摘要
提出了基于RBF-ARX模型的压铸机多变量压射过程非线性预测控制算法。在参考历史数据和机理模型建立RBF-ARX全局非线性动态模型的基础上,以快速收敛的结构化非线性参数优化策略离线辨识估计和优化模型参数,确定了预测控制系统的结构和多步向前预测控制输出。与传统PID控制算法的速度设定值跟踪效果对比表明基于RBF-ARX模型的非线性预测控制算法动态特性好,控制精度高。
A nonlinear predictive control algorithm was proposed for multivariable die-casting processes of die-casting machine based on RBF- ARX model. By constructed global nonlinear dynamic RBF - ARX model referenced on historical data and mechanism model, the structured nonlinear parameter optimization method is developed to estimate and optimize parameters of RBF- ARX model off-line,finally determined the predictive system structure and steps forward output. Speed set point tracking comparison with traditional PID control shows nonlinear predictive control algorithm based on RBF - ARX model has good dynamic characteristics and high control precision.
出处
《电气传动》
北大核心
2011年第8期33-36,60,共5页
Electric Drive
基金
辽宁省高等学校优秀人才支持计划(2008RC25)
辽宁工程技术大学研究生科研立项资助(Y201000402)
关键词
压铸机
非线性预测控制
有源自回归模型
径向基函数神经网络
die-casting machine
nonlinear predictive control
fuzzy auto-regressive with extra inputs model
radial basis function neural network