Synchronized Predictive Control Of Networked Multi-Motors

Abstract:A motor is a common power transmission device in industrial production, and is widely applied in textile, iron and steel, automotive, aerospace industry and other fields. The synchronous control of multi-motors has strong functionality and good adaptability, which can meet complex requirements of control tasks. In a complex system, there exist many nodes, complex wiring and maintenance difficulties. Therefore, there are very strict requirements on the scalability and flexibility of the system structure.The nodes in traditional synchronized predictive control of multi-motors are connected directly through hardware circuits. It causes a complex operation and high error rate of changing control system nodes. Also, it is difficult to guarantee the reliability of the control system. Current research on synchronous control structures has not analysed the structure itself. So, it cannot meet the demands on scalability and flexibility of the control system. This paper proposes a networked synchronized control structure in view of the issue that the existing synchronized control structure cannot meet the requirements of flexibility and scalability, and applies a networked predictive control algorithm to control of multi-motors to overcome the delay and packet dropout in a network environment. The main work is as follows:First, the current synchronized control structure is analyzed in the anti-interference and synchronous precision aspects, and the relative coupling control structure is used to ensure synchronous precision. The advantages of networked control systems are analyzed, the time delay and packet dropout of the network communication are considered. The networked synchronized control structure is presented by combining networked control and relative coupling control.Second, the system identification toolbox in Matlab is used to establish the motor model, based on measured input-output data of motors. To compensate for network induced delays and data packet dropout, a networked predictive control algorithm is designed. The simulation of synchronized predictive control of networked multi-motors is carried out on the Matlab platform for the established motor system model. The feasibility of the proposed networked synchronized control structure and efficiency of the adopted networked predictive control algorithm are illustrated.Last, real-time synchronized control of multi-motors is achieved in a practical multi-motor control test platform based on the iNetConl04system. The feasibility of the proposed networked synchronized control structure and efficiency of the adopted networked predictive control algorithm are further validated.