基于PSO＋CMAC的铝热连轧精轧机组张力系统控制研究

针对河南某公司“1＋4”铝热连轧机升级改造过程中产生张力波动问题，分析了影响张力波动因素，结合前后张力对前滑值的影响，完善了连轧张力公式，完全模拟铝热连轧从F1机架到F4机架穿带建张过程，并构建出四机架张力动态框图。由于连轧张力系统的非线性、耦合性，使得常规PID控制器难以满足要求，所以提出利用粒子群算法（PSO）对PID控制器参数进行优化，然后与小脑神经网络（CMAC）复合进行控制，解决了铝热连轧穿带过程中控制器参数难以在线实时改变的问题。仿真结果表明，PSO＋CMAC复合控制的使用大幅度地减小系统超调量、过渡时间，降低了机架间张力耦合程度。

Tension Control Study in the Finishing Mill of Aluminum Hot Tandem Rolling Based on PSO + CMAC

Aimed at the tension variation during transforming process of 1 ＋ 4 aluminum hot tandem rolling u- nit in one factory of Henan, the factors affecting tension variation was analyzed. Considering influence of tension on for- ward slip value, the tension formula of continuing rolling process was improved. The tension forming process of continu- ing hot rolling from stand F1 to F4 was simulated, and the tension diagrams of the four stands were constructed. Due to non-linear relation of the aluminum hot tandem rolling tension system, and coupling normal PID controlling could not meet the requirement of the system, so using Particle Swarm Optimization （PSO） to optimize PID controller parame- ters, the mixed control of Cerebellar Model Articulation Controller （ CMAC ） coupled with PID was proposed to solve the problem of online parameters control. By simulation, it could be found that the using of PSO ＋ CMAC mixed control significantly reduced surplus adjustment, transition time, and tension coupling degree among stands was also reduced.