11:20AM Bidirectional Series-Resonant DC-DC Converter with Fault-Tolerance Capability for Smart Transformer [#801] Levy Costa, Giampaolo Buticchi and Marco Liserre, University of Kiel, Germany
The Series-Resonant dc-dc converter (SRC) is widely used in several application and it became very popular in Smart Transformer application. In this application, fault tolerance is a highly desired feature and it is obtained through redundancy. This paper proposes a reconfiguration scheme for the SRC for the case of failure in one semiconductor, which could drastically reduce the need of redundancy. Using the proposed scheme, the full-bridge based SRC can be reconfigured in a half-bridge topology, in order to keep the converter operational even with the failure (open circuit or short circuit) of one switch. The theoretical analysis is carried out for the unidirectional SRC and then extended to the bidirectional topology, since bidirectionality is required in smart transformer application. To verify the feasibility of the proposed scheme, the converter is tested experimentally in a 700 V to 600 V prototype with 10 kW of output power. A IGBT short-circuit fault is tested and the results confirms the effectiveness of the proposed approach.
11:45AM Analysis and Design of Planar Inductor and Transformer for Resonant Converter [#432] Yueshi Guan, Na Qi, Yijie Wang, Xiangjun Zhang, Dianguo Xu and Wei Wang, Harbin Institute of Technology, China
With the increasing demand of low-profile, high-power density for power electronics system, the planar magnetics (inductors and transformers) begin to be adopted in many fields. The copper tracks on printed circuit board (PCB) are mostly adopted as the windings of the planar magnetics which cannot be changed flexibly once the PCB is produced. So the design method of planar magnetics is an important issue. In this paper, the Response Surface Method (RSM) is adopted to design the planar inductor. The method can exactly obtain the inductance and resistance by the formulas obtained from experimental or simulation test results which can simplifying the design process. For the planar transformer, the Modular Layer Model (MLM) is adopted to build the final equivalent model of the transformer, by which the magnetizing inductance, leakage inductance and resistance can be calculated. The proposed design methods of planar inductor and planar transformer are verified by a resonant converter prototype.
Modeling and Control of Grid Connected Converter II
10:30AM Combined DC Voltage Control Scheme for Three-port Energy Router Based on Instantaneous Energy Balance [#460] Gaohui Feng, Zhengming Zhao, Liqiang Yuan and Kai Li, Tsinghua University, China
DC voltage control is crucial for the multi-stage converter, because it contributes to reduce the DC capacitance and improve the reliability of the whole system. A combined control scheme, including two parts, is proposed in this paper for controlling the two DC bus voltages respectively in the Energy Router. Firstly, the energy feed-forward control scheme is used to control the low voltage DC bus, which connects the dual active bridge, the inverter and the bidirectional DC/DC converter. Compared with the conventional feed-forward control scheme, the energy feed-forward control scheme considers the instantaneous energy changes of the inductors in the inverter and the DC/DC converter together. Secondly, for the high voltage DC bus, which is connected to the single-phase PWM rectifier, a new control scheme is proposed based on the energy balance relationship among the source, the load and the passive elements (the inductors and the capacitors). The proposed combined control scheme can improve obviously the transient performances of the two DC bus voltages. Simulation and experimental results have confirmed their superior performances.
10:55AM Grid-Voltage Sensorless Control of a Converter Under Unbalanced Conditions: On the Design of a State Observer [#180] Jarno Kukkola and Marko Hinkkanen, Aalto University, Finland
This paper deals with grid-voltage sensorless synchronization and control under unbalanced grid conditions. A three-phase grid-connected converter equipped with an LCL filter is considered, and no other signals than the converter currents and the DC-link voltage are measured for control. An augmented adaptive state observer is proposed for estimation of the positive- and negative-sequence components of the grid voltage. The proposed observer is tested as a part of a sensorless control system. Experimental results show that the proposed method works well even in highly unbalanced grid conditions.
