The Key Problems Of Synchronous Acquisition Of Industrial Information

The full launching of China Manufacturing 2025,and the promotion and use of the "Manufacturing + Internet" public platform have enabled the application of Wireless Sensor Network(WSN)technology is extend to the industry to form a new research field.Industrial Wireless Sensor Network(IWSN)is mainly used to monitoring industrial process,collaborating industry management and reducing some unnecessary waste.Meanwhile,it collects key data to optimize the production process,so as to improving production efficiency and achieving greater benefits for enterprises.The acquisition of accurate data plays a key role,and this thesis considers the key problem of synchronous acquisition of industrial information: clock synchronization problem.Considering the requirement of the IWSN application for high-performance clock synchronization algorithms,and the existing WSN clock synchronization algorithm can not take into account the convergence speed,synchronization error and network scalability.Based on the theory of network control,a state decoupling observation model is designed.Then according to the principle of separating mechanism,the Luenberger estimator and the controller based on the output feedback model predictive control(MPC)are designed separately,a distributed clock synchronization control algorithm is established.It improves the anti-interference ability of the clock synchronization algorithm,speeds up the synchronization convergence speed,and quantifies the synchronization error.The main contribution of the thesis are as follows: 1)An absolute clock state space model based on standard time is established from the viewpoint of network control,to get rid of the state coupling and control coupling between nodes.A clock synchronization method based on output feedback model predictive control is established,and the clock synchronization problem is transformed into a standard clock fixed-point tracking problem,then realize distributed control which does not rely on the network transmission control information,and improve the robustness of clock synchronization algorithm to the network structure and the adaptability of large scale network;2)In order to reduce the on-line computational complexity of the system,boundary treatment of the Gauss noise from a probability point of view and do the exception handling for cases beyond the boundary,to ensure the optimal control problem is feasible.For the positive definite invariant set,a substitution method for calculating the approximate positive fixed invariant set by the state decoupled Luenberger observer as a theoretical calculation tool is proposed,and the on-line calculation of clock synchronization control method is realized;3)With the nominal system as the center of the output feedback MPC control strategy,the thesis uses the feedback control strategy and the control set theory,quantitative analysis the error bound between the perturbation system state,the estimated system state and the nominal system state by establishing error positive definite invariant set,and proves an exponential convergence performance of clock synchronization control algorithm.