摘要
针对甲醇精馏过程动态反应慢、机理复杂、参数间耦合严重等问题,设计了一种基于支持向量模型的预测控制器。通过建立支持向量机模型的结构,利用历史数据对模型进行训练,模型训练完成后作为系统的控制器。通过Matlab软件对该模型进行仿真,试验结果表明:该方法控制精度高,具有良好的抗干扰性能,同时该模型具有自学习能力,可以更快地提供甲醇产品估算值,改善控制效果,对解决甲醇精馏控制中遇到的系统复杂、非线性高、机理建模误差大的问题,有一定的改善作用。
Aiming at the problems of slow dynamic response, complex mechanism, serious coupling among parameters for methanol distillation process, one kind of predictive controller based on support vector model has been designed. Through establishing the structure of support vector model, and the model is trained with historical data, and is used as system controller after training. The model is simulated with Matlab software. The test results show the method has high control accuracy, good anti-interference performance. Meanwhile, the model has self-learning ability, and can provide methanol product estimate value more quickly to improve control effect. It has certain improving effects to solve the problems of complex system, high nonlinear and big mechanism modeling error for methanol distillation.
出处
《石油化工自动化》
CAS
2015年第1期18-21,共4页
Automation in Petro-chemical Industry
关键词
甲醇精馏
支持向量机
预测控制
建模
methanol distillation
support vector model
predictive control
modeling
