当前位置:
X-MOL 学术
›
Automatica
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
Multivariable extremum seeking with distinct delays using a one-stage sequential predictor
Automatica ( IF 4.8 ) Pub Date : 2021-04-24 , DOI: 10.1016/j.automatica.2020.109462 Michael Malisoff , Miroslav Krstic
中文翻译:
使用一阶段顺序预测器以明显的延迟进行多变量极值搜索
更新日期:2021-04-26
Automatica ( IF 4.8 ) Pub Date : 2021-04-24 , DOI: 10.1016/j.automatica.2020.109462 Michael Malisoff , Miroslav Krstic
We provide new designs for multivariable extremum seeking for static maps with arbitrarily long time delays. We allow different delays in each input channel and output delays, and gradient- and Newton-based frameworks. Our sequential predictor based approach provides an alternative to earlier prediction methods, and leads to extremum seeking designs where a dynamical extension eliminates the need for distributed terms. Our source seeking example illustrates the good performance of our methods.
中文翻译:
使用一阶段顺序预测器以明显的延迟进行多变量极值搜索
我们提供了针对多变量极值的新设计,以寻求具有任意长时间延迟的静态地图。我们在每个输入通道和输出延迟以及基于梯度和牛顿的框架中允许不同的延迟。我们基于顺序预测器的方法为较早的预测方法提供了一种替代方法,并导致了极值搜索设计,其中动态扩展消除了对分布式项的需求。我们的寻源示例说明了我们方法的良好性能。