In order to prove the validity of prediction of penetration depth using Hablanian plot, we have compared experimentally-obtained penetration depths with those predicted by Hablanian plot and linear multiple regression analysis. The experimental results with ~1200 data points were obtained from bead-on-plate welding tests by the laser job shop. Stainless steel plates and four types of industrial CW lasers were used in these tests. From linear multiple regression analysis, it was found that dominating factors were laser power, welding speed, and spot diameter. On the other hand, we can predict the penetration depth using the 3^th-degree polynomial fitting curve (standard curve) expressed by Hablanian plot with two dimensionless parameters. The fitting difference between the standard curve and a data point results in the error of prediction. From Hablanian plot, it showed (1) that this fitting difference was dependent on some factors (such as laser power, welding speed, power density, and penetration depth), (2) that the fitting difference was, respectively, negative and positive in the lower and middle welding speed ranges, and (3) that the penetration depth could be precisely predicted by correcting the fitting difference in welding speed from the standard curve compared to the results predicted by linear multiple regression analysis.

In the mechanically induced long-period fiber grating (MLPFG) fabricated by applying the periodical pressure on an optical fiber, the small bending occurs due to the periodical pressure, which influences the losses of the core and cladding modes. We propose the theoretical model of the MLPFG for estimating its transmittance based on the transfer matrix method and investigate the effect of the losses of the core and cladding modes in the MLPFG on the transmittance, theoretically and experimentally. We numerically clarify that the transmitted light spectrum of MLPFG shows only the main attenuation lobe, no side lobe and the attenuation bandwidth is broadened as the cladding mode loss increases. We also clarify that the coupling coefficient, and the losses of the core and cladding modes in the fabricated MLPFG can be estimated from the measured transmitted spectrum based on our model. Moreover, we measure the transmittance of two types of MLPFGs that fabricated with a screw, weights and metallic plates, and that fabricated with a screw and a heat-shrinkable tube. The measured spectra show good agreements with the calculated ones using our model. Our model will be useful for designing MLPFGs applied to the sensors.

AI techniques are required for realizing society 5.0. For the issues of societal implementation for AI, an AI hardware, which is enhanced security performances including authentication, and so on, is needed in order to reduce security risks. This study proposes a new physical unclonable function (PUF) based on neural network (NN) called NN PUF. The proposed NN PUF uses a difference of calculation time in NN due to production variations of semiconductor. Evaluation experiments using a field programmable gate array (FPGA) prove the effectiveness of the proposed NN PUF.

In muscle rehabilitation and training, it is important to evaluate a patient's muscle state. In general, training is based on physical therapy, but there has not been quantitative assessment of the actual effects on muscles. The authors previously tried to quantify muscle fatigue and defined “Muscle-fatigue time (MFT)” as the quantitative evaluation index of muscle fatigue based on electromyography and the use ratio of the muscle fiber type. MFT is the time when the use ratio of the slow muscle fiber (Type I) and the fast muscle fiber (Type IIb) intersect. Furthermore, the authors also proposed a method to estimate MFT by modeling the relationship between each muscle load and MFT. This paper examines the appropriate evaluation method for MFT containing variation, and generalization performance of the constructed model.

MRI diagnoses various diseases by using a very strong magnetic field. In order to prevent electronic equipment malfunction caused by this magnetic field, the International Non-ionizing Radiation Protection Committee regulates access around the MRI laboratory, MSR (Magnetically Shielded Room) is used for shielding the magnetic field. Since the conventional MSR for MRI has a shape that the upper, lower, and side surface are hermetically sealed with shields, it gives closure to patients. Recently, open-type MSRs have been researched which have openings on the side, which can take in outside light and scenery into the MSR and can see the state of examination from the outside. However, open-type MSR has a problem that the volume increases from the shape of the shield at the opening. In this research, in order to propose an open type MSR with a volume equal to or close to that of the conventional model, we analyzed the shield cross sectional area and gap width as a parameter. Three-dimensional static magnetic field analysis using finite element method was used.

Brain-computer interface enables people who cannnot move their own body freely to manipulate machines and helps their communication and life. Recent BCI uses multimodal stimuli to increase bit rate of the system, so it is important to reveal when and which part of the brain part is activated to discriminate target stimuli. Though magnetoencephalography measures brain activity with high spatial and temporal resolution, there is a few index to estimate spatiotemporal locality of response. We propose an index to estimate spatiotