Research On Ball Mill Control Method Based On Fractional Order Brain Emotional Learning Model

Ball mill is the key device to break and grind bulk stuff,it is widely used in thermal power plant,mine and metallurgy thanks to its characters of vast application scope,safe and steady operation and not picky about material.Keeping ball mill running safely and effectively has been key to get more economic benefits for manufacturing unit.However,ball mill is such an object with multiple variables,strong coupling and timely variant,so traditional control methods cannot get better control result on it.Therefore,it is urgent and valuable to implement intelligent control on ball mill pulverizing system.Brain Emotional Learning(BEL)model is a kind of new machine learning model,it imitates the mechanism of brain emotional and learning and has better ability to study and adjustment as well as stronger robustness.This model gets much attention once it is created,many methods based on BEL have been widely used in many different fields in recent years.This paper focuses on ball mill problems of safe operation and low efficiency,and uses BEL model to control ball mill intelligent.The main tasks are as follows:(1)Aiming to the characteristics of strong coupling and time variation of ball mill system,an improved control method of brain emotional learning model(BEL)was proposed.Fractional calculus is adopted to depict emotional input function and emotional reward function of BEL model,which makes the selection of input signals of BEL model and improves control precision of BEL controller.An intelligent controller based on fractional order BEL is designed by using the method of multivariable inverse decoupling.The simulation results show that this method has better control performance,is resistant and model-insensitive as well as better robustness.(2)To improve the stability and control precision of ball mill control system,this paper proposes a method to optimize the parameters of BEL model,which adopts teaching-learning-based optimization(TLBO)method to optimize every parameter to be more rational and improves system precision.The simulation result suggests that method proposed in this paper could get more precise parameters and optimal solution faster.