This paper describes a bi-directional isolated dc/dc converter considered as a core circuit of 3.3-kV/6.6-kV high-power-density power conversion systems in the next generation. The dc/dc converter is intended to use power switching devices based on SiC and/or GaN, which will be available on the market in the near future. A 350-V, 10-kW and 20kHz dc/dc converter is designed, constructed and tested in this paper. It consists of two single-phase full-bridge converters with the latest trench-gate Si-IGBTs and a 20-kHz transformer with a nano-crystalline soft-magnetic material core and litz wires. The transformer plays an essential role in achieving galvanic isolation between the two full-bridge converters. The overall efficiency from the dc-input to dc-output terminals is accurately measured to be as high as 97%, excluding gate drive circuit and control circuit losses from the whole loss. Moreover, loss analysis is carried out to estimate effectiveness in using SiC-based power switching devices. The loss analysis clarifies that the use of SiC-based power devices may bring a significant reduction in conducting and switching losses to the dc/dc converter. As a result, the overall efficiency may reach 99% or higher.
Fast responses without any steady-state errors are required for control of a sinusoidal PWM inverter. Its reference voltage waveform is not constant but sinusoidal. The output voltage is controlled to coincide with a reference waveform. However the reference varies to be a further value when the output reaches the value. So there is always a deviation between the reference and the output voltages. It is necessary to design a compensator by considering these sinusoidal variations of the reference waveform to get a good steady-state and fast transient response. This paper proposes a sort of digital repetitive control systems for a sinusoidal PWM inverter control. It is based on the internal model principle to realize a response with no deviation for a periodic sinusoidal reference input. It has a simple numerator and a denominator z2-2z cos ωT + 1 of a transfer function which is equal to the z function of a sinusoidal waveform of the angular frequency ω and the sample time T. Compensator and feedback gains of the inverter are determined by the optimal control principle. The proposed method is investigated for actual performances and it is validated through theoretical and experimental results for linear and nonlinear loads by using a DSP system at the switching frequency of 20kHz.
Almost all methods for sensorless drive of permanent-magnet synchronous motors (PMSMs), which estimate rotor phase using fundamental components of stator voltage and current, cannot operate properly at low speed including standstill. For proper sensorless drive of PMSMs, they must be started and driven up to a certain speed by other methods. This paper proposes a new simple vector control method for a starter, which can start and drive PMSMs from rest up to the 50% rated speed. The proposed vector control method has the following useful features: 1) It is constructed on basis of “MIR strategy" together with feedback current control, and has a new specific stability function preventing rotating oscillation; 2) It can start up PMSMs with less than 50-70% rated load; 3) It can prevent over-current even at very low speed including standstill; 4) It can be applied to both of salient and nonsalient pole PMSMs; 5) It does not require motor parameters of high precision; 6) It is simple and required computational load is very light; 7) It starts up PMSMs with pre-specified maximum current and automatically reduces current amplitude to the necessary minimum; 8) It attains pre-specified ratio of effective/reactive currents at steady state; 9) It can estimate properly rotor phase as well as speed at speed of 10% rating or more. This paper presents the new vector control method in detail from principle and confirms its usefulness by experiments.
This paper presents a new circuit topology of full-bridge soft-switching PWM inverter linked DC-DC power converter composed of conventional full-bridge high frequency PWM inverter with high frequency transformer and an active quasi-resonant snubber consisting of an additional power switching device in series with DC busline and a lossless capacitor in parallel with DC busline. Under this proposed high frequency soft-switching PWM inverter linked DC-DC converter, four power switches in the full-bridge arms and DC busline series switch can achieve ZVS at turn-off commutation. By developing the advanced soft-switching PWM high frequency inverter type DC-DC converter, although the conduction power loss of DC busline series power switch increases a little, the total turn-off switching loss of full-bridge high frequency inverter power modules can be sufficiently lowered more and more in the higher frequency range of 60kHz. As a result, when the switching frequency of high frequency inverter power stage using IGBT power modules is designed so as to be more than about 10kHz, the more the switching frequency of inverter increases, the more this high frequency soft-switching DC-DC converter has remarkable advantage as for the power conversion efficiency as compared with the conventional hard-switching PWM inverter DC-DC converter. Its practical effectiveness of high power density and high performance is actually proved for TIG arc welding equipment in industry.
It is well-known that the separately-excited DC motor has effective torque (current) reduction characteristics in response to rapid increase in the rotational speed of the motor. These characteristics have been utilized in adhesion control of electric railway trains with separately-excited DC motor. Up to now, we have proposed a new skid prevention method for EVs, utilizing these characteristics and have made experiments with the hardware skid simulator “Motor-Generator setup”. In this paper, we applied this skid prevention control to our new vehicle “UOT CADWELL EV" equipped with BLDC motors and showed its effectiveness.
