This paper is focused on the distributed estimation issue in the form of set-membership (SM) for a class of discrete time-varying systems suffering mix-time-delays and state-saturations. The phenomena of time-delays and state-saturations are introduced to better describe insightful engineering. During local measurements transmission between sensors over a resource-limited sensor network, to prevent data collisions and resource-consumption, a newly dynamic event-triggering strategy (DETS) is designed to dispatch the local measurements transmission for each sensor to its neighbors. Compared with the most existing static ETSs, this DETs can mitigate the total number of triggering times and enlarge interval time between consecutive triggering instants. Then, some novel adequate criteria for designing the desired event-based SM estimators are derived such that the plant’s true state always resides in each sensor’s ellipsoidal region regardless of the simultaneous presence of bounded noises, mixed time delays and state-saturations. Subsequently, a recursive optimization algorithm is formulated such that the minimal ellipsoids, the estimators gains and event-triggering weighted matrices are acquired simultaneously. A verification simulation is presented to illustrate the advantages of the design approach of the developed state estimator.

本文重点研究一类具有混合时间延迟和状态饱和度的离散时变系统的集合成员（SM）形式的分布式估计问题。引入时间延迟和状态饱和现象以更好地描述有见地的工程。在传感器之间通过资源有限的传感器网络进行本地测量传输期间，为了防止数据冲突和资源消耗，设计了一种新的动态事件触发策略 (DETS)，将每个传感器的本地测量传输分派给其邻居。与大多数现有的静态 ETS 相比，该 DET 可以减少触发次数的总数并扩大连续触发时刻之间的间隔时间。然后，推导出了一些用于设计所需的基于事件的 SM 估计器的新的适当标准，使得植物的真实状态始终位于每个传感器的椭球区域，而不管是否同时存在有界噪声、混合时间延迟和状态饱和。随后，制定了递归优化算法，以便同时获取最小椭球、估计量增益和事件触发加权矩阵。提供了一个验证模拟来说明所开发的设计方法的优点 同时获取估计器增益和事件触发加权矩阵。提供了一个验证模拟来说明所开发的设计方法的优点 同时获取估计器增益和事件触发加权矩阵。提供了一个验证模拟来说明所开发的设计方法的优点状态估计器。