预测PI和准预测PI控制算法在片烟复烤机上的应用  

Application of Predictive and Quasi-predictive PI Control Algorithms in Strips Redryer

作　　者：任正云[1] 韩佰恒 王小飞 金军辉 夏明东 

机构地区：[1]东华大学自动化系 [2]秦皇岛烟草机械有限责任公司 [3]上海派克软件公司,上海市徐汇区漕宝路82号200235

出　　处：《烟草科技》2009年第11期21-25,共5页Tobacco Science & Technology

基　　金：国家自然科学基金"组合积分系统先进控制理论研究与应用"的资助课题(资助号:60674088)

摘　　要：采用传统的PID控制算法对打叶复烤中的片烟复烤机进行控制,由于烤房温度波动大,难以保证冷房含水率与出口含水率稳定。为此,采用预测PI和准预测PI控制算法对其进行控制以保证控制品质和精度。烤房温度是大惯性、大滞后过程,采用预测PI控制算法可以避免调节速度慢、波动大的问题;冷房含水率与出口含水率均为多输入单输出过程,按照频率域准则将多输入单输出过程简化为单输入单输出的组合积分过程,并采用自适应抗干扰的准预测PI控制算法进行控制,使生产完全闭环自动控制,烤房温度、冷房含水率和出口含水率的控制精度均能达到设定值的±0.3%,满足了工艺要求。两家打叶复烤厂的应用结果表明,采用预测PI和准预测PI控制算法后,片烟含水率稳定地达到规定的工艺技术指标,提高了控制精度,实现了闭环全自动操作,无需操作人员的人工干预,减轻了劳动强度,返料次数由原来的每班至少2次减少为0,降低了能源消耗。It is hard to keep the temperature of the drying section, the moisture contents in tobacco in cooling section and ordering section stable by traditional PID control algorithm due to the drying in the strips redryer is a process of large thermal inertia and large time delay. The control system＇ s performance and accuracy can be improved by using predictive PI control algorithm and quasi-predictive PI control algorithm. The moisture con- tents in tobacco in cooling section and at output end are multiple-input and single-output processes, they are simplified into a single-input and single-output combined integrating process according to criteria of frequency domain, and controlled by self-adaptive anti-disturbance quasi-predictive PI control algorithm. The closed loop fully automatic control of production ensures the control accuracy of drying section temperature and moisture contents in tobacco in cooling section and at the output end within the range of ± 0.3% of set values. The results of application in two tobacco processing facilities demonstrate that by using the mentioned control algorithms, the moisture content in redried strips stably met the technical indexes standards without operator＇ s interference and energy consumption decreased also.

关 键 词：片烟复烤机 PID控制算法 含水率 预测PI控制算法 准预测PI控制算法 组合积分系统 

分 类 号：TS431[农业科学—烟草工业] TS443