This paper focuses on selection of wind turbine generation systems that include generators, converters and gears. We study three systems: permanent magnet generator (PMG) system, doubly-fed generator (DFG) system and synchronous generator (SYG) system in terms of the system efficiencies and running costs. The system efficiencies and running costs are calculated with considering relation between wind power and wind condition. According to these results, the 1-step gear PMG system is the best choice to large wind turbine system.
This paper focuses on realization of high performance motion control based on acceleration control. Characteristics of acceleration control are investigated and the relationship between the performance and sampling frequency of the system is discussed. It is then described that a higher sampling frequency is required for acquisition of output information than for update of the input value. Based on the consideration, a new multirate sampling method for the acceleration control system is proposed. Disturbance observer for the multirate system is designed by adopting a new definition of disturbance torque. A bilateral system is introduced as a preeminent example of the system that requires robust acceleration control. Both simulations and experiments are performed on the system, and the validity of the proposition is verified by the results.
A linear generator is being developed for on-board power source in the superconducting Maglev. This system generates not only power by using harmonic magnetic field of ground coils but also magnetic damping by controlling zero-phase current of generator coils. We have studied the characteristics of this system mainly applying to vertical damper. This paper proposes the sensor-less combined vertical and lateral damper and reveals its characteristics by numerical examples.
One of fundamental problems in the factory automation is how to obtain linear motion. Linear motors produce directly the linear motion force without a motion-transform mechanism. Linear d.c. motors (LDMs) have excellent performance and controllability. However, the dynamics of small-sized LDMs is adversely affected by the dead-band due to the friction between brushes and commutators. In this paper, it is described that the design of the two-degree-of-freedom positioning system with a LDM using model reference type sliding mode controller (SMC). The proposed positioning system consists of a fixed gain feedforward controller and a SMC used as a feedback controller. The objective of the SMC is to repress the influence of nonlinear characteristics (the dead-band and parameter variations etc.). The tracking performance can be improved as the fixed gain feedforward controller makes a dynamic inverse system in the feedforward path. The effectiveness of the proposed system for improvement of the tracking performance is demonstrated by experimental results.
The objective of this study is to find out a simple, time and cost-effective method for the assessment of state of batteries. In this study, the targets of estimation are NiCd, NiMH batteries that are widely used in portable equipments, but are difficult to estimate exact state. In our previous studies, the mathematical model was successfully derived from short-time measurement of voltage and current of battery using proposed signal processing scheme suitable for batteries characteristics. This paper presents a novel estimation method of state of NiMH battery using parameter fluctuations of internal impedance. In the proposed method, an equation for estimation was obtained as a function of parameters. Using this equation, state-of-charges of NiMH batteries were experimentally estimated with error less than 10%.
This paper discusses an instantaneous calculation method of matrix converter using Venturini control algorithm. This algorithm can stabilize output voltages in the low speed operation because a narrow PWM pulse disappears on the condition of low output voltages, and it can drive by few computational complexities. This algorithm calculates matrix elements, which include trigonometric functions using the angles of input voltages, mathematically. Conventional methods estimated the angles using the zero crossing points of input voltages. The proposal method calculates matrix elements directly using the instantaneous values of input voltages. Especially, this method can calculate trigonometric functions include triple angle accurately and easily, even when a power supply has distortion. A permanent magnet synchronous motor as a load is driven desirable by using the proposal method. In addition, the simulation and experimental results show that the proposal method can stabilize output voltages when any phase input voltage has distortion.
The driving force of automobiles is transmitted by the frictional force between the tires and the road surface. This frictional force is a function of the weight of the car-body and the friction coefficient μ between the tires and the road surface. The friction coefficient μ is also a function of the following parameters: the slip ratio λ determined by the car-body speed and the wheel speed, and the condition of the road surface. Slippage of automobiles which causes much damage often occurs during accelerating and braking. In this paper, we propose a new drive control system which has an effect on acceleration and braking. In the drive control system, a non-linear controller designed by using a Lyapunov function is used. This non-linear controller has two functions: first one is μ control which moves the car-body, another one is λ control. The controller is designed in order that μ and λ work at noslip and with slip respectively. As another controller, a disturbance observer is used for estimating the car-body speed which is difficult to be measured. Then, this lead to the proof of the stability condition of the combined system which consists of two controllers: the non-linear controller and the disturbance observer. Finally, the effectiveness of this control system is proved by a very satisfactory simulation and experimental results for two cases.
This paper describes large-scale magnetic field analysis by means of the hybrid finite element-boundary element (FE-BE) method. The FE-BE method is well-suited for solving open electromagnetic field problems that comprise movement, non-linear media and eddy current. In general, however, large memory and computational costs are required due to the dense blocks in the system matrix generated by the BE part of the hybrid formulation. To overcome the above difficulties, we introduce the fast multipole method (FMM) to the FE-BE formulation developed by ourselves. Furthermore, we propose a novel preconditioning technique suitable for the hybrid FE-BE method with the FMM. Some numerical results that demonstrate the effectiveness of the proposed approach are also presented.
