执行器饱和多速率采样网络化控制系统研究

Many problems are induced for the introduction of the communication networks into control closed-loop. In order to achieve the high control performance, the networks bandwidth should be well used when the information is transported among the control nodes. The available networks bandwidth is real-time changed based on the environment of networks load. Iit is difficult to obtain good control effect in the environment of changing networks bandwidth by using traditional fixed period sampling approach, which is suit for the networks environment with fixed bandwidth. Moreover, saturation is very ubiquitous in all practical control systems due to the inherent physical limitations of the devices. Actuator saturation can lead to performance degradation and even instability of the closed-loop systems. Therefore, this project studies the multi-rate sampling networked control problems subject to actuator saturation. A new networked control model with actuator saturation is given based on the delta operator approach. Some problems on state estimation and controller design are also deeply investigated in the project. Firstly, a delta operator system model with actuator saturation is proposed to describe networked control systems. In the new model, the sampling period values are used explicitly, by which the available networks bandwidth is real-time reflected. Secondly, both filtering and control of networked systems are studied, some new state estimation and control strategies which are dependent on multi-rate sampling periods and actuator saturation are proposed. Finally, an experimental platform on internet of things will be consummated to illustrate the effectiveness and the potential of the developed techniques. The research results of this project are expected to enrich the contents of multi-rate sampling theory, develop networked control systems with actuator saturation, and provide new ideas and methods for control practice in internet of things environment.

网络嵌入到控制闭环给系统的分析与设计提出了诸多挑战。网络负载的多变导致网络可用带宽实时变化。适用于固定网络带宽的传统定周期采样控制方法很难在多变带宽环境下取得好的控制效果。由于设备内在物理局限性，饱和现象在控制系统中非常普遍。执行器饱和不但会使系统的动态性能降低，甚至会导致闭环系统不稳定。因此，本项目提出带执行器饱和约束条件的多速率采样网络化控制策略，基于Delta算子系统建立网络化控制新模型，进一步对滤波和控制等问题进行研究。首先，建立带执行器饱和的Delta算子系统模型，新模型融入的显式采样周期参数可实时反映和利用网络带宽。其次，研究网络化系统滤波与控制问题，提出依赖于执行器饱和与多速率采样周期的状态估计新方法和控制新策略。最后，将研究结果应用于物联网实验平台，进行实验验证。研究成果将丰富多速率采样理论，发展执行器饱和网络化控制系统研究内容，同时对物联网的应用提供新的研究思路和方法。

网络嵌入到控制闭环给系统的分析与设计提出了诸多挑战。由于设备内在物理局限性，饱和现象在控制系统中非常普遍。执行器饱和不但会使系统的动态性能降低，甚至会导致闭环系统不稳定。因此，本项目提出带执行器饱和约束条件的多速率采样网络化控制策略，基于Delta算子系统建立网络化控制新模型，进一步对滤波和控制等问题进行研究。首先，建立带执行器饱和的Delta算子系统模型，新模型融入的显式采样周期参数可实时反映和利用网络带宽。其次，研究网络化系统滤波与控制问题，提出依赖于执行器饱和与多速率采样周期的状态估计新方法和控制新策略。最后，将研究结果应用于物联网实验平台，进行实验验证。研究成果将丰富多速率采样理论，发展执行器饱和网络化控制系统研究内容，同时对物联网的应用提供新的研究思路和方法。在本项目的支持下，申请人出版了英文专著1部，发表（含录用）SCI收录论文56篇，EI收录论文6篇，其中在IEEE Trans系列汇刊上发表（含录用）论文15篇。