状态量化的随机系统容错控制策略研究  

Research on Fault-Tolerant Control Scheme of Stochastic Systems with States Quantization

作　　者：吴健 杨亚东 钱立冬 WU Jian;YANG Yadong;QIAN Lidong(School of Computer and Information,Anqing Normal University,Anqing 246133,China;School of Information Engineering,Yangzhou University,Yangzhou 225127,China;State Grid Anhui Province Electric Power Co.,LTD,Feidong County Power Supply Company,Hefei 231600,China)

机构地区：[1]安庆师范大学计算机与信息学院,安徽安庆246133 [2]扬州大学信息工程学院,江苏扬州225127 [3]国网安徽省电力有限公司肥东县供电公司,安徽合肥231600

出　　处：《安庆师范大学学报（自然科学版）》2024年第1期10-16,共7页Journal of Anqing Normal University(Natural Science Edition)

基　　金：国家自然科学基金(61603003)。

摘　　要：本文研究了具有状态量化和时变输入时滞的随机系统的自适应神经容错控制问题。通过构造一种补偿系统来消除延迟造成的不利影响,并利用径向基函数神经网络在线处理系统的未知项问题。为了补偿状态量化带来的不连续问题,本文构造了一个引理来建立信号量化前后的关系,进而提出一种自适应神经容错控制算法。基于构造的控制器和量化器,随机系统内的信号均概率性有界;最后,利用仿真实例验证了所提控制算法的可行性。This article investigates the issue of adaptive neural fault-tolerant control for the nonlinear stochastic systems with state quantization and time-varying input delay.By designing a compensation system to eliminate the adverse effect caused by input delay,and the radial basis function neural networks are applied to handle some unknown terms.In order to resolve the discontinuity problem brought by states quantization,this article constructs one lemma to establish the relationship between unquantized signals and quantized signals,and then proposes an adaptive neural networks fault-tolerant control algorithm.Under constructed control law and quantizer,all signals maintain bounded in probability.At last,the practicality of the developed control scheme is validated by a simulation experiment.

关 键 词：随机系统 自适应神经控制 状态量化 输入时滞 传感器故障 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]