In an electrolytic-capacitor-less single-phase to three-phase inverter, the resonance between the line inductance on the source side and the DC-link capacitor causes an input current ripple. In this paper, a new approach for suppressing the input current ripple due to this resonance is described. This paper proposes a new control method that does not emulate the action of the damping resistor for passive damping. The proposed method cancels the resonance by using a DC-link capacitor discharge actively. In the proposed method, a cancellation voltage pulse that cancels the resonance is added to one of the three-phase voltage references for the inverter. The cancellation voltage is calculated by feeding back the variables relative to the discharge. The effectiveness of the proposed method is validated by experimental results.

In this paper, a maximum torque per ampere (MTPA) control strategy for direct-torque-controlled interior permanent magnet synchronous motor (IPMSM) drives is proposed. For current-controlled IPMSM drives, many methods have been proposed, in which the periodic signals are injected into reference current vectors. On the other hand, the direct-torque-controlled IPMSM drives require the torque and stator flux linkage magnitude as their references. The proposed method injects signals into the reference flux in a manner similar to that used for the signal injections in current-controlled IPMSM drives. The main feature of the proposed method is the ability to search the optimal current directly and successively using a local search algorithm. Therefore, variations in the parameters can be considered. The simulation and experimental results validate the proposed method.

To operate motors at a higher efficiency over a wider range of speeds, we have developed a technique that quickly reverses the salient poles in permanent magnet (PM) motors. With this technique, PM motors can change negative salient poles into positive salient poles by applying the magnetization of the PM. Furthermore, we discuss the combination of reversible salient poles with variable magnetic force to increase the efficiency of the motor under partial load. This paper proposes a novel PM motor that uses our proposed technique to reverse saliency and describes the principle of reversible pole saliency as well as the resulting motor characteristics. When the proposed motor operates with a negative salient pole, a higher torque is produced at low speeds; when the motor reverses to a positive salient pole, higher power and efficiency are produced at high speeds. These results confirm that the proposed motor can reverse the saliency direction of the pole and that this reversibility allows the motor to achieve high torque at low speeds and high power at high speeds. Consequently, the motor can operate over a wide range of speeds.

A double-winding permanent magnet synchronous motor (PMSM), which has two groups of three-phase windings on its one stator, has a coupling effect between the windings. This paper explains the deterioration of the current control performance of a double-winding PMSM by using a voltage equation of the motor. The proposed coupling cancellation control method improves as demonstrated by the results of simulations and experiments. A design method for the coupling cancelling control with a Bode diagram is also discussed.

This paper proposes a position sensorless control system for interior permanent magnet synchronous motors (IPMSMs) within the inverter over-modulation range. In general, using the position sensorless control method in the middle- and high-speed regions requires q-axis inductance, which varies significantly owing to magnetic saturation. Therefore, parameter mismatch may occur, which decreases the motor efficiency and causes control system instability. Furthermore, in the inverter over-modulation range, the performance of the auto current regulator deteriorates significantly because disturbance components, which consist of low-order harmonic current components, are fed back to the controller. As a result, the operation range is not appropriately expanded if no measurements are obtained for these components. First, the robustness against magnetic saturation is increased by constructing a position sensorless control system based on the L_d model. Next, a comb-filter-based harmonic current countermeasure method is proposed. The position sensorless control system can expand the operation range in this method. Finally, the effectiveness of the proposed control system is confirmed by several experiments.

This paper proposes a novel control strategy for a step-down rectifier that applies a modular multilevel converter (MMC). This paper first discusses the circuit configuration and the proposed control method. The main part of the control system is based on the control method for the single-phase power factor correction converter (PFC converter), which means that the conventional control techniques of the single-phase PFC converter can be simply applied to the MMC. The control elements which achieve step-down rectification and capacitor voltage balancing are added to the main part. Besides, in capacitor voltage balancing of the proposed control method, it is not necessary to design the control parameters. In addition, the operation of the proposed step-down rectifier is confirmed by the simulation. The simulation results show that the proposed converter achieves step-down rectification from the grid voltage of 6.6kV to the DC voltage of 400V. Moreover, the step-down rectifier maintains all capacitor voltages constant. Finally, a fundamental operation is confirmed by experiments using a miniature model. As the results, the step-down rectifier converts from the input grid voltage of 200V to the output DC voltage of 75V. Besides, the proposed step-down rectifier maintains the capacitor voltage of each cell to the voltage command of 130V. The maximum error between the voltage command of the cell capacitor and the measured voltage is less than 2%.

This paper proposes a new current reconstruction method that uses a DC-link current sensor for an inverter. In a typical current reconstruction method, an error of the average motor current is likely to occur because there are only two reconstruction points in one carrier period. In addition, the phase current harmonics increases because the carrier wave and voltage references are changed in a region where current reconstruction is difficult. This paper proposes three methods for current reconstruction. The first method calculates the average motor current more accurately by oversampling. The oversampling is synchronized to pulse-width modulation (PWM) switching and is conducted by combining the conventional method and the second method. The second method realizes the difference detection by using the resistance-capacitance (RC) circuit. The third method calculates the current in the region where current reconstruction is difficult by calculating the phase-shifted voltage references. The current calculation method does not increase the current harmonics in this region because it does not change the carrier wave and voltage references. The effectiveness of these methods and the motor control with the reconstructed current were confirmed in experiments.

An individual's feeling of bone quality during a drilling procedure is important for predicting the prognosis of a dental implant. At present, the estimation of an individual's feeling is subjective, and a new, definitive, data-based classification scheme is needed to identify bone quality objectively. A tele-robotic drilling system was developed to solve this problem. This paper presents a model for predicting vibration during bone drilling. The forces from the drill are modeled by considering the indentation process that occurs under different form computed tomography (CT) numbers. This method can be applied to a cutting-force presentation method using a haptic drilling system. An experiment was conducted to confirm the validity of the proposed method. The result shows that the experimental simulation achieved the same force response as the real cutting. The mean of the error between the real cutting force and the predicted cutting force was 3.5%. The system allows dental students to learn and practice procedures such as cutting the jaw bone. The advantages of this system are that it has a high force-output quality and a fine position-sensing ability. Therefore, dental students can simulate a realistic force response from the bone if the CT number has been acquired. This system can also be used for real implantation surgery because it allows doctors to experience the force response before operating.