The higher efficiency and the low noise in output voltage and current has been required in some applications, i.e. audio-video equipment, medical equipment and so on. This paper proposes a power amplifier in which a PWM inverter is used as a main circuit. In the proposed power amplifier, a hybrid filter composed of a simple and general LC filter and a small capacity linear amplifier is connected to the output of the inverter. The linear amplifier is inserted in series to the filter capacitor to improve the filtering effect of the LC filter. Switching ripples and LC resonances are considerably suppressed by controlling the amplifier using the proposed method and a low distortion switching power amplifier is realized. The effects of the proposal circuit are verified by simulations and experiments. As a result, the proposed circuit achieves low noise about THD=0.68% in simulation and THD=1.7% in experiment.
At the contact point of current collector of railway, it is difficult to investigate the relation between the wear and the electric current because of many influence factors and the sliding conditions. Therefore we need to show this relation by means of smaller current conditions than railway to see the effects of electric current on friction and wear. Experimental device is pin-on-wire type, which we improve pin-on-disk type sliding system into. It is set in a direct-current circuit and the pin rubs wires under an electric current. At this point, there is not only the mechanical phenomenon by adhesive wear, but also the electric phenomena including fusion caused by the heat from constriction resistance and the arc discharge caused by the contact loss during sliding test. To examine the effects on the wear of contact materials by them, we investigate the details classifying contact loss, which has a great influence on the wear of contact materials, to two kinds of contact loss, arc contact loss and complete contact loss. As a result of it, we can show the relations between the power supply voltage and the outbreak number, the outbreak ratio of two types of contact loss, and relations between the wear and the power supply voltage, materials by means of the former two relations. And we can see the effects of fusion and arc discharge on the sliding surface by optical microscope.
A winding-current interference model of self-sensing active magnetic bearings (AMBs) and an improvement technique of position-sensing characteristics are proposed. In the self-sensing AMB systems, the electromagnets work not only as actuators suspending the rotor but also as position sensors. The self-sensing position signal, however, includes errors because of nonlinearity of the magnetic circuits. In the proposed model, the current is directly transmitted to the self-sensing position signal. This transmission means degradation of the self-sensing characteristics. This winding-current interference to the self-sensing signal is reduced by the proposed compensation method. The characteristics of the proposed self-sensing method agree well with a reference sensor signal up to over 1kHz.
With the realization of the wide range of voltage ratings up to 6.5-kV, IGBTs have been widely applied to various inverters as a switching device due to their features of fast switching, low switching losses and easy driving. Accordingly, a short-circuit protection for IGBTs is one of the most critical issues. This paper describes two short-circuit protection methods, we call them, “Active-controlled cut-off method" and “Passive-controlled cut-off method". The active-controlled cut-off method enables a device to cut off a short-circuit current safely by control of a gate-to-emitter voltage waveform following a memorized pattern in consideration of its short-circuit safe operating area (SCSOA) and maximum short-circuit withstand time. A short-circuit test with a 2.5-kV rated IGBT are conducted, and effectiveness of the active-controlled cut-off method is verified. For more simple protection circuit configuration suitable for inverters with lower cost and smaller size, the passive-controlled cut-off method is given. This circuit is composed only of a stray inductance in main circuit line. Experimental results with simulation and experimental results are shown here.
This paper presents novel tuning formulas for two-degree of freedom PID-family controllers. The structure of the proposed control system consists of a plant or process to be controlled, P or PI or PID controllers and a pre-filter. The tuning formulas are derived based on the CDM design method by solving CDM algebraic equations obtained from the known controllers' structures and the approximated plant model based on the experimental test. The proposed formulas are summarized in the tuning table and can be used easily. To apply the tuning rules, the plant or process is firstly tested to find the values of critical gain and critical period experimentally in the same manner as Ziegler-Nichols' second method. The parameters of P, PI or PID controller and the associated pre-filter are then obtained by substituting those values in the formulas. The experiments in controlling the level process with long pipeline controlled by the controllers tuned by the proposed rules and the well-known Ziegler-Nichols' formulas are conducted and compared. The proposed control systems result in responses with shorter settling time and considerably smaller or no overshoot. Overall, the proposed formulas are well applicable despite their simplicity.
This paper discusses a 6.6-kV transformerless STATCOM intended for installation on industrial and utility distribution systems in the near future. In addition, this paper provides experimental results obtained from a laboratory model rated at 200V and 10kVA. The authors propose such a control method as to superimpose a 6th-harmonic zero-sequence component on each of three-phase voltage references. This helps to stabilize the voltage of the inner midpoint in the dc link. As a result, the laboratory model installs two bi-directional buck-boost choppers on the dc link for the porpose of stabilizing the voltages of two outer midpoints. Experimental results obtained from the laboratory model verify the validity of the system design, giving promise of the viability of the 6.6-kV transformerless STATCOM.
The paper presents a power generation system based on the cogeneration of heat and electricity with a novel high speed turbogenerator. The machine consists of a single stage steam turbine and a directly coupled permanent magnet synchronous generator in one constructional unit. A PWM IGBT rectifier is the load to the generator and a PWM IGBT three-phase four-wire inverter feeds the power into the low voltage mains. In order to increase the turbine efficiency at light load, variable speed operation of the turbogenerator is realized. Different control schemes for mains parallel operation and stand alone operation are presented. The control schemes allow for the use of a lookup table based control with a speed-power-characteristic or for the use of a maximum power point tracker. Measurement results from the successfully tested turbogenerator set are presented.
In recent days, minimally invasive surgery by using endoscope has spread and robots that support endoscopic surgery are being focused on. However, existing surgery robots could not transmit information of tactile sensation to operators. To improve the safety of robotic surgery, it is required to transmit the tactile sensation to the operator. To attain that need, a teleoperation method called bilateral has developed and we have achieved to transmit keen tactile sensation with one degree-of-freedom forceps robots. In this paper, multi degrees-of-freedom bilateral forceps robot system will be developed to make surgery robot adaptable to actual situation of the surgery. In bilateral teleoperation of multi degrees-of-freedom robots, environmental information of master and slave should be exchanged in common coordinates. In this paper, mode coordinates is used as common coordinates. Mode coordinates are closely related to the motion of the robot. Method to decouple the mechanical interference on mode coordinate is proposed. The paper also proposes a controller implemented on FPGA to achieve short processing time and hard-realtime control. As a result, multi-degrees-of freedom force feedback is attained. The experimental results show the viability of the proposed method.
Sensor fusion extends robot ability to perform more complex tasks. An interesting application in such an issue is pushing operation, in which through multi-sensor, the robot moves an object by pushing it. Generally, a pushing operation consists of “approaching, touching, and pushing"(1). However, most researches in this field are dealing with how the pushed object follows the predefined trajectory. In which, the implication as the robot body or the tool-tip hits an object is neglected. Obviously on collision, the robot momentum may crash sensor, robot's surface or even the object. For that reason, this paper proposes a soft pushing operation with dual compliance controllers. Mainly, a compliance control is a control system with trajectory compensation so that the external force may be followed. In this paper, the first compliance controller is driven by estimated external force based on reaction torque observer(2), which compensates contact sensation. The other one compensates non-contact sensation. Obviously, a contact sensation, acquired from force sensor either reaction torque observer of an object, is measurable once the robot touched the object. Therefore, a non-contact sensation is introduced before touching an object, which is realized with visual sensor in this paper. Here, instead of using visual information as command reference, the visual information such as depth, is treated as virtual force for the second compliance controller. Thus, having contact and non-contact sensation, the robot will be compliant with wider sensation. This paper considers a heavy mobile manipulator and a heavy object, which have significant momentum on touching stage. A chopstick is attached on the object side to show the effectiveness of the proposed method. Here, both compliance controllers adjust the mobile manipulator command reference to provide soft pushing operation. Finally, the experimental result shows the validity of the proposed method.
In this paper, we investigate the methods of eddy current analysis for stator core-ends of turbine generators from both results of experiment and finite element analysis. First, the 3-D finite element analysis considering the lamination of the core is applied to the simple model that simulates the end part of the turbine generator. The calculated losses and flux distributions caused by the perpendicular flux are compared with the measurement to verify the validity of the calculation method. The measured and calculated results agree well. Next, the calculation method is applied to the 250MVA class turbine generator with the considerations of the laminated stator core. The field and loss distributions at stator core-ends of the turbine generators are clarified.
The authors present a new pre-compensator used for control of a stepping motor in the oscillation-damping type full-step drive system. The compensator is a low-pass filter, in which the cut-off frequency varies with reference speed variation, and is used for adjustment of reference angle waveform. The effect of this system for rotor oscillation damping is verified by the experiment.
This paper proposes a novel detection method of active and reactive currents in single-phase circuits. This method uses only two multipliers and LPFs for detecting active and reactive currents with a PLL circuit. In the proposed method the fundamental active-reactive component in single-phase circuits can be detected by LPFs in the time domain like the instantaneous active-reactive power theory in three-phase circuits. The basic principle of the proposed detection method is discussed in detail, and then confirmed by digital computer simulation. Digital computer simulation results demonstrate that the proposed method can detect the active-reactive current in single-phase circuits in the time domain.
For a self tuning strategy of a temperature controller, an optimization method of the Laguerre time scaling factor in consideration of the model error is presented in this paper. The relationship between a thermal model and the time scaling factor is described. The thermal resistance and capacity can be identified.
The authors previously proposed a new maximum power point tracking strategy based on limit cycle phenomena (LCMPPT) for a grid connected PV system with dc linked two power conversion stages. This paper enhances its principle for the systems with single power conversion stage such like a battery charging system. Since, it needs only one current sensor, it is cost effective being easy to implement by using a low cost PIC microprocessors. Experimental results in proto-type system have been shown to prove its validity and usefulness.