Starting methods for the inverter that drives a permanent magnet machine with only current sensors, while the rotor is rotating, (“free-run startup techniques") are proposed. The proposed methods are based on whether current flows or does not when one switch of the inverter is turned on, which depends on the electrical angle of the emf and the reverse blocking function of the freewheeling diodes. The merit of the proposed methods is that the calculation to determine the rotor position is simple. The variety of the methods is with the types of current detection. Proposed are the methods not only for the phase current detection but also for the DC-bus line current detection that are utilized in low-cost drive systems. Theoretical analysis, design issues, and experimental verifications of the proposed methods are set forth.
In some substations and sectioning posts of Shinkansen power system in JR East area, the electrical transient phenomena when the Shinkansen train passes through the changeover section are reported during the last a few years. In this paper, the closing surge phenomena of the Shinkansen changeover section are elucidated based on both the detail measurement of the surge phenomena at substations and computer simulation. It is clarified that the surge propagation and the reflection at open end of the contact wire system cause the over-voltage at changeover section and result in the discharge of the lightning arresters at some substations.
The used cutting oil contains not only fine metals but also sludge and other compounds which cause the source of odors as time elapses. In the present study, the two types of electrolytic bubble flotation reactors, such as the flow type reactor and the batch type reactor, were investigated to remove these compounds and to regenerate the cutting oil with the aim of developing the maintenance-free and innovative cutting oil treatment process. When the flow type reactor was used, 39% of fine aluminum particles and sludge were removed based on weight basis from the filter catch. When the batch type reactor was used, 75% of these were removed. The cutting oil can be used for regeneration and its characteristic was not deteriorated when the treatment time was less than 40 s at 20V. When the cutting oil regeneration system with the batch type reactor followed by the filter was used, the complete removal of contaminant was obtained.
Recently, QAM (Quadrature Amplitude Modulation) transmission is paid to attention in digital modulation method, and it has been used for wireless LAN and digital broadcasting. QAM is a modulation method that puts information on carrier amplitude and phase. So QAM can achieve high-capacity data transmission that the use efficiency of frequency band is good. But QAM has a bad characteristic that is weak to the noise and the interference because distance between each symbol is short. Then we propose the method that not transmission data but CDMA signal, obtained after transmission data is modulated by spread code and multiplexed, is given to each symbol of QAM. And we studied about railway signal system on track circuit as a sample to apply this method. As result of computer simulations, we verified to be able to achieve the improvement of signal to noise ratio by optimizing the constellation map of QAM. We report with the method of a synchronous acquisition.
This paper describes the high-capacity data transmission system for advanced ATC (Automatic Train Control) System on track circuit. A conventional ATC system is very high safety, but this system has not realized the high-capacity data transmission because of the high return current noise. Therefore, we develop the high-capacity data transmission system using spread spectrum communication in consideration of the characteristic of track circuit. And we verified to be able to take about 2000 bps, with the result that we carried out computer simulations changed the transmission speed about this system. In addition, we studied about the method of a synchronous acquisition. As a result, we established a good method by giving the same data for a synchronous acquisition to some channels. We report this method with result of computer simulations.
This paper deals with the dynamic behavior of a 21-level (line-to-line) BTB system based on series connection of sixteen converter-cells under a single-line-to-ground fault condition. A laboratory system rated at 200V and 20 kW was designed and built up to confirm the validity of the system. Experimental results obtained from the system lead to the following conclusions: i) The BTB system has a capability of reducing both negative-sequence voltage and current appearing during the fault. ii) A dc-capacitor voltage regulator that has been already developed by the authors makes it possible to suppress the dc-voltage fluctuation sufficiently. iii) Flux fluctuations in transformers used for the system strongly depend on a voltage depth during the fault. These would make a significant contribution to putting the BTB system into practical use.
This paper describes an instantaneous harmonic-reactive-power based IPM motor control system without a rotor position sensor and three techniques to improve its drive performance. The first technique is a harmonic-current injection control method using resonant regulators in a current control loop, which prevents the injected harmonic currents from being distorted mainly due to spatial harmonics of the motor. The second one is a method to compensate for coordinate transformation errors of the feedback currents and the voltage commands caused by a position estimation error. This method is effective to suppress a current oscillation as well as speed ripples that depend on the position estimation error. The last technique is devoted to compensation for the disturbance load torque, which utilizes acceleration information in the position estimator. Also, this paper presents performance of position control without any position sensors. In the paper, all techniques were examined with a DSP based fully digital prototype and effective performance improvements were confirmed through the tests.
A high-power-factor electrical ballast for HID lamp requires 6 switching devices. The cost ratio of switching devices and its driving circuits in electrical ballast is not small. Therefore, this paper proposes a new switching method for a ballst which uses 4 switching devices. In experimental results, the proposed method obtains the power factor 0.990 with output power 400[W]. This paper verified that the proposed method has the fine driving performance on a high power HID lamp.