一种基于即时学习的多模型在线建模方法

An Online Multiple-model Modeling Method Based on Lazy Learning

针对复杂非线性系统的建模问题,基于空间划分树(SP-Tree)和即时学习(lazy learning)的思想,设计了一种多模型在线建模方法。该方法基于分解-合成策略,根据系统输入输出数据,采用即时学习算法建立当前时刻的最佳局部模型,随着系统工作点的移动,滚动建立系统的多个模型,实现对非线性系统的准确建模。在建立邻域的过程中,采用一种基于SP-Tree数据结构的数据库进行分层递阶搜索,有效地提高了在线建模的实时性。最后,通过对一个仿真案例的研究验证了该算法的有效性。

An online multiple-model modeling method based on spatial partition tree and lazy learning is suggested for complex nonlinear system. The new method establishes the optimum local model of the system based on lazy learning algorithm, which is on the basis of divide-and-conquer principle and input-output data. As working points changing, multiple local models were built to realize the exact modeling for the global system. To select local neigh- borhoods of the query points, a hierarchical searching strategy result, the real-time performance of the modeling is improved. proposed method. based on spatial partition tree is present, and as a Simulation results showed the effectiveness of the