Abstract: Based on DSP Builder and Quartus II, this paper introduces the application of modern DSP technology in electro-hydraulic servo systems. A digital PID controller was designed using DSP Builder in conjunction with a strip coiling electro-hydraulic servo system. Simulation results show that the proposed method is effective. Keywords: DSP Builder; Electric-hydraulic servo system; FPGA [b][align=center]The Design on Electric-hydraulic Servo-system Controller Based on DSP Builder CHAO Zhiqiang, LI Huaying, CHEN Qiang, SU Ligang[/align][/b] Abstract: The technology of modern DSP for electric-hydraulic servo-system based on DSP Builder and Quartus Ⅱ was introduced. With the strip steel batching electric-hydraulic servo2 system, the digital PID controller was designed using DSP Builder. The simulation result shows that the method is effective. Keywords: DSP Builder; Electric-hydraulic servo-system; FPGA 0 Introduction With the development of electro-hydraulic servo control theory, many advanced control strategies have been applied to the field of electro-hydraulic servo control. For example, literature [1] describes a filter-type two-degree-of-freedom control algorithm with incomplete differential PID as the basic operation, and conducts simulation research on the characteristics of hydraulic servo system for flight simulation turntable. Reference [2] studied the application of PID control based on RBFNN in electro-hydraulic position servo system. Reference [3] adopted sliding mode variable structure control for electro-hydraulic position servo system and designed sliding mode plane with optimal control theory, and achieved good results. However, most of the literature is theoretical and simulation research, and most industrial applications still mainly use analog circuits to implement PID control algorithm. The main reason is that the methods to implement these advanced control algorithms are currently implemented by the lower computer responsible for control using programs, and the computer is prone to crashes, power failures, etc., which reduces the reliability and safety of the hydraulic system. The author introduces a method to design an electro-hydraulic servo system controller using FPGA-based DSP technology. This method overcomes some shortcomings of traditional servo controllers, can implement many complex real-time control algorithms in hardware, and adjust the control algorithm according to the quality of the control effect, thereby improving the control effect, calculation speed and reliability of the controller. Using this method, the designer does not need to be very familiar with FPGA (programmable gate array) and VHDL (hardware description language), and can design the required servo controller in Matlab. 1. Overview of Modern DSP Technology In recent years, digital controllers designed using digital signal processing (DSP) technology have been increasingly applied to electro-hydraulic servo systems. For a long time, DSP processors, represented by the TMS320 series from Texas Instruments (TI), were almost the only choice for digital signal processing applications. However, facing today's rapidly changing DSP application market, the immutability of their hardware structure has long proven inadequate. Modern FPGA-based DSP technology uses FPGAs and other programmable gate arrays to implement digital signal processing algorithms. It is an object-oriented DSP system, allowing users to customize and configure their own DSP systems according to their needs. However, the highly specialized nature of FPGA-based DSP system development significantly limits its application. Currently, new design tools and workflows exist for developing and applying DSP systems using FPGAs. The world's two largest FPGA manufacturers, Xilinx and Altera, have both launched their own DSP solutions. DSP Builder is a system-level tool for DSP development launched by Altera. MathsWorks' Matlab is a powerful mathematical analysis tool. Simulink is a toolbox within Matlab used for graphical modeling and simulation. DSP Builder, a toolbox within Simulink, enables the design of DSP systems using FPGAs through Simulink's graphical interface. The basic modules in DSP Builder are described at the algorithm level, making them easy for users to understand from a system or algorithmic perspective, even without a thorough understanding of FPGAs and hardware description languages. This provides convenient top-down, algorithm-level design for engineers in traditional control systems to develop reliable FPGA-based control system chips. 2. Mathematical Model of Electro-hydraulic Position Servo System An electro-hydraulic servo system is a system that combines electrical and hydraulic control methods. A typical block diagram of an electro-hydraulic system is shown in Figure 1. [align=center] [img=430,108]http://www.chuandong.com/uploadpic/THESIS/2009/5/2009050813435793528O.jpg[/img] [/align] [align=center]Figure 1 Block diagram of an electro-hydraulic servo system[/align] [align=left] The control element can be a hydraulic control valve or a hydraulic servo variable pump, etc., and the actuator can be a hydraulic cylinder or a hydraulic motor, etc. Based on the parameters of the electro-hydraulic servo valve and hydraulic cylinder in the strip coiling electro-hydraulic servo system in the literature [5], the author studies how to use Matlab and DSP Builder to design the controller of the electro-hydraulic servo system. 2.1 Electro-hydraulic Servo Valve The electro-hydraulic servo valve is considered as a second-order oscillating element, and its transfer function can be written in the following form: [img=330,81]http://www.chuandong.com/uploadpic/THESIS/2009/5/20090508134503789658.jpg[/img] Where: Ksv is the flow gain of the servo valve; ωsv is the natural frequency of the servo valve; ξsv is the damping ratio of the servo valve. A TR2h7/20EF type moving-coil two-stage slide valve position feedback electro-hydraulic servo valve was adopted. Its main parameters are: rated current ΔiR = 0.13A; oil supply pressure ps = 415MPa; rated flow rate qR = 0.15 ×10⁻³ m³/s; zero-position leakage flow rate qc = 8.13 ×10⁻⁶ m³/s; chatter current amplitude and frequency are 25mA and 50Hz, respectively. Experiments show that the servo valve's natural frequency ωsv = 112 rad/s and damping ratio ξsv = 0.6. The transfer function of the servo valve is obtained as follows: [img=327,94]http://www.chuandong.com/uploadpic/THESIS/2009/5/20090508134605501105.jpg[/img] Let the sampling period of the control system be 0.11 s, the pulse transfer function of the servo valve is obtained as follows: [img=339,74]http://www.chuandong.com/uploadpic/THESIS/2009/5/2009050813473813860K.jpg[/img] 2.2 Hydraulic Cylinder-Load If the load is an inertial load, the transfer function of the hydraulic cylinder-load link can be written in the following form: [img=304,104]http://www.chuandong.com/uploadpic/THESIS/2009/5/2009050813481113126H.jpg[/img] Where: XP is the piston displacement of the hydraulic cylinder; QL is the load flow rate; AP is the effective working area of ​​the hydraulic cylinder; ωn is the natural frequency of the hydraulic cylinder; ξh is the damping ratio of the hydraulic cylinder. The technical parameters of the hydraulic cylinder are: piston diameter D = 0.1125m, piston rod diameter d = 0.106m, piston stroke H = ±0.1075m, effective working area of ​​the hydraulic cylinder AP = 9145 ×10 - 3m2, total compression volume of the system Vt = 2HAP +V pipe ≈ 2148 ×10 - 3m3. If the elastic modulus of the hydraulic oil is βe = 7 × 10⁸ Pa and the inertial load mass is mt = 2175 × 10⁴ kg, then the natural frequency of the hydraulic cylinder-load link is: Since the viscous damping coefficient of this link and the flow-pressure coefficient of the servo valve involved are both small, we take ξh = 0.12. The transfer function of the hydraulic cylinder-load link can then be obtained as: [img=329,109]http://www.chuandong.com/uploadpic/THESIS/2009/5/20090508135029120820G.jpg[/img] Let the sampling period of the control system be 0.11 s, the pulse transfer function of the servo valve can be obtained as: [img=447,65]http://www.chuandong.com/uploadpic/THESIS/2009/5/2009050813505695254W.jpg[/img] 3 Design of PID Controller for Electro-hydraulic Servo System Based on DSP Builder 3.1 Control System Structure Design The servo controller design can start from the system level, which is completely independent of hardware. First, the powerful system design and analysis capabilities of Matlab and the modules provided by DSP Builder are used to complete the structural design of the control system. The controller in this paper adopts a positional PID controller, and the model shown in Figure 2 is built in Simulink. [img=498,133]http://www.chuandong.com/uploadpic/THESIS/2009/5/20090513103214833973.jpg[/img] Figure 2 Top-level Simulink model of the control system. In Figure 2, the top-level model has PID Controller as the PID control subsystem, Input as the control input, Feedback as the feedback input, and Function1 and Function2 as the discrete mathematical models of the servo valve and hydraulic cylinder-load, respectively. It is worth noting that the Mask Type in the PID subsystem must be set to SubSystem AlteraBlockSet; otherwise, only Simulink simulation can be performed, and SingnalCom2piler analysis cannot be performed. The PID control subsystem is the part that implements the PID algorithm, and its block diagram is shown in Figure 3. Figure 3 Block diagram of PID subsystem. Both the control input and feedback input use 16-bit precision. Since DSP Builder does not yet support floating-point arithmetic, a bit-level conversion method is used for the PID coefficients to achieve precise adjustment of the proportional, integral, and derivative coefficients. First, the PID coefficients are converted to integers and amplified to 24 bits. Then, the bus converter unit in IO&Bus after the parallel adder unit converts the accumulated data to 16 bits, representing floating-point numbers that are multiples of 1/256 = 010039, thus enabling floating-point arithmetic in the FPGA. 3.2 Control System Simulation In this example, PD control is used, with a proportional coefficient of 1715 and a derivative coefficient of 4. Correspondingly, Kp = 1715 × 256 = 4480 and Kd = 4 × 256 = 1024 are set. The closed-loop step response and sinusoidal response of the system are shown in Figures 4 and 5. [img=491,223]http://www.chuandong.com/uploadpic/THESIS/2009/5/2009050813585465484H.jpg[/img] 3.3 FPGA Implementation of the Controller Double-click the SingnalCompiler module in the ServoSystem model, and follow the prompts to select the device, synthesis, and optimization tools. Here, we select the EP2C8 FPGA, Quartus II as the synthesis tool, and Balanced as the optimization method. Considering both computational speed and resource consumption, we compile and generate ServoSystem1qpf. Open ServoSystem1qpf in Quartus II, and you can see the automatically generated VHDL source code from SingnalCompiler. Complete the compilation and adaptation process in Quartus II. The generated pof and sof files can be directly used for FPGA programming configuration. For the configured controller, connect the Input terminal to the computer's given value, the Feed2back terminal to the displacement feedback A/D chip, and the Output terminal to the D/A output. 4. Conclusion This paper takes the design of a strip winding electro-hydraulic servo system controller using the FPGA system-level design tool DSP Builder as an example to introduce the application of modern DSP technology in electro-hydraulic servo systems. This method can resolve the limitations of discrete components and the contradictions in real-time performance and reliability of software implementation when complex control algorithms are applied in practical electro-hydraulic servo systems. With system-level design tools like DSP Builder, the design begins with Matlab system-level simulation completely independent of hardware. This allows engineers in traditional control fields to quickly apply algorithm-level concepts to control system design, enabling them to focus their limited energy on system-level algorithm design and avoid getting bogged down in repetitive and tedious circuit design. It is foreseeable that with the development of control theory and electronic technology, this method will be widely used in the design of future electro-hydraulic servo control systems. References [1] Cao Jian, Li Shangyi, Zhao Keding. Two-degree-of-freedom PID control of electro-hydraulic servo system [J]. Machine Tool & Hydraulics, 2001 (6). [2] Zhang Dehua, Wu Yonghong, Duan Suolin. PID control based on RBFNN and its application in electro-hydraulic position servo system [J]. Journal of Taiyuan University of Science and Technology, 2006, 27 (5). [3] Tang Qingbo, Zhang Wenhan. Research on the application of optimal sliding mode control in electro-hydraulic servo system [J]. Hydraulics & Pneumatics, 2007 (1). [4] Pan Song, Huang Jiye, Zeng Yu. Practical tutorial on SOPC technology [M]. Tsinghua University Press, 200513. [5] Wang Jiwei, Wu Zhenshun. Fundamentals of Control Engineering [M]. Higher Education Press, 200118.