Speed Control of BLDCM for Industrial Sewing Machine Based on dSPACE
By the prompt development of textile industry, the design of t the equipment of textile industry is developing towards high precision, multifunction and low cost. ISM is one of the key equipment in textile industry. It is composed of motor servo system and the traditional sewing machine, so it has some additional functions besides all of the functions provided by traditional sewing machine, for which the efficiency of product line can be improved greatly by its high automation, as well as the product quality. The precision of ISM is mainly determined by the control precision of the motor servo system. Therefore, it is essential of the research on the motor servo system to developing an ISM system. Presently, classic PID control theory is mainly adopted in the research on speed control of the motor servo system of ISM. Based on the classic PID control theory, increment-PI control algorithm or fuzzy PID control theory are also adopted in this research field [6,12,13]. Considering speediness, stabilization precision and overshoot of the ISM’s motor servo system which is a characterization of non-linear and strong coupled system, a comparison result among classic PID, increment-PI and fuzzyPID shows that the effect of control by Fuzzy PID control algorithm is the best, increment-PI algorithm takes second place and the classic PID control algorithm is the worst . The precise mathematical model of the object will be not required if fuzzy PID control algorithm is adopted. Instead, the fuzzy control rules, which would be used to determine the control volume, are built mainly based on the experience of the developers and experts. However, if the developer is not enough familiar with ISM, the application of fuzzy control will cost huge time on making the fuzzy control rules for ISM’s motor servo system, and the fuzzy control rules may be half-baked or even incorrect. Consequently, the effect of fuzzy control will be much worse than classic PID control . An algorithm is needed when the developers just have a few knowledge of the ISM, which will spend much less time to realize the motor control. The differentiator in classic PID control improves the system’s fast response characteristic, but it amplifies the external disturbance, which has a bad impact on the reliability of the system. Hence, the differentiator is deleted in the application of increment-PI control considering the complex industry environment for the ISM’s well working. This method is simple and practical, but the dynamic response, speed smooth and reliability of the system needs improving if this method is applied in ISM . Combined various matured control theories with the author’s experiences in practical engineering, the control algorithm is improved based on the increment-PI control algorithm mentioned above. And a HILS with traditional sewing machine and motor is implemented by dSPACE. It is showed that the effect of the improved control algorithm is satisfied.
The conventional algorithms applied in Industrial Sewing Machine (ISM) have some disadvantages such as less precision, lower speed response and more complicated due to the special requirement of the ISM. Taking the mechanical characteristics of ISM into consideration, a changing-dead-zone multi-segment-PID with digital filter algorithm is presented in this paper, which is applied in the speed control of the Brushless DC Motor (BLDCM) servo system. The improvement process of the arithmetic is specified in the paper. The mathematical model of the speed control of servo system is built by MATLAB/Simulink, and the online simulation is implemented on the software-hardware platform provided by Hardware-in-theLoop Simulation (HILS) tool dSPACE. The simulation result is showed that the improved algorithm is more effective in the dynamic performance of the system compared with the increment-PI, and it also works well in the application of the practical product of ISM.
The schematic diagram of the hardware design of ISM’s servo system shows in Fig. 1. The driving motor of the ISM adopts BLDCM with four poles and three-phase whose power reaches 400W. According to the characteristics of the BLDCM and the function requirement of the ISM, PIC18F4431 microcontroller is adopted as the core control solutions of the hardware of this system. BLDCM is driven by square wave with full-bridge style. In Fig.1, the PWM signal which is used to drive the BLDCM working is produced by ISM’s motor servo system. And the servo system implements the functions according to the defined signal of the control panel. The speed signal is detected by the HALL sensors of the BLDCM.
INTRODUCTION OF DSPACE
DSPACE is a real-time HILS system with a softwarehardware platform based MATLAB/Simulink which is developed by Company dSPACE. Gapless link is realized between dSPACE and MATLAB/Simulink excellently . HILS is a technique of real-time simulation aimed at practical process, which could help simulation with the practical object in the simulation loop, helps adjusting the control parameters online, and improving the performance of the real system [1,4,14]. The HILS system dSPACE is composed with two parts, software and hardware. Ds1104, a single board system, is based on to discus in this paper. The software of dSPACE is mainly included: 1) RTI: Real-Time Interface. It is used to translate the source code and download the object code into the hardware of dSPACE. But it should complete the work with MATLAB/Simulink which is produced by Company MathWorks together. 2) ControlDesk: dSPACE’s well-established experiment software, provides all the functions to control, monitor automate experiments and make the development of controllers more efficient. 3) MLIB/MTRACE:MATLAB-dSPACE Interface Libraries. It gives the access to dSPACE real-time processor hardware from the MATLAB workspace. 4) RealMotion: real-time animation analyze software. 5) CLIB: Communicating with real-time microcontroller and PC. In the hardware of dSPACE, real-time microcontroller and I/O interfaces are integrated into one real-time simulation system, which have higher calculated capacity. It integrates analog input interface, analog output interface, digital I/O and etc. Then HILS can realize on the hardware platform connected with data interface. HILS need build mathematical model before all the work is doing, which can not be done in dSPACE system. Generally, mathematical model is built in the visual modeling tool MATLAB/Simulink, then translated into object files by MATLAB/Simulink/RTW that could be identified by the dSPACE system. Real-time data monitor interface created by ControlDesk composed dSPACE hardware control unit is connected realtime data that need to be monitored. Curve graph and table are generated in the monitor interface which gives the referenced information. Before the development of hardware of ISM’s servo system is completed, control algorithm can be simulated and modified by dSPACE, combined with the sewing machine and driving motor.
The changing-dead-zone multi-segment-PID algorithm with digital filter described in this paper is a simple and practical control algorithm, which is summarized by the practical engineering experience. And the dynamic performance of the system can be implemented well by this algorithm. Combined with BLDC Motor and ISM, this algorithm is implemented in the dSPACE simulation platform which is a tool of HILS. From the simulation result, it is much better in the speed response and stability of the control system with the algorithm discussed in the paper compared with the increment-PI. Therefore, changing-dead-zone multi-segmentPID algorithm with digital filter is adapted to speed control of the ISM’s BLDCM servo system with a definite precision. Besides, most of the industrial equipment with four-bar mechanism and crank slide mechanism could use the algorithm, which would be also adapted to them. Currently, the ISM, whose speed control is implemented by this algorithm, has been a product, which receives customers’ consistent approved.