基于分数阶BEL模型的球磨机控制方法研究

发布时间：2018-05-04 11:23

本文选题：球磨机 + 大脑情感学习模型　；参考：《江西理工大学》2017年硕士论文

【摘要】：球磨机是一种破碎并研磨大块物料的关键制粉设备,其具有适用范围广、运行安全稳定、对物料不挑剔等特点,广泛应用于火电厂、矿山、冶金等领域。保证球磨机能够安全高效的运行,成为了生产单位取得较好经济效益的关键。然而球磨机是一个多变量、强耦合、时变性的控制对象,传统自动化控制方法很难实现对其较好的控制效果。故而,实现球磨机制粉系统的智能化控制具有重要的工程价值和现实意义。大脑情感学习模型是一种新型的机器学习模型,其模拟大脑的情感和学习的机制,具有较强的学习能力和调节能力,鲁棒性较好。一经提出就引起较大关注,近年来基于大脑情感模型的控制方法已经广泛应用在不同的领域。本文围绕保证球磨机安全运行、提高其工作效率的问题,利用大脑情感学习模型对球磨机进行智能控制,主要内容如下:(1)基于球磨机系统的强耦合、时变性的特点,提出了一种改进的大脑情感学习模型(brain emotional learning,BEL)的控制方法。采用分数阶微积分对BEL模型的感官输入函数和情感暗示函数进行描述,使得BEL模型输入信号选择更为合理,提高了BEL控制器的控制精度。利用多变量逆向解耦的方法,设计了基于分数阶BEL的智能控制器。仿真结果表明:该方法具有较好的控制性能、良好的抗干扰性能及模型不敏感,具有良好的鲁棒性。(2)为了提高球磨机控制系统的稳定性和控制精度,提出了一种优化大脑情感学习模型参数的方法,采用教与学优化算法对系统的各个参数进行优化,使得系统的各个参数更合理,提高了系统的精度。仿真结果表明:该方法对参数选取的精确度较高,能更快寻找到最优解。
[Abstract]:Ball mill is a key powder making equipment for crushing and grinding large pieces of material. It has the characteristics of wide application, safe and stable operation, not picky on material and so on. It is widely used in thermal power plant, mine, metallurgy and other fields. To ensure the safe and efficient operation of the ball mill, it has become the key to achieve better economic benefits in the production units. The machine is a multi variable, strong coupling and time-varying control object. The traditional automatic control method is difficult to achieve good control effect. Therefore, it is of great engineering value and practical significance to realize intelligent control of the ball mill pulverizing system. The brain emotion learning model is a new model of machine learning, which simulates the brain. The mechanism of emotion and learning has strong ability to learn and adjust, and has good robustness. Once put forward, it has aroused great concern. In recent years, the control methods based on brain emotion model have been widely used in different fields. This paper focuses on the problem of guaranteeing the safe operation of the ball mill and raising the efficiency of its work, and using the emotional learning of the brain. The main contents of the model are as follows: (1) based on the strong coupling and time-varying characteristics of the ball mill system, an improved brain emotional learning (BEL) control method is proposed. The fractional order calculus is used to describe the sensory input function and the emotional implication function of the BEL model. The input signal selection of the BEL model is more reasonable and the control precision of the BEL controller is improved. By using the multivariable inverse decoupling method, the intelligent controller based on the fractional order BEL is designed. The simulation results show that the method has good control performance, good anti-interference performance and model insensitivity, and has good robustness. (2) (2) In order to improve the stability and control precision of the ball mill control system, a method of optimizing the parameters of the brain emotion learning model is proposed. The parameters of the system are optimized by teaching and learning optimization algorithm. The parameters of the system are more reasonable and the precision of the system is improved. The simulation results show that the accuracy of the method is accurate for the selection of parameters. Higher, can find the optimal solution faster.

【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273;TM621

【参考文献】