Apply of Multi-scale Model Method Based on Wavelet Transform
Liu Jiwei, Shuangxi Hu, Chen Changxi, Shilei Du · 2021 China Automation Congress (CAC) · 2021
With the development of technology, demands of control accuracy and operation optimization get more and more higher. High precision model is the footstone for control and optimization. A multi-scale model method based on wavelet transform is proposed to establish appropriate model for complex product process. Wavelet transform is used to decompose parameters into different scales, then several models are set up. The form of the models can either mechanism based or data based. Then all models’ output would be added to obtain a more accurate predictive value. The simulation results of coal mill modeling showed, comparing with traditional signal-scale model method, performance of multi-scale model method is better.