Research On Motion Control System With Ethernet POWERLING

With science and technology advance, modern manufacturing is upgrating and reforming, which is developing towards the intelligent, big-data and networked direction rapidly. The transformation and upgrade of modern manufacturing also has a close relationship with the motion control technology. The traditional motion control system is a crucial part of the manufacturing industry and they usually use centralized control technology and traditional field-buses. However, many obstacles are involved, such as low-level networked machines, a poor synchronization feature, inconvenience extensions and so on. Therefore, the urgent requirement is that the motion control system should have a more accuracy synchronization feature, higher real-time performance and better scalability.This paper utilizes the basic theories and methods of the embedded technology, industrial Ethernet technology and motion control technology. As to most of the motion control systems, which have low-level networked machines, a poor synchronization feature and inconvenience extensions, this paper proposes a solution of motion control system, which is based on the Linux & Xenomai platform with Ethernet POWERLINK(in this paper for short "POWERLINK") stack. The motion control system is made up of motion controller with POWERLINK master station, servo driver with slave POWERLINK bus interface and servo motor, and etc. The system mainly realizes the network communication function and motion control function.First of all, it builds a reliable and efficient real-time embedded operating environment. In this paper, the software environment chooses "the host machine and the target machine" architecture. The host machine is based on embedded Linux system and the target machine is based on Zedboard development board. Based on embedded technology, ARM technology and FPGA technology, it completes the construction of the cross-compile environment and the transplantation and custom of the operating system platform. It evaluates the real-time performance of the Linux system through Xenomai, which finally makes it meet the real-time demand of motion control system.Then, it describes the implementation of motion controller with POWERLINK master station in detail. Combined with the architecture characteristics of real-time and kernel layer in the Linux & Xenomai system and based on Zedboard development board, it realizes the POWERLINK master station of motion control system. Most of functional modules are implemented on the ARM, including POWERLINK network communication module and motion control module. In order to improve the communication cycle of POWERLINK, network management state machine, data link layer and cyclical data object in the protocol stack are realized in the FPGA. Network communication module mainly realizes POWERLINK protocol stack and motion control module is mainly programmed according to the CIA 402 and CIA 401 specification. The whole system communicates in real-time through the Ethernet POWERLINK.Next, this research analyzes the network delay of the Ethernet POWERLINK and designs a feedforward & feedback PID compensation algorithm. In order to improve the adverse impact of the network delay on the system real-time performance, it simplifies the mathematical models of the speed loop and position loop of the system. A feedforward & feedback PID compensation algorithm is designed. It establishes the simulation model with the simulation software, carries out simulation analysis and realizes the network delay compensation algorithm by programming. Simulation results show that the performance of signal tracking of the system with algorithm compensation is improved greatly.At last, this paper sets up the experimental platform of the motion control system which is based on Linux & Xenomai system with Ethernet POWERLINK. It analyzes the results after an experiment. The experiment includes the test related to network communication with Ethernet POWERLINK and the test related to servo driver control. The experimental results show that the overall system is feasible, and can achieve the desired network communication and motion control functions. Meanwhile, position tracking performance of the system with algorithm compensation is improved obviously. Not only the hardware structure of the system is simple, but also the software resources of the system are rich. It is more reliable and reconfigurable, which is extremely competitive in the current market.