11:20AM Current-Mode Boundary Controller with Reduced Number of Current Sensors for a Three-Phase Inverter [#527] He Yuanbin, Chung Shu-hung, Ho Ngai-man and Wu Weimin, City University of Hong Kong, Hong Kong; University of Manitoba, Canada; Shanghai Maritime University, China
Cascaded boundary-deadbeat controller has been proven to be effective in controlling single-phase grid-connected inverter with LCL output filter. Such architecture mitigates filter resonance and offers good stability under stiff and weak grid conditions. However, its merits are offset by requiring many sensors and high-precision intra-cycle information of the circuit variables to dictate the states of the switches. Moreover, the system will also be in variable switching frequency operation, due to coupling circuit variables amongst phases, in controlling three-phase three-wire inverter. This paper presents a boundary controller that utilizes second-order switching surface to track directly the output current of the three-phase three-wire grid-connected inverter with an LCL filter. By applying the 60-degree discontinuous pulsewidth modulation scheme for a fictitious decoupled dual-buck structure in each operation sector, two separate sets of switching criteria with reduced number of current sensors, fixed frequency operation and recovered intra-cycle information of circuit variables for dictating the states of the switches of two half-bridge legs are formulated. A 3kW, 127V/50Hz prototype is built to validate the effectiveness of the proposed control method.
11:45AM Positive- and Negative-Sequence Current Controller for Grid-Tied Converters With LCL Filters [#575] Diego Perez-Estevez, Jesus Doval-Gandoy, Alejandro Yepes, Oscar Lopez and Fernando Baneira, University of Vigo, Spain
Traditionally, the current control of grid-tied converters with LCL filter is based on proportional-resonant or proportional-integral controllers, which often need an additional active damping method to achieve stability. These solutions do not permit to place the closed-loop poles in convenient locations when dealing with such high-order plants. This constraint results in degraded reference-tracking and disturbance-rejection responses. On the other hand, the existing methods based on direct pole placement or other modern control strategies, do not control with zero steady-state error both positive and negative sequences of the grid current, but only the positive one. This limitation is undesirable under unbalanced grid conditions. This paper presents a current controller for grid-tied converters with LCL filters based on direct discrete-time pole placement. The proposed controller makes it possible to control both positive and negative sequences of the grid-side current with zero steady-state error. Contrarily to the classical resonant controllers, the closed-loop poles can be placed in convenient locations, yielding a fast response with negligible overshoot and low controller effort. Moreover, no additional damping methods of the resonance are necessary to achieve stable operation, regardless of the switching frequency and LCL filter used. Simulation and experimental results that validate the proposal are presented.
10:30AM Realization of Quadrature Signal Generator Using Accurate Magnitude Integrator [#1067] Zhen Xin, Changwoo Yoon, Rende Zhao, Poh Chiang Loh and Frede Blaabjerg, Aalborg University, Denmark; China University of Petroleum (Hua dong), China
Second-Order Generalized Integrator based Quadrature Signal Generator (SOGI-QSG) has been widely used in single- or three-phase power converter systems due to its simplicity and flexibility. However, its dynamic response is not only decided by its damping gain but also influenced by the input-signal parameters, especially when a fast response is required for usages such as grid synchronization. As a result, the parameter design of the SOGI-QSG becomes complicated. Theoretical analysis shows that it is caused by the inaccurate magnitude-integration characteristic of the SOGI-QSG. To solve this problem, an Accurate-Magnitude-Integrator based QSG (AMI- QSG) is proposed. The AMI has an accurate magnitude-integration characteristic for the sinusoidal signal, which makes the AMI-QSG possess an accurate First-Order- System (FOS) characteristic in terms of magnitude than the SOGI-QSG. The parameter design process of the AMI-QSG can thus be as simple as the typical FOS. Besides, the structure of the AMI-QSG is further configured to be able to extract the dc component and harmonic components. The effectiveness of the proposed structures and the correctness of the theoretical analysis are evaluated by experimental results.
