Search

The series strings of reused battery cells connected in parallel have different available voltage ranges due to the imbalance between the SoC (State of Charge) and SoH (State of Health) of these cells. Therefore, the utilization capacity of this battery system is limited due to overcharge and over discharge protection of any one of all cells. This paper proposes hot-plug switches to solve this problem. Series battery strings are connected in parallel via switches that are self-controlled according to the state of the battery cell. Experiments have shown that the utilization efficiency of the reused battery system is improved. The proposed switches allow batteries to be hot-swapped without inrush current into the DC bus, and a reliable and redundant power storage system can be constructed.

This paper presents the measurement and simulation of surface charge decay on a polymethyl methacrylate (PMMA) circular disc. The dependency of charge reduction on varied thickness was investigated. For the experiment, a rod-plane electrode arrangement was used to generate the positive corona discharge on the dielectric surface of PMMA. The measured surface potential was used to calculate the electric field by using the finite element method (FEM). The simulation model is axisymmetric having boundary conditions based on the measured potential. The surface charge density was then calculated from the normal electric field in the PMMA side and that in the air side. The comparison of estimated results and simulation results of surface charge density was made. The surface potential distribution was initially bell-shaped then decreased with time. The mechanism of electric conduction through the volume of insulation and charge neutralization by ambient gas ions influenced the surface potential decay. In the case of thicker PMMA, the reduction of surface potential distribution was fast in comparison with other cases. The estimated surface charge density was similar to the simulation results.

The problem with conventional snoring sound identification methods is that their performance declines when the snoring sound is identified in the actual environment. Therefore, it is necessary to cope with the stationary and nonstationary environmental sounds that cause the decrease. In this research, we tried to cope with stationary environmental sounds by spectrum subtraction method for noise suppression. Non-stationary environmental sounds were regarded as one class for each type of environmental sound. We tried to identify the snoring sounds by multikernel learning, which is a multiclass extension of a support vector machine and by multilayer perceptron, which is a kind of neural network.

Hemodialysis patients are generally provided with an arteriovenous fistula (AVF), but problems such as stenosis of the blood vessel can occur. Listening to shunt murmurs is an effective way for hemodialysis patients to check their own shunt function. However, manually judging shunt function is difficult and requires correct knowledge and experience. Therefore, a system that can be easily handled by patients and automatically identifies shunt function is required. In this study, we proposed a method for classifying shunt stenosis using the Support Vector Machine (SVM). To record shunt murmurs, we used the electronic stethoscope and the microphone with stethoscope chest piece. We used the resistance index (RI) obtained from the ultrasonic diagnostic equipment as a class label and the normalized cross correlation coefficient, ratio of the frequency power and Mel-frequency cepstral coefficients (MFCC) as the feature. We examined the effectiveness of the proposed method by comparing the accuracy and F-measure. It was found that the classification accuracy of the RI using the SVM was lower than that obtained by human judgment.

A slot antenna backed by a quarter cylindrical cavity operating in 920 MHz-band was designed. The antenna was provided with a slot line for tuning in addition to the radiating slot in order to attempt the extension of the variation range of the input impedance. Next, it was schemed that an RFID chip having a complex impedance was directly loaded on the surface of the slot antenna. Then, a set of configuration parameters was sought related to the tuning stub for impedance matching by electromagnetic field simulations. As a result, it was confirmed that matching could be performed even with the chip having not 50Ω but a complex impedance by the tuning slot stub, and thereby that the slot antenna backed by a quarter cylindrical cavity could be used as an antenna for the RFID. A chip mounted prototype antenna was fabricated and its communication range was measured.

We measured electron and ion swarm parameters in gas mixtures of Ar or He with Dry Air at atmospheric pressure. This paper presents the average drift velocity of three kinds of charged particles and the effective ionization coefficient of electrons using two types of gas mixtures: Ar with Dry Air (Ar/Dry Air) and He with Dry Air (He/Dry Air). The Pulsed Laser-induced Plasma method was used in this experiment because this way can measure discharge current easily. Each gas were injected into discharge chamber to total gaseous pressure 760 Torr (1013 hPa) converted by 20 °C. As a result, the average drift velocity of electrons in Ar/Dry Air depended on mixing rate of Ar and changed remarkably at low reduced electrical field: E/P = about 5 – 18 V/cm Torr. On the other hand, it showed almost same value in a range of E/P = about 23 – 33 V/cm Torr. In case of the average drift velocity of ions, the drift velocity in He/Dry Air was faster than that in Ar/Dry Air at same E/P. Moreover, the drift velocity of positive and negative ions in Ar/Dry Air almost had no effect on injected Ar gas amount. Regarding the effective ionization coefficient of electrons, Ar/Dry Air and He/Dry Air depended on Air concentration. Especially, the critical reduced field strength Ec/P depended on Air concentration, remarkably.

Location prediction is essential to facilitate proactive support services. However, predicting a location that the user has not visited previously based on location history is difficult to predict the location where the user has never visit. Smartphones handle a significant amount of important personal data such as location, schedule, and email data. This paper proposes a location prediction method based on personal data acquired from smartphones. Experimental results based on personal data acquired over one year demonstrate that the system can predict user location precisely.

Recently, many researchers focus on glass fiber reinforced plastic (GFRP) and carbon fiber reinforced plastic (CFRP) materials because these materials have superior mechanical properties. Therefore, their electrical insulation performance must be revealed. Especially, the discharge which occurred under insulation barriers has not been revealed. So, we have investigated the discharge phenomena under GFRP and CFRP insulation barriers. In this study, we used a needle-plane short air gap in the discharge chamber. The insulation barrier was placed 2 mm below the brass needle electrode and 5 mm above the plane electrode. Synthetic air (N2 (79%)/O2 (21%)) or room air was injected into the chamber at atmospheric pressure. The steep impulse voltages were repeatedly applied to the needle electrode. As a result, the surface discharge (SD), it developed in both gases on GFRP insulation barrier. However, it didn't develop on CFRP insulation barrier. Regarding the discharges under GFRP insulation barrier (barrier-plane discharges: BPDS), they mainly appeared in its center area at synthetic air. On the other hand, the discharges under CFRP insulation barrier appeared in overall area at both gases. In addition, the discharges were not occurred with the standard lightning impulse voltage.

We demonstrate that the lasing wavelength and emission intensity of GaInAsP photonic crystal nanolaser can be controlled by immersing it to an electrolyte and applying a bias voltage. This method gives an option to the control of semiconductor lasers by the iontronic effect.

This paper presents the influence of circuit parameters on the small signal characteristics of the control to output transfer function of the LLC current resonant converter. The equivalent source model is employed for the analysis. To derive the dc gain and the internal resistance, circuit simulator is used. The iron loss that increases as the output current increases is considered in order to obtain more accurate results. The experiments are conducted that are in good agreements with the analytical ones. As a result, the major influence of the characteristic impedance on the internal resistance is found out.

High-voltage equipment needs to withstand various environments. It is well known that charged body between electrodes affects insulation performance. In the worst case, the charged body caused electrical breakdown. Therefore, it is necessary to study discharge mechanism in case that the charged body exists between the electrodes. To study it, we need the technique which control the charge amount of the charged body. In this work, we fixed a millimeter-sized particle between electrodes. In addition, high voltage was applied to the electrodes, and quasi-uniform field was produced. Then, we irradiated position of 2 mm above or below the metallic particle with YAG-Laser to control the charge amount of it. Consequently, the charge amount and polarity of a metallic particle can be controlled by laser.

Surface discharge transforms flashover through streamer to leader. Flashover causes damage to electric equipment and fire accident. Therefore, surface discharge velocity and distance have been researched by using Dust Figure and Lichtenberg Figure method for many years. Recently, surface discharge luminescence as time resolving pictures can be measured by optical equipment. In this study, a quartz glass plate having a back side electrode and luminescence permeability was used as a dielectric material. In this experiment, pictures of positive surface discharge luminescence on the quartz glass plate in atmospheric air were directly taken from its backside by ICCD camera. Additionally, surface discharge velocity on the quartz glass plate in atmospheric air was led from the distance of surface discharge development after 50 ns passed from the start of discharge. In this experiment, the steep positive impulse voltage was applied respectively 15–30 kV. As a result, positive surface discharge velocity was led from pictures of its luminescence and we confirmed that the more the applied voltage increased, the faster surface discharge velocity becomes.

The optimal parameters of noise suppression for detection of snoring activity are analyzed and we improve performance of detection of snoring activity in this paper. For detection of snoring activity, we use a Support Vector Machine which is one of machine learning. By training of grand truth and features, the SVM model is obtained. By applying test date to the SVM model, it is classified into snoring or non-snoring class. Using a mobile device, sleeping sound is recorded. The features for machine learning are computed from sleeping sound. To improve the detection performance, noise suppression is performed before features extraction. We examine the relation between the noise suppression parameters and the performance of detection of snoring activity. We investigate the optimal parameters of noise suppression for detection of snoring activity.