Model-free adaptive control of cement mill load based on VRFT method

Starting from the direction of improving the operating efficiency of the cement pre-grinding system and reducing the energy consumption of cement production, this paper proposes a model-free adaptive control method for mill load based on the analysis of the cement pre-grinding system. First of all, this paper determines the key variables that affect the pre-grinding efficiency through process analysis and uses wavelet filtering to filter the data. After that, a dynamic relationship model between the speed of the circulating fan and the current of the bucket elevator was established by using the neural network method. We designed the controller using the MFAC method based on full-format dynamic linearization, followed by the VRFT method to tune the controller parameters. Finally, the simulation proves that the method has a better tracking effect.