基于数据驱动技术的非线性系统建模方法研究与应用

Research and Application of Nonlinear System Modeling Method Based on Data-Driven Technology

针对实际工业系统多存在非线性耦合、时变、滞后等特性,难以建立精确机理模型,提出了一种基于数据驱动的方法建立系统的预测模型。采集过程运行中的历史数据分别建立非线性系统的RBF、LS-SVM和KPLS 3种预测模型,仿真实验表明所建数据驱动模型具有较好的预测精度,能够被用于控制、预报和评价生产过程和设备的运行状态。

For practical industrial systems, there are many characteristics such as nonlinear coupling, time varying, and retardation. It is difficult to establish an accurate mechanism model, and a data -driven method is proposed to build a predictive model of the system. The historical data during the collection process were used to establish thine prediction models of RBF, LS - SVM and KPLS for nonlinear systems. Simulation experiments showed that the data - driven model has better prediction accuracy and can be used to control, forecast and evaluate the process of production, and equipment operating status.