کنترل مقاوم بر اساس کنترل پیش بینی تعمیم یافته به حلقه جریان موتور رلوکتانسی سوییچ شده Robust Control Based on Generalized Predictive Control Applied to Switched Reluctance Motor Current Loop

۱ Introduction SRMs are an alternative and modern solution to electromechanical conversion with variable speed. The availability of highfrequency switching devices and improvements in machine design, associated with SRM intrinsic simplicity, reliability, low cost, high power capacity, and fault tolerant operation have made it a viable replacement for conventional motor drives [1,2]. SRMs have been traditionally controlled by either open loop hysteresis or closed-loop pulse-width-modulation (PWM) current controllers. Each scheme presents advantages and drawbacks with regard to parametric variations, accuracy, robustness, and dynamic response over the entire speed range. The hysteresis current controller is popular because of its inexpensive, simple, and easy-to-use architecture [3–۵]. However, there are well known disadvantages, such as variable switching frequency and high ripple current, making it undesirable for many applications. On the other hand, PWM controllers provide better control loop characteristics compared to their hysteresis counterparts, although they are more complex to be designed and require more computation effort, such drawbacks can be overcome by using digital signal processors (DSPs). In addition, in order to achieve improved efficiency the SRM must operate under magnetic saturation [6]. This effect associated with the current level and the variation of magnetic reluctance with respect to rotor position results in highly nonlinear dynamics of all the machine relevant characteristics. For the high performance operation of SRM drives, the current loop plays a major role, mainly at low speeds where the torque is only limited by the current. As a consequence, to achieve good torque control, accurate current command tracking control is required [1,2]. However if the current is not properly modulated and switched at the correct rotor position all negative effects e.g., high torque pulsation and acoustic noise are intensified. As the speed increases, the back electromotive force (EMF) increases to a level where the available voltage becomes insufficient to regulate the current, while the control system should naturally assume single pulse mode to achieve the maximum available voltage for highspeed operation. The torque can then be controlled only by properly adjusting the angles of the current pulses [6]. The ability of tracking the dynamic setpoint and recovering from load disturbances without torque ripple are two important challenges for high performance in SRM drives [7,8]. Several researchers have proposed current profiling-based methods to minimize torque ripple [1,2,9,10]. Often, proportional-integral (PI) current regulators have been applied in SRM drives with limited performance [2,11]. However, due to the aforementioned nonlinear plant characteristics, good performance, and stable operation are difficult to achieve over the entire operating range. The nonlinear characteristic of the SRM model represents a challenge to classical and advanced control techniques, thus motivating several researchers to propose current control techniques to overcome this drawback [2,11–۱۴].