Then a residual compensation approach based on artificial neural network is proposed to further improve the prediction performance. Finally, the authors validated the effectiveness of the proposed predictor by using online prediction simulation cases in industrial control systems.