Search

In this paper, a new reactive power compensation method with a state observer and repetitive control is proposed for voltage regulation of the low voltage distribution line. Reactive power compensator has a single-phase full bridge inverter structure and consists of DC link voltage controller to maintain a constant DC voltage and current controller to inject a reactive current for voltage regulation. A state observer is proposed to extract the second harmonic component of the DC link voltage and then a dc component, which is calculated by eliminating the DC link ripple component, is applied as the voltage controller feedback component. By using the state observer, the feedback 120 Hz DC link voltage ripple component inherited in the single-phase system is reduced to 2V from 66V in the peak-to peak magnitude. In order to inject a clean reactive current, PI controller is employed for the fundamental component control and a repetitive controller is connected in parallel with the PI controller to compensate for lower-order odd harmonics, where both controllers are implemented in the synchronous reference frame. The proposed control method is verified in a ±5kVar single-phase reactive power compensator by simulation and experiment, and shows that THD is reduced to 2% from 9% even in the small DC link capacitance.

This paper proposes design rule to determine an effective passive damping resistor of grid-connected inverter with LCL filter for industry applications, where it considers the stability of current control and minimum power loss by defining three standard criteria based on stability analysis. Despite of losses due to resistor, passive damping is the simplest way to reduce filter resonance, which is widely used in the industrial field. However, there is no clear criterion for determining damping resistor. This paper suggests three standard criteria to determine an effective damping resistor that can guarantee system stability for industrial applications. The analytic equivalent model takes into account system delays that are composed of processing and PWM delays. The applicable range of damping resistors that should be able to attenuate the filter resonance are mathematically derived from equivalent modeling and parameters of the system. To verify the validity of the effective damping resistor that is determined, 7 kW grid-connected inverter is simulated and implemented. Experimental results are matched to simulation results and analysis; in particular, the stability of current control is completely guaranteed with a low total harmonic distortion of 3.6% and minimum power loss of 28.79 W (0.41%) with effective damping resistor.

To date, the unbalance phenomenon of the three-phase power system has been attracted much attention due to the difference of load condition between main and teaser phase. Herein, we propose a model predictive control (MPC) with low switching frequency for a railway power compensator to balance active and reactive power in high-power railway system. It is found that the rate of unbalance is reduced from 10.7% to 7.5% corresponding to downscale Main 5kW and Teaser 1kW-220V with low frequency of 600Hz. Finally, both simulation and experiment results are presented to verify the effectiveness of MPC.

In this paper, a reactive power compensator (RPC) is used for stabilization of the low voltage distribution line, and a novel reactive power control method of the RPC is proposed to reduce size as well as maintain its long lifetime. In the single-phase system, a 2nd harmonic component is inevitably generated in the DC link voltage, and this phenomenon gets even worse in case that the capacity of DC link capacitor decreases in order to reduce the size and cost of the RPC. A reactive power control is performed by controlling the d-axis current in the virtual synchronous reference frame, and the voltage control for maintaining the DC link voltage is implemented through the q-axis current control. The proposed method for compensating the ripple voltage effect consists of an extraction unit of DC link ripple voltage, high pass filter (HPF), and time delay unit. The HPF removes an offset component of DC link ripple voltage caused by an integrator, and a time delay unit compensates the phase leading effect due to the HPF. The compensated DC link voltage is used as a feedback component of voltage control loop in order to supply the stable reactive power. The performance of the proposed algorithm is verified through simulation and experiments. In case that DC link capacitor is reduced to 375uF, an effect of ripple voltage is reduced to 8V pp from 74V pp in the voltage control loop, and total harmonic distortion (THD) of the current is improved.

The neutron measurement system (NMS) used in research reactors to measure thermal neutron flux emitted from the reactor core, provides the neutron power signal that controls the reactor power with control rods. The neutron power signals from NMS are key components of the reactor protection system (RPS) and the reactor regulating system (RRS). Therefore, the NMS having accuracy and reliability is a significant component in the reactor operation. In particular, the linearity in the transition range between the pulse count and Campbell signal, as well as, the log power and linear power signal is considered a top requirement for reactor power control. However, analogue-based NMS has had difficulties in fitting the curve of different signals, such as, the pulse count and Campbell signal. Moreover, the full power alignment cost much trial and error. Digital NMS in research reactors, herein, which are being applied to Jordan Research and Training Reactor (JRTR), can improve the linearity performance and save time by using implemented calibration method. Herein, the detailed calibration of digital NMS is introduced and the implementation method is described in detail with actual test results.

