研究基于线性量化的网络控制系统的状态观测器的设计问题.在设计观测器时,考虑观测信号的线性量化以及量化后的数据经过非完美的频道传输时有数据损失发生.针对量化误差和数据损失2种不稳定因素,基于Bernoulli模型求得误差的最小值设计观测器,所设计的观测器能够保证闭环系统有界稳定,且增益阵最优.应用数值仿真验证了该方法的有效性和可行性,得出了量化密度、数据丢包概率与所设计的观测器增益矩阵之间的关系,在理论上增大了网络控制系统的稳定性和可实现性.
The problem of the design of state observer was considered for network control system based on the linear quantization.Two cases were considered,which were the quantized measurement and the data-loss occurring when the quantized signal was transmitted in the unreliable channel.The observer with the minimal error variance was designed based on the Bernoulli model aimed at the quantization error and the data-loss.The observer can not only guarantee the bounded-input and bounded-output stability for the closed-loop system,but also the optical gain matrix.Numerical results illustrated that the approach was effective and feasible.The relationship between the quantized density,the data-loss probability and the gain matrix was shown,and the network control system became more stable and realizable.