Research Of Redundancy Technology In Network Control System Based On Markov Process

As a technical means to enhance system reliability, redundancy technology ensures continuous operation of the system by improving the structure of the network control system and using idle backup module to replace failed component in case of controller failure and communication network failure. Because the promotion of performance and reliability in a single component is limited, a redundant design is taken by using group components to replace the single one, which can make sure that the entire system is running when a single component fails and normal operation of system will keep going on while the fault can be repaired in off-line condition.In order to improve the system reliability, a research on the architecture of master-slave network control system is carried out and its hybrid dual-redundancy technology is developed. Based on network redundancy and device redundancy, analyzes on network structure, communication protocol and systems running control are put forward. Using online self-diagnosis to find fault reason, automatic switching strategy is implemented on choosing more stable equipment and communication link, which realizes self-recovery of failure in system. Then, resources synchronization is carried out to make sure that the resource in active equipment is consistent to one in backup equipment and the disturbance of production process will reduce when role-switching happens, which promises the control system can keep going on after switching.According to improve the reliability of control system, a formula for reliability analysis is derived combined with the Markov process theory. Based on backup component in system and fault switching process, a hybrid-redundancy equipment failure model is established by creating a state transition matrix to solve differential equations. The impact of device parameters on system reliability is analyzed, which provides a theoretical basis for construction of project and the transformation of equipment. On the other hand, a dual-redundancy working status switching model is established based on Markov process. After compared with non-redundancy system model, it can be proved that system reliability degree in hybrid-redundancy system is improved by73.59%, which enhances the reliability significantly.An example of control system is put up to verify for this hybrid redundancy method. With the plug-in fault testing, hot switch time in master-slave system and time-consuming in each stage are recorded to study the fault recovery process of system. The results show that the system can be recovered into normal operation by hybrid redundancy function through fault detection and role switching when some modules fail. In addition, hot standby switching method can effectively reduce system failure recovery time while resource synchronization technology lowers the effects of switching process on user programs executed by controller, which improves the performance of system.