Yazar "Korpe, Ugur Ufuk" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Data-Driven TD3 Control of IM Considering Magnetic Saturation and Temperature Effect
    (Ieee, 2024) Korpe, Ugur Ufuk; Gokdag, Mustafa; Gulbudak, Ozan
    Induction machines (IM) are still widely used in the industry due to their advantages, such as low maintenance requirements and improved robustness. The field-oriented control (FOC), direct torque control (DTC), and model predictive control (MPC) techniques are used to control IM in high-performance control applications. The common disadvantage of these control techniques is that the control performances are negatively affected by changes in machine parameters, and machine parameters vary non-linearly depending on the magnetic saturation and temperature. To solve this negative affect, the control technique can be optimized by using a parameter estimation methods. Another solution to eliminate these negative effects is to design a reinforcement learning (RL)-based controller that regulates the control variables without the knowledge of machine parameters. In this study, IM speed control is performed using a twin-delayed deep deterministic policy gradient (TD3) agent. The dynamic and steady-state performance of the designed controller are compared with the traditional control techniques. Extensive simulation results have shown that the dynamic and steady-state performance of the designed controller is better than other control techniques.
  • Küçük Resim Yok
    Öğe
    Modulated Model Predictive Control of Permanent Magnet Synchronous Motors with Improved Steady-State Performance
    (Ieee, 2021) Korpe, Ugur Ufuk; Gokdag, Mustafa; Koc, Mikail; Gulbudak, Ozan
    Finite Control Set Model Predictive Control (FCS-MPC) is an optimal control strategy that predicts the future trends of the control goals by assessing the discrete-time model of the system. FCS-MPC has many advantages, such as it has a fast dynamic response, and nonlinearities can be controlled by the customized cost function. Besides the featured benefits of the FCS-MPC strategy, the ripple in the output variable (in most cases, control variable) may be problematic due to the uncontrolled switching frequency. For that reason, the MPC-based closed-loop strategy offers a better regulation performance at high-sampling frequency. However, the selection of a low sampling rate causes an unpleasant distortion or poor power quality. A modulated model predictive control method is proposed in this work to suppress the unwanted distortion in the control variable. In the proposed method, a space vector modulator is integrated into the FCS-MPC-based control method to attain a fixed-switching frequency. By doing so, the distortions and unwanted harmonics are significantly decreased. In this paper, a modulated model predictive control (M2PC) method is proposed for controlling the permanent magnet synchronous motor. The proposed method calculates the dwell-time of the modulator stage by assessing the multi-objective cost function. The noticeable lower distortions in the stator currents are obtained by the proposed routine. All theoretical concepts are verified by extensive simulations. Based on the simulation results, the proposed method provides a better control performance for permanent magnet synchronous motors (PMSM). Furthermore, the proposed modulated MPC strategy offers superior steady-state performance compared to the conventional MPC method in all regards.
  • Küçük Resim Yok
    Öğe
    Speed Control of IM Using RL-Based TD3 Agent
    (Ieee, 2024) Korpe, Ugur Ufuk; Gokdag, Mustafa; Gulbudak, Ozan
    Induction machines (IM) have been used daily since the 19th century. The induction machine should be effectively controlled to achieve high performance. Since the late 20th century and 21st century, field-oriented control (FOC), direct torque control (DTC), and model predictive control (MPC) techniques have been used in high-performance control applications. However, these techniques are dependent on the motor parameters and inverter model. These parameters change non-linearly depending on magnetic saturation, temperature, and operating point which negatively affects the performance of the system. In order to eliminate the negative effects of parameter changes, reinforcement learning (RL)-based methods have become increasingly popular in the literature in recent years. In this study, for the first time, the speed control of IM is performed using TD3 agent, which is one of the RL-based methods. The dynamic and steady-state performance of the control system designed with the TD3 agent is compared with the traditional FOC technique. Extensive simulation results have shown the robustness of the proposed drive system.