10:55AM A New Instantaneous Point on Wave Voltage Sag Detection Algorithm and Validation [#243] Yujia Cui, Ahmed Sayed-Ahmed, Prathamesh Vadhavkar, Brian Seibel and Russel Kerkman, Rockwell Automation, United States
In industrial applications, power quality has been an issue drawing increasing concerns due to its severe consequences on system performance and downtime cost. Voltage sags are classified as one of the most common power quality issues. In order to guarantee system operation under several line sag scenarios, international standards such as SEMI F47, IEC-61000-4-34, and IEC- 61000-4-11 have been established as guidelines for electrical/electronics manufacturers. In this paper, an innovative point on wave sag detection is introduced. Although the main focus of this work is centered on applications related to regenerative motor-drive systems, this approach can be utilized in a myriad of other applications such as grid-tie inverters, uninterrupted power supplies and advanced relay protection. In addition, the introduced technique is very effective at detecting repeated line sag conditions. The introduced detection method has been experimentally validated using a 20 HP regenerative motor-drive system setup under various line sag scenarios.
11:20AM Voltage Quality Enhancement with Minimum Power Injection [#1203] Darlan Fernandes, Fabiano Costa, Joao Martins, Alberto Lock, Edison da Silva and Montie Vitorino, Federal University of Paraiba, Brazil; Federal University of Bahia, Brazil; Federal University of Campina Grande, Brazil
This work proposes the usage of a repetitive-based control to dynamically restore the voltage applied to sensitive and critical loads of power systems. Besides, sag and voltage swells, the proposed control can intrinsically mitigate harmonic distortions. Furthermore, the filter is able to work out on sinusoid references and, thus, avoids the need of employing the dq transform. A recursive least-squares is also included to the control system in order to assure the synchronization of the injected voltages to be restored. The design of the control parameters along with the system stability are discussed throughout the paper. Also, this work aims to analyze two different types of voltage insertion into the grid through the series compensator. In the first one, the voltage injection is in synchronization with the pre-faulted grid voltage, while the second injects the voltage in a manner that the compensator spends a minimum of active power. Both types produce reference voltages that are used by the repetitive control. Additionally, simulated and experimental results are presented and corroborate the method efficacy.
11:45AM A Universal Variable On-time Compensation to improve THD of High-frequency CRM Boost PFC Converter [#896] Zhehui Guo, Xiaoyong Ren, Handong Gui, Yu Wu, Zhiliang Zhang and Qianhong Chen, Nanjing Univ. of Aeronautics and Astronautics, China; The University of Tennessee, United States
The critical conduction mode Boost PFC converter utilizes the resonance to achieve soft-switching, however, leading to high THD and poor power quality, especially at high switching frequency applications. This paper presents a systematic operation analysis of the CRM Boost PFC converter considering the parasitic parameters and investigates the negative influence of resonance process to THD. In order to improve input current THD, a variable on-time control method is proposed to keep the average inductor current following the intended sinusoidal input current. The proposed implementation combines the advantages of digital control and analog control, effectively decreasing the total cost of the system. And the proposed variable on-time compensation method is verified with the experimental results well.
Stability in Power Converters I
Thursday, September 22, 10:30AM-12:10PM, Room: 102E, Chair: Jian Sun, Andrea Formentini
10:30AM On Impedance Modeling of Single-Phase Voltage Source Converters [#1228] Shahil Shah and Leila Parsa, Rensselaer Polytechnic Institute, United States
This paper explores modeling of ac-side impedance for grid-connected single- phase voltage source converters (VSC). It shows that the conventional impedance, in the form of one-dimensional transfer function, may not completely capture the linearized dynamics of the converter. Resonance analysis using such impedance may fail for oscillations involving multiple frequency components. This problem is addressed by extending the impedance concept and defining a two-dimensional admittance for single-phase VSC. This admittance, in the form of a two-by-two transfer matrix, accurately captures the frequency cross-coupling effect that plays a critical role in low- frequency oscillations. A new form of signal-flow graphs is also introduced to represent the linearized dynamics of the VSC. These graphs visually demonstrate: a) the flow of perturbations through converter, b) why the conventional impedance fails in completely capturing the converter dynamics, c) how the two-dimensional model better captures them, and d) how to extend the concept further to higher dimensions without repeating the modeling process. Proposed two-dimensional and the conventional one-dimensional impedance models are developed for a single-phase VSC and they are validated using detailed circuit simulations.
