Analysis And Optimization Of Networked Control Systems Based On Hyper-Sampling Mode

With the rapid development of computer network technology, researchers combine control theory with computer network technology to contribute to the creation and development of the networked control systems. As a major research field, Stability and dynamic performance analysis has been a hot issue of research on networked control systems, this thesis has a major research in the optimization problem of the dynamic performance. The sampling period and the sampling mode will affect the performance of networked control system which is a kind of discrete digital control system. Therefore, based on the sampling period as the main research object, this thesis employs the hyper-sampling mode and adopts the symbolic operation as the main computing tools to realize the optimization of system dynamic performance. The main works are given as follows:(1) This thesis discusses the single-sampling mode which is a special case of the hyper-sampling mode, namely it samplings only once within a sampling period. By calculating the optimal performance index to obtain the optimal sampling period, this thesis can achieve the goal of optimizing the system dynamic performance. In the process of calculating, it shows that that there is no positive correlation relationship between the performance index and the sampling period within a certain range of sampling time. On account of a phenomenon appearing in the simulation, the discuss is continued and the experimental results show that increasing the sampling period to a certain degree can improve the dynamic performance of systems and saving systems resources at the same time within a certain range of sampling time.(2) This thesis mainly studies the optimal effect on system dynamic performance by applying to the hyper-sampling mode. By the same performance index function, the optimal sampling period and the corresponding optimal sub-sampling periods can be obtained, the experimental results show that there is still no positive correlation relationship between the performance index and the sampling period. Then the simulation within the same average sampling frequency(ASF) shows that there may be exists the hyper-sampling sequence that makes the dynamic performance better than the one under the hyper-sampling mode, and along with the increasing numbers of sub-sampling, system dynamic performance is also improved obviously.(3) Based on the close connection of the time-delay systems and networked control systems, this thesis makes a discussion about the relationship between the time delay and the sampling period within the networked control system model. Through the analytical calculation of some examples, this thesis can be found that critical stable value of the time delay may be greater than critical stable value of the sampling period, and these two ones may be acted upon each other.This thesis mainly studies the optimization of sampling period, and employ a new method for optimization of the sampling period. The simulation results show that by means of optimizing the sampling period, the dynamic performance of the system is improved.