In this paper, verification of operation and control of a vertical stabilization (VS) converter for a superconducting coil by using a real-time digital simulator (RTDS) is described. The VS converter supplies voltage (±1000 V) and current (±22.5 kA) to the superconducting coil for plasma current VS, and four-quadrant operation must be achieved without a zero-current discontinuous section. To remove any zero-current discontinuous sections, circulation current control is required, but it is hard to analyze and design it because of the characteristics of thyristor nonlinearity. In this paper, delay time of the thyristor and processor is implemented using a zero-order hold, and an equivalent model is designed and analyzed. Circulation current control and difference current control are similarly analyzed because both equivalent circuits are the same, and are satisfied with a 54° phase margin and a 11 ms step response. In addition, gamma control is designed to restrict the firing angle to prevent commutation failure and commutation failure is prevented by the proposed gamma control, even if a short circuit or sag occurs in the grid. To verify the feasibility of the VS converter, a real Zynq-based controller is assembled and a power system, including the VS converter, is implemented in an RTDS for the hardware-in-the-loop simulation. Moreover, the small-scaled VS converter is built and tested. Experimental results are well-matched to simulation results and analysis.

In this paper, verification of a four-series connected central solenoid (CS) converter sequence control of a nuclear fusion reactor to reduce reactive power by using a real-time digital simulator is described. The unit CS converter supplies voltage (±975 V) and current (±45 kA) to the CS superconducting magnets to induce current in plasma. The output voltage of the unit CS converter is limited because of the rated voltage of thyristor switches. However, the CS superconducting coil needs more operating output voltage. Thus, the CS converter must be connected in four series to provide the required operating output voltage. Adequate reactive power support is important to improve the electrical grid stability, because four-series connected CS converter generate large amounts of reactive power. However, other reactive compensators require high costs and a large area. Hence, sequence control is applied to the voltage control of the four-series connected CS converter in order to reduce the reactive power. The sequence controller calculates and transmits the individual voltage command to each unit CS converter according to the total voltage command via safe ring-type communication channel. To verify the feasibility of the proposed sequence control of four-series connected CS converter, five real Zynq-based controllers are assembled and a power system, including the four unit CS converters, is implemented in a real-time digital simulator for hardware-in-the-loop verification. Maximum reactive power is reduced by 16%, from 225 to 190 Mvar.

This paper proposes a new design method of LCL filter for a grid connected single phase inverter to improve reduction of switching ripple current. LCL filter is designed from the viewpoint of Total Harmonic Distortion(THD), reactive power and minimization of magnetic materials. The converter side filter inductor L1 depends on a ripple ratio of the switching frequency component of the converter side current to the rated current, which varies by the modulation index and PWM method of the inverter. Filter capacitor limits the amount of reactive power absorbed from the grid. The grid side filter inductor L2 depends on an attenuation ratio of the switching frequency component of the converter side current and the grid side current. The validity of filter design is verified by simulation of current control stability and current ripple characteristics by applying designed LCL filter to the single phase inverter, which can be applied to a railway application.

In this paper, a power conversion system for an energy harvester of electromagnetic generator for a tire application is proposed, and the energy harvesting system is experimentally verified. The power conversion system with the proposed energy capturing method consists of a full-bridge rectifier, a boost converter and a lithium-ion battery, and is designed to weigh less than 3g with a size of 25mm by 14mm due to a restricted installation inside a tire. In this power conversion system, a novel intermittent energy transfer method is proposed so as to charge the battery with minimal losses, which can be occurred in the power conversion. Since power of the energy harvester is very small, performance is analyzed to find the optimal parameters through of simulation and experiment.

In this paper, a novel control algorithm that is more efficient than the conventional method by using a variable step size based on perturb and observe(P&O) is proposed. The conventional P&O method is simple, but there would be highly vibration near the maximum power point. To solve these disadvantages, when the solar irradiance or the temperature change is great, the PV voltage is kept invariable by the constant voltage mode. After that mode is over, it switches to the fast mode to track the maximum power point speedily. In case of operating point is reached at the maximum power point, the duty size is reduced by the variable step mode. As a result, the oscillation of the operating point is reduced and the power generation efficiency is increased. The novel algorithm is verified by simulation and experiment with a PV simulator and PV PCS.

This paper proposed reactive power variation (RPV)strategy using positive feedback (PF)to identify unintentional islanding condition of distributed generations (DGs)and cease the successive operation of them. The concept in this paper uses the relationship between the shift of point of common coupling (PCC)frequency and the injected reactive power after islanding occurrence. The structure of the proposed RPV method has a constant reactive power component and a certain reactive power term due to PF using the frequency deviation between the rated frequency and the measured frequency. When islanding occurs, PF increases continuously the positive or negative reactive power in order to deviate PCC frequency outside over/under threshold levels, where PF reduces significantly detection time. The proposed RPV strategy is verified in experiments through 1.7kW 3-level T-type inverter. Detection times of islanding are measured to 53ms and 150ms in over/under threshold levels, respectively.

In this paper, leakage current generation is analyzed through an equivalent model of the common mode voltage considering a significant parasitic capacitance existing between the photovoltaic array and ground, and combined two methods to reduce the leakage current, which are a new PWM method and leakage current bypass path with a damping network, are proposed. In the first, the leakage current is reduced significantly by a PWM method using two carriers with 180 degrees - phase difference without losing the voltage utilization rate. In the second, the leakage current is further reduced by adding a bypass path from LCL filter to the mid-point of DC link with a damping network. The combined two methods are verified by simulations and experiments with 1μF stray capacitance. The proposed method using the PWM method is verified to reduce the leakage current by 73% compared to the conventional SVPWM, and the bypass circuit with the damping network further reduces the leakage current by 93%, respectively.

In this paper, energy capturing performance of Duty-cycle control and current control method are compared in the photovoltaic (PV) system under the same operating points and verified by simulation and experiment. Maximum power point tracking (MPPT) algorithm is based on the perturb and observe (P&O) algorithm which has advantages of simple implementation, controllable availability and fast response. As a precondition for the performance comparison between two control methods, the same variable width of the PV voltage is used for every cycle of MPPT execution by adjusting the variation of width of Duty-cycle reference to be equalized to the variation width of current reference. To compare and verify the performance of two control methods, simulations and experiments are carried out to compare PV voltage, PV current and captured energy in PV Power Conditioning System(PCS). Under the same operating points of the PV curve, two control methods generate the same magnitude of captured energy during a cycle of MPPT execution. For instance, captured energy for a cycle is 3.26[kWs] for Duty-cycle control and 3.22[kWs] for current control in experimental results.

– In this paper, zero sequence equivalent circuit of Yg-Yg three phase core-type transformer is analyzed. Many problems by iron core structure of the three phase transformer due to asymmetric three phase lines, which includes line disconnection, ground fault, COS OFF, and unbalanced load are reported in the distribution system. To verify a feasibility of zero sequence impedance of Yg-Yg type three phase transformer, fault current generation in the three phase core and shell-type Yg-Yg transformer is compared by PSCAD/EMTDC when single line ground fault is occurred. As a result, shell-type transformer does not affect the flow of fault current, but core-type transformer generate an adverse effect by the zero sequence impedance. The adverse effect is explained by the zero sequence equivalent circuit of core-type transformer and Yg-Yg type three phase core-type transformer supplies a zero sequence fault current to the distribution system.Key Words : Three phase core-type transformer, Zero sequence impedance†Corresponding Author : Department of Electrical Engineering, Chungnam National University, Korea.E-mail: hjcha@cnu.ac.kr* Department of Electrical Engineering, Chungnam National University, Korea.Received : December 16, 2015; Accepted : April 29, 2016

An LCL-type Isolated dc-dc converter operating for constant output voltage is analyzed, including the effect of a high frequency transformer using ac complex circuit approximation. Its solution is derived and is used to obtain the characteristics of the proposed converter. The analyses show through converter modeling, phasor diagram and gain comparison that inclusion of a high frequency transformer results in introduction of magnetizing inductance and leakage inductances at conventional LCL dc-dc converter with ideal transformer. The theoretical and simulation results are presented in case of the wide variations in input voltage and load current in detail. Analysis and simulation results observed that introduction of a transformer in the dc-dc converter had considerable effect on the performance, especially in the case of low output voltage and large load.

The railway vehicle has an auxiliary power unit for supplying power to the associated electronic control devices and passenger service unit. Typically, input voltage from the catenary for rolling stock is highly fluctuating according to the substation capacity, vehicle propulsion and regeneration. Especially, the frost and freezing on contact wire in winter can cause a blackout inside vehicle, and also brings about electronic components damaging and the system down. To prevent this problem, a large filter and capacitor is used. But this is not a perfect solution, because it is increasing weight of the unit. In this paper, a new algorithm is proposed to suppress the disturbance without adding devices. Simulation and experimental results show that the proposed algorithm has performance to suppress the disturbance at the sudden input voltage variations.

In this paper, the characteristics of duty-cycle control method and current control method are compared in PV system under the same maximum power point tracking (MPPT) mode condition and verified by simulation and experiment. MPPT algorithm is based on perturb and observe (P&O) algorithm which has advantages of fast response, simple implementation and controllable availability. As a precondition for the performance comparison between duty-cycle control and current control method, the variable width of the PV voltage is designed to maintain the same value during one cycle of MPPT mode by setting variation of duty reference corresponding to variation of current reference through a series of processes. To verify the performance of two methods, simulations and experiments are performed to compare voltage, current and captured energy in PV PCS system. Under 1.7kW maximum power point condition, two control methods are verified in the same value of instantaneous power and instantaneous power integration through simulation and experiment in one cycle of MPPT. The operating point of two methods is located in the same point. The instantaneous power integration value is 3.26[kWs] for duty-cycle control and 3.22[kWs] for current control in experimental results.