10:55AM Design Consideration of Volt-VAR Controllers in Distribution Systems with Multiple PV Inverters [#1617] Mahsa Ghapandar Kashani, Yonghwan Cho and Subhashish Bhattacharya, North Carolina State University, United States
Advanced control techniques such as Volt-VAR Control (VVC) are required for integration of multiple distributed renewable energy, such as Photovoltaic (PV) resources, on an electric distribution system. However, undesired interactions have been observed among these Volt-VAR controlled PV inverters which leads to oscillation and instability of the system. In this paper, an analytical approach to study the stability of local voltage control in high PV penetrated distribution systems with advanced Volt-VAR control functions is employed. The transient of inverter Volt-VAR Control interactions and dynamics of the interconnected feed-back loops in the distribution circuits are investigated. It is shown analytically that the Grid impedances, droop slope, PI controller parameters, response time and delay time in the VVC are the main factors affecting the dynamic response of the system, and the absence of a standard selection criteria for inverter and controller parameters under different Grid impedances results in undesired potential interactions among the PV inverters and distribution power system.
11:20AM Extended Stable Boundary of LCL-Filtered Grid-Connected Inverter Based on Grid-Voltage Feedforward Control [#1673] Minghui Lu, Zhen Xin, Xiongfei Wang, Remus Beres and Frede Blaabjerg, Aalborg University, Denmark, Denmark
For the LCL-filtered grid-connected inverter, it has been reported that the digital time delays will narrow the stable region of current control loop when the inverter- side current is used for implementing the feedback control. A sufficient stable condition is that the filter resonance frequency should be designed under one-sixth of sampling frequency. However, the low resonance frequency leads to a comparatively large filter inductance or/and capacitance. To extend the stable boundary to the region above fs/6, this paper proposes a novel voltage feedforward scheme for the LCL- filtered inverter. Theoretical analysis is then provided to validate its feasibility and stability. Compared to other widely used active damping strategies, no extra sensors are needed because the filter capacitor voltage, which is used for voltage feedforward control, is also sampled for phase-locked loop in this paper. Simulations and experimental results are provided for verifying the theoretical analyses.
11:45AM Allowable Bus Impedance Region for MVDC Distribution Systems and Stabilizing Controller Design Using Positive Feed-Forward Control [#1266] Jonathan Siegers, Silvia Arrua and Enrico Santi, University of South Carolina, United States
Advances in switching power electronic converter technology have brought about a resurgence of interest in the use of DC power distribution systems for a variety of applications. However, the notional power electronic based DC distribution system is a complex and extensively interconnected system consisting of multiple power converters. A number of system-level challenges related to stability arise due to interaction among multiple converter subsystems. The recently proposed Passivity Based Stability Criterion (PBSC) coupled with active damping by a Positive Feed-Forward (PFF) control technique has successfully been demonstrated to provide meaningful stability analysis and stabilizing controller design via the insertion of active damping impedances. However, the PBSC provides no explicit indication of the relative stability and damping of the system under study. This paper establishes an Allowable Impedance Region concept in the s-plane to relate the magnitude of the system bus impedance Nyquist contour to the expected dynamic performance of the system. This analysis technique provides information such that an appropriate damping impedance for insertion into the system bus via PFF control may be easily identified, thus ensuring the passivity of the system and guaranteeing stability and desired dynamic performance.
Non-Conventional Machine Configurations I
Thursday, September 22, 10:30AM-12:10PM, Room: 101A, Chair: Robert D. Lorenz, Ronghai Qu
10:30AM A Novel Stator-Consequent-Pole Memory Machine [#1030] Yang Hui, Lin Heyun, Zhu Z. Q., Fang Shuhua and Huang Yunkai, Southeast University, China; University of Sheffield, United Kingdom
This paper proposes a novel stator-consequent-pole memory machine (SCPMM), in which the AlNiCo permanent magnets (PMs) with low-coercive-force (LCF) are alternately placed between the adjacent stator teeth. This new machine has the merits of simplified online PM magnetization, robust rotor and easy thermal management. Meanwhile, the energy-efficient flux regulation can be achieved since the LCF magnets can be repetitively magnetized or demagnetized with negligible excitation loss. The parallel magnetic circuit topology between PM and armature reaction fields permits the LCF PMs to well resist the irreversible demagnetization risk. In addition, the SCPMM benefits from the ease of manufacture since the stator and rotor assemblies are similar to the switched reluctance machines. Afterwards, the machine configuration and operating principle are introduced, respectively. The available slot/pole combinations are analyzed. The electromagnetic performances of the SCPMMs having various slot/pole combinations are investigated and compared. A prototype is manufactured and tested to experimentally validate the finite-element (FE) analysis.
10:55AM A Novel Variable Flux Memory Machine with Series Hybrid Magnets [#239] Hao Hua, Z.Q. Zhu, Adam Pride, Rajesh Deodhar and Toshinori Sasaki, University of Sheffield, United Kingdom; IMRA Europe SAS, UK Research Centre, United Kingdom
This paper proposes a novel variable flux memory (VFM) machine, in which the constant permanent magnet (PM) (CPM) with high coercive force and the variable PM (VPM) with low coercive force are alternatively located in the interior-PM (IPM) rotor. Thus, the VPMs and CPMs are magnetically connected in series, with which the CPMs can assist the VPMs to withstand the unintentional demagnetization caused by armature reaction. Therefore, a high armature current can be applied to the machine. Meanwhile, the reluctance torque is retrieved. Thus, a high torque density can be obtained. The demagnetization and re-magnetization characteristics are investigated, in which the working points of VPMs are illustrated. Furthermore, the advantages of improved efficiency of the proposed VFM machine are demonstrated. A prototype machine is manufactured and tested to validate the predictions.
11:20AM On the Feasibility of Carbon Nanotube Windings for Electrical Machines - Case Study for a Coreless Axial Flux Motor [#1679] Vandana Rallabandi, Narges Taran, Dan M. Ionel and John F. Eastham, University of Kentucky, United States; University of Bath, United Kingdom
The latest developments in carbon nanotube (CNT) wires and yarns attract great interest for potential application to electromagnetic devices, such as electrical machines and transformers. The CNT material properties are largely different from copper and aluminum in terms of electrical conductivity, mass density, and thermal transfer, creating a new design paradigm for which the traditional rules and device topologies no longer apply. This paper proposes a brushless permanent magnet multidisc axial flux construction with coreless stator and special windings and minimal rotor back iron, as a suitable topology for CNT winding application. Specific analytical closed-form sizing equations, as a function of winding electric conductivity, machine dimensions, and operating speed/frequency, are derived and employed in a systematic comparative study over a range of kW power ratings and speeds between 1,000 and 10,000 rpm. The numerical study is complemented by 3D and 2D electromagnetic FEA. The results show that the designs with CNT windings may have substantially higher specific power per mass, particularly at high rotational speeds and/or supply frequency, where the combined effect of DC and AC conduction losses in the windings is significant.
11:45AM A Novel Simplified Structure for Single-Drive Bearingless Motor [#827] Hiroya Sugimoto, Itsuki Shimura and Akira Chiba, Tokyo Institute of Technology, Japan
This paper presents a novel design of simplified structure in one-axis actively positioned bearingless motor. To enhance torque density, while having extreme high passive stiffness, simple and compact structures are designed. Basic principle of active axial force generation and the calculated results in 3D-FEM analysis are presented. The axial force, torque and passive stiffness are compared in five structures. In addition, a minimization method of an interference torque and force is presented.