Koseki-Ohnishi Laboratory

WPT for electric vehicles (EVs) is taken as an example of an application of the proposed method. In recent years, WPT for EV has been studied all over the world to extend its cruising range and improve convenience. Analyses done in previous researches are useful and straightforward for evaluation of the specific combination of a transmitter (TX) system and a receiver (RX) system in such cases as the commercial system compliant with a well-defined standard or the industrial system originally designed as a pair system.

Meanwhile, the combination of TX and RX is not always identical to what designers expect in such cases that a user charges the user's EV with a public wireless charging facility in a parking spot. The unexpected combination of the TX and RX is not able to guarantee to charge the battery sufficiently. Moreover, in practice, the resonance condition of WPT is not always satisfied due to variation in the values of the elements of the compensation circuit. For these reasons, a practical compensation circuit is often designed as the non-resonant system. Since such a non-resonant compensating circuit requires a complicated analysis, its power transmission characteristics have not been sufficiently studied yet. As a result, the design of a non-resonant compensation circuit is complicated and time-consuming.

In this research, we propose a visualized method that can intuitively grasp and evaluate the power transmission characteristics of various combinations of compensating circuits by dividing the circuits into the TX side and RX side. In addition, we aim to apply this method to the compensation circuit design of dynamic wireless power transfer for EV by considering the transient phenomena of the circuit.

The wireless power transfer system for electrical vehicle can build a strong magnetic field in medium distance space. This AC magnetic field will heat any metal dropped into this area which is a safety problem in using WPT system. The traditional methods can just detect some big metal and can be affected by the receiving coil misalignment and sometimes have dead zone which limits their application.

To solve these problems, I am trying to use new methods to design the search coil and improve the detection method to find the metal. Now I have succeeded in distinguishing the metal object and receiving coil misalignment. Besides, I am also working to improve the robustness of the detecting system In order to promote the practical use of this technology.

Magnetic levitation is a technology that enables non-contact support of objects by magnetic force, and this is useful for realizing high-speed movement and establishing maintenance-free systems because of characteristics of zero friction or wear. Our research group proposed a new configuration that combines a transverse flux type permanent magnet linear synchronous motor and an attraction type normal conduction magnetic levitation. This configuration has higher thrust density and load capacity than those of others and it is expected that this is applied to a transport device etc.

Assuming this device is actually used as conveyance, the mass and the center of gravity change due to gripping objects, which degrades control performance and causes coupling forces. In this research, in order to solve these problems, the estimation method of the mass and the xy coordinates of center of gravity of levitating body without influence of modeling error and individual difference of permanent magnet of linear motor is proposed. The performance of this method is evaluated using simulator and a real device.

In order to achieve high acceleration motion by a linear motor, it is important that the z coordinate of the center of gravity is used for decoupling control with magnetic levitation. In this research, estimation of the z coordinate of the center of gravity and decoupling control considering the magnetic force of levitation and thrust force of linear motor are researched.

In this research, we deal with wave power generation called point absorber type that has a structure like a floating body attached to the movable part of a linear synchronous generator. By installing this on the sea, the up and down movement of the waves interlocks with the floating body to generate power. Although various researches and demonstration experiments have been carried out up to now, wave power generation has not been widely used because the power generation cost (yen / kWh) is large. In this research, by studying the design of the linear synchronous generator and its control, we aim to improve the generation efficiency and prolong the life of the generator and reduce the generation cost.

What is required of the linear synchronous generator is to reduce a force called a detent force, to facilitate generator control, and to increase the power density per unit volume. In this research, we focus on a two-stage vertical and horizontal flux type linear synchronous generator with two mover magnet plates. We succeeded in reducing the detent force by 83% by shifting the electrical phase of the two mover magnet plates. In addition, by changing the phase arrangement of the armature winding, a decrease in output was also prevented. As mentioned above, the method to reduce the detent force effectively without lowering the power density can be established, and the usefulness of the proposed method is shown by the measurement of the prototype.

What is required of the control of the power generation system is to be able to perform highly efficient power conversion and to operate while satisfying physical constraints such as the generator stroke. Therefore, we focused on impedance control, which has been studied for a long time, and model predictive control, which has been actively researched in recent years, and have studied by numerical calculation. As a result, it was found that the performance is significantly reduced when a wave exceeding the performance of the generator is applied. In the future, we plan to construct a control algorithm that always satisfies the physical constraints of the generator, consider installation of a mechanical generator protection device, etc., and carry out desktop experiments and water tank experiments using a prototype.

Magnetically levitated transport carrier systems have the potential of being extensively used in fields such as precision machinery, semiconductor fabrication, factory conveyance etc. in the future. Since these applications require high precision, magnetic levitation can provide complete contactless motion without the inherent presence of friction due to linear bearings allowing higher precision performance as well low wear and tear. However, magnetically levitated carriers suffer from high control complexity and expensive hardware, since replacement of linear bearings puts a 5 DOF control requirement on it. For that a conventional magnetically levitated carrier always has more actuators than required DOF i.e. 6 to 7 electromagnets controlling heave z, guidance y, pitch ��y, roll ��x and yaw ��z. Owing to the unstable nature of electromagnetic levitation, these DOF request active stabilization which increases the hardware, cost and control complexity. Furthermore, literature shows that for structural integrity, generally magnetically levitated carriers have bulky frames and require extra non-magnetic material which increases carrier weight and reduce effective loading capacity.

Overcoming these problems, we are developing a novel 6 DOF compact carrier comprising of a magnetically levitated transverse flux permanent magnet linear synchronous motor. The proposed structure is compact with almost entire volume composed of magnetic cores, orthogonally arranged and used for either thrust production or levitation. For a six DOF motion, only four actuators are employed for thrust, pitch, roll and heave active control and by virtue of the proposed motor, yaw and lateral motion are passively stabilized.

In most of the railways across the world, single phase 25kV traction power supply is obtained by connecting the single-phase traction transformer at substation across the two phases of the utility grid in a cyclic manner. The asymmetric loading of this single-phase load in different sections of traction distribution network causes voltage unbalance at the point of common coupling (PCC) with the three-phase grid system. This adversely affects other consumers of the connected network. With the ever-increasing traction load and introduction of high-speed railways in different countries this power quality problem is attaining newer dimensions.

In this research, we propose to model the moving trains along with the power supply system to study the extent of unbalance in the grid due to each influencing factor: traction transformer configurations like single phase, Scott etc., headway, speed and total weight of the train. Further it is proposed to integrate optimum scheduling of the traction load for restricting the unbalance within the IEEE limits in an already established system where it is economically not feasible to entirely change the infrastructure like transformer.

In this study, three phase induction motor torque control using a disturbance observer and PID controller is proposed for wheel slip control and re-adhesion as delay in wheel slip control by conventional methods not only causes rail and wheel burns and increased maintenance but also sometimes results in stalling of trains and poor riding index. Force of adhesion existing between rail and wheel is responsible for transferring the angular tractive force acting on wheel to a longitudinal force which is required to pull the train and is also responsible for slip velocity. Based on typical characteristic of adhesion coefficient and slip velocity, a two mass rail and wheel load model driven by DC motor controlled by proposed method is analyzed using Simulink for better appreciation and to extend the understanding further for analyses of real train model. In proposed control method change in rail and wheel adhesion is observed as change in load torque by disturbance observer to adjust input torque obtained by motor accordingly. Also literature review is done for the re-adhesion methodologies governed by the concept of rate of change in slip velocity and conservation of angular momentum. Future work is envisaged to do Simulink based performance analyses of proposed control and re-adhesion method for real train model.

In a highly frequent urban train service, suppression of delay propagation due to excessive dwell time is significant. In this research, we propose a speed control method to minimize the interval between departure and arrival at a station under uncertainty of preceding train’s departure and evaluate the performance.

In this study, an approach of power management in urban railway system (DC railway system) is proposed. The proposed method aims to provide an improvement of energy-saving operation by combining the design of driving strategy, train scheduling, and use of Energy Storage System. Basically, the design of driving strategy provides efficient energy-saving operation of each train and the design of train scheduling provides effective regenerative power usage among trains. However regenerative power can be utilized by exchanging among trains. There is considerable amount of surplus regenerative power. To manage surplus regenerative power, energy storage system is installed at suitable location with optimal capacity. By integrated design of software and hardware approach, power management can be more flexible and effective leading to increment of usable regenerative power. The numerical case study of Bangkok Rapid Transit System was performed and evaluated as a preliminary process for developing the proposed method. The numerical results showed that the improvement of energy-saving operation can be achieved.

This research focuses on energy savings in traction of rolling stocks and applies energy saving driving operation. Increasing maximum acceleration rate of the rolling stock is proposed. By applying it, energy saving running is achieved by decreasing powering energy because of decreasing the rate of powering in total running time and lowering maximum speeds. The authors performed on-track tests and obtained results by applying proposed energy-saving 6.0 % of energy saving effect against conventional running profiles and 4.7 % against energy saving running profiles without increasing acceleration rate. Effectiveness of increasing acceleration rate is proven by on-track tests as well as numerical evaluations.

都市鉄道において通勤時間帯の混雑や列車の低速化は大きな課題となっている， 本研究では，この課題を抜本的に解決する方法として，区間的に線路を増やし，急行列車の導入及び列車の高頻度化を図る方法を提案する． 本論文では，列車計画の評価法として旅客の移動時間や乗換・混雑による不効用の和，設備計画の評価法として追加する設備の費用を考慮し， これらを二軸的に評価することによってよりよい評価の選定を行う． これらの計画の選定に向け，先行研究の損失時間ダイヤを用い，これを拡張して線路の増設が必要な最小限の区間を求める方法を示す． 全体の列車計画においては，急行の停車駅の問題と緩行の待避の問題の２つに課題を分割した上で， 前者を全探索によって，後者を最小費用流問題の拡張によって計算する． ケーススタディを通じて本手法で作成された計画の比較を行い，設備計画として考慮する年数が長くなるにつれて区間的線路増備の効果が大きくなること， および先行研究に挙げられる３線による運行計画および待避設備を用いない緩行のみの列車計画と比較して， 旅客の不効用および設備費用が中間となるような鉄道計画が可能であることを示した．

This research is about high precision positioning system’s disturbance rejection. High-precision positioning system contains two parts, the actuator side and table side, which are connected by high rigid components, such as high rigidity leaf spring. The table side which is the non-collocated side has also been mounted with an encoder to guarantee the precision position of itself.

By feedbacking the table side encoder information into the actuator side to design the controller which not only guarantees the stability but also disturbance rejection performance of the whole system is conducted in this research. What’s more, the frequency response data only is used for designing controller which can eliminate the influence of model uncertainty and increase the feedback bandwidth.

Boost converters are conventionally used as constant or discretely variable voltage source. It is expected as a variable voltage source applied for electric drive for mass motor drives based on pulse amplitude modulation control. The transfer characteristics of boost converters are nonlinear and nonminimum phase which results undershoots of output voltage to step voltage reference. Because of this characteristics, precise control of boost converters as variable voltage source is difficult. In this research, we apply a pre-actuated feedforward which is effective for compensating the nonminimum phase LTI systems to boost converters. It has been confirmed that the proposed method can reduce tracking error of 71 % and settling time of 33 % in a step response.

This research aims to apply the pneumatic cylinder to the stage of a flat panel display (FPD) manufacturing machine. The machine is for patterning circuit patterns on a glass plate in order to fabricate the display. Recently, increasing in size of the glass plate, improving the definition of the display and reducing the cost are important factors. Therefore, the long-stroke micrometer order high-speed positioning control is required on the coarse stage. As the overheat and the big mass inhibit the improvement of performance, we aim to replace the electric motor with the pneumatic cylinder. However, the elements in the pneumatic driving system, such as valves, chambers, and a cylinder have the difficulties like the nonlinearity, the time delay and the resonances.

In this research, we focused on the nonlinear compensation of the valve which is the most basic element in the pneumatic driving system. Three steps constitute our proposal. The first step is to constitute the inverse model based on adaptive identification. The second step is the feedback control with a fast flowmeter. The third step is the feedforward control. Conventionally, the variation of the input-output characteristics makes the nonlinear compensation of a valve difficult. However, the accuracy of the inverse model is increased by adaptive identification and hence it is expectable that the two-degree-of-freedom control becomes effective. From the experimental result, the mass flow rate error reduced down to 10 percent comparing to the conventional method with the inverse model previously constructed.