Kyo-Beum Lee

Torque ripple reduction in DTC of induction motor driven by three-level inverter with low switching frequency

A torque ripple reduction technique of direct torque control (DTC) for high power induction motors driven by three-level inverters with the inverter switching frequency limited around 0.5-1 kHz level is presented. It is noted that conventional two-level DTC algorithms to reduce torque ripple are devised for applications with relatively high switching frequency above 2-3 kHz and cannot accomplish satisfactory torque ripple reduction for three-level inverter systems with such lower switching frequencies. A new DTC algorithm, especially for low switching frequency inverter system, illustrates quite reduced torque ripple characteristics all over the operating speed region. Simulation and experimental results show effectiveness of the proposed control algorithm.

Sensorless DTC-SVM for Induction Motor Driven by a Matrix Converter Using a Parameter Estimation Strategy

This paper presents a new direct torque controlled space vector modulated method to improve the sensorless performance of matrix converter drives using a parameter estimation scheme. The flux and torque error are geometrically combined in a new flux leakage vector to make a stator command voltage vector in a deadbeat manner. A new sensorless method of estimating the rotor speed, flux, stator resistance, and rotor resistance is derived and verified with experimental results. Common terms in the error dynamics are utilized to find a simpler error model involving some auxiliary variables. Using this error model, the state estimation problem is converted into a parameter estimation problem assuming the rotor speed is constant. The proposed adaptive schemes are determined so that the whole system is stable in the sense of Lyapunov. The effectiveness of the proposed algorithm is verified by experiments.

Method for Detecting an Open-Switch Fault in a Grid-Connected NPC Inverter System

This paper proposes a fault-detection method for an open-switch fault in the switches of grid-connected neutral-point-clamped inverter systems. The proposed method can not only detect the fault condition but also identify the location of the faulty switch. In the proposed method, which is designed by incorporating a simple switching control in the conventional method, the fault condition is detected on the basis of the radius of the Concordia current pattern, and the location of the faulty switch can be identified. By using the proposed method, it is possible to detect the open-switch fault and identify the faulty switch within two fundamental periods, without using additional sensors or performing complex calculations. Simulations and experiments are carried out to confirm the reliability of the proposed fault-detection method.

New Modulation Techniques for a Leakage Current Reduction and a Neutral-Point Voltage Balance in Transformerless Photovoltaic Systems Using a Three-Level Inverter

Transformerless topologies of many topologies are widely used in photovoltaic (PV) systems because these topologies have many advantages in terms of the weight, size, and efficiency. A three-level inverter has an outstanding performance and is advantageous in the switching device selection than a two-level inverter. In the transformerless PV systems using the three-level inverter, the PV systems should suffer from the leakage current and generate the neutral-point voltage unbalance. To solve two problems, this paper proposes modulation techniques to reduce the leakage current and balance the dc-link voltages. The cause of the leakage current in the three-level inverter is analyzed. The proposed technique LMZVM using the large, medium, and zero vectors reduces the common mode voltage that causes the leakage current than that of the conventional PWM. Moreover, the proposed technique LMSVM using the large, medium, and small vectors balances the dc-link voltages with reduced CMV as the same in LMZVM. The effectiveness of the proposed techniques is verified by comparing its results with those of the convectional PWM. The results are obtained through simulations and experiments.

Study and Handling Methods of Power IGBT Module Failures in Power Electronic Converter Systems

Power electronics plays an important role in a wide range of applications in order to achieve high efficiency and performance. Increasing efforts are being made to improve the reliability of power electronics systems to ensure compliance with more stringent constraints on cost, safety, and availability in different applications. This paper presents an overview of the major failure mechanisms of IGBT modules and their handling methods in power converter systems improving reliability. The major failure mechanisms of IGBT modules are presented first, and methods for predicting lifetime and estimating the junction temperature of IGBT modules are then discussed. Subsequently, different methods for detecting open- and short-circuit faults are presented. Finally, fault-tolerant strategies for improving the reliability of power electronic systems under field operation are explained and compared in terms of performance and cost.

Performance Improvement of LCL -Filter-Based Grid-Connected Inverters Using PQR Power Transformation

The demand for three-phase pulsewidth modulation (PWM) inverters in applications such as power control or grid connecting has been on the increase in recent years. Such inverters are connected to the grid via an L filter or an LCL filter to reduce the harmonics caused by the switching. An LCL filter can reduce the harmonics induced by low switching frequency and generates a satisfactory level of grid-side current using a relatively low inductance, as compared to an L filter. The additional poles introduced by the LC part induces resonance in the system, leading to stability problems; this paper presents a compensation method using power theory to improve these issues, so that the performance of the designed LCL filter system can be improved. The effectiveness of the proposed algorithm is verified by simulations and experiments.

Diagnosis and Tolerant Strategy of an Open-Switch Fault for T-Type Three-Level Inverter Systems

This paper proposes a new diagnosis method of an open-switch fault and fault-tolerant control strategy for T-type three-level inverter systems. The location of the faulty switch can be identified by the average of the normalized phase current and the change of the neutral-point voltage. The proposed fault-tolerant strategy is explained by dividing into two cases: the faulty condition of half-bridge switches and neutral-point switches. The performance of the T-type inverter system improves considerably by the proposed fault-tolerant algorithm when a switch fails. The proposed method does not require additional components and complex calculations. Simulation and experimental results verify the feasibility of the proposed fault diagnosis and fault-tolerant control strategy.

Dynamic Performance Improvement of AC/DC Converter Using Model Predictive Direct Power Control With Finite Control Set

This paper presents a control scheme for the dynamic performance improvement of an AC/DC converter using the model predictive direct power control (MPDPC) with a duty cycle. In the MPDPC, the active and reactive power is simultaneously controlled with a single cost function. If either of the two control targets has a large power variation, the control weight is concentrated on one side, which causes mutual interference. Because of such mutual interference, the control dynamics of the AC/DC converter deteriorates. Due to the control weight being concentrated on one side using the single cost function, even if the control dynamics of the other side decreases, the dynamic performance of the system is improved by reconfiguring the cost function that has the weighting factor to minimize the decline of the system dynamics that is caused by the mutual interference. The effectiveness of the proposed control scheme is verified by comparing its results with those of the conventional MPDPC. The results are obtained through the simulations and experiments.

Torque-Ripple Minimization and Fast Dynamic Scheme for Torque Predictive Control of Permanent-Magnet Synchronous Motors

This paper proposes a simple and effective method to reduce torque ripples for the torque predictive control (TPC) of permanent-magnet synchronous motors (PMSMs). The conventional TPC analyzes the relationship among the electrical torque, stator flux, and stator voltage using the magnitude of the stator voltage vector of PMSMs to obtain the angle of the reference voltage vector and accurately control the torque. In addition, the stator-flux control uses the hysteresis method. However, the voltage vector that can be chosen in an inverter is limited because the conventional TPC fixes the magnitude of the reference voltage vector, and thus, a large torque ripple is generated in the low-speed region. The proposed TPC does not fix but varies the magnitude of the reference voltage vector using both the torque and flux error information. Therefore, it not only has the fast dynamic of a direct torque control but also can reduce effectively the torque ripple. The proposed method is proven by the simulation and experimental results, and the proposed algorithm provides an excellent steady-state response and fast dynamics.

Improvement of low-speed operation performance of DTC for three-level inverter-fed induction motors

A direct torque control algorithm for three-level inverter-fed induction motors is presented. Basic voltage selection methods similar to a two-level inverter provoke some problems such as stator-flux drooping phenomenon and undesirable torque control deterioration appeared, especially at low-speed operation. To overcome these problems, an algorithm with the basic switching sectors subdivided and intermediate voltage vectors applied is proposed in this paper. This algorithm basically considers applications in which direct torque-controlled induction motors are fed by three-level inverters with maximum switching frequency lowered around 1 kHz. An adaptive observer is also employed to bring better responses at the low-speed operation, by estimating some state variables and motor parameters which take a deep effect on the performance of the low-speed operation. Simulation and experiment results verify effectiveness of the proposed algorithm.