基于HTCPN和隐马尔科夫的网络控制系统模型分析

针对具有随机时延的网络控制系统(NCSs),研究其建模与系统分析问题.运用层次时间颜色Petri网(HTCPN)对以PLC为控制器的NCSs进行结构保留形式建模,并以新颖的方式将网络丢包和乱序现象与系统模型结合.重点分析系统中传感器-控制器、控制器-执行器随机时延分布情况,建立时延与网络中数据包运行情况间的关系.在此基础上,利用Baum-Welch算法建立两种时延的隐马尔科夫模型(HMM),进一步验证了HTCPN模型与HMM的等效性.

Model analysis of network control systems based on HTCPN and hidden Markov

The modeling and system analysis of networked control systems(NCSs) with stochastic delay are studied. Firstly, the structure conserving model of the NCSs with the PLC controller is built by using hierarchical timed colour Petri net(HTCPN), which combines network packet loss and disorder with the system model in a novel way. Secondly, the random delay distributions from the sensor to controller and the controller to actuator are analyzed, and the relationship between the time delay and the operation of data packets in the network is established. On this basis, the Baum-Welch algorithm is used to establish two time delay hidden Markov models(HMM) to further verify the equivalence between HTCPN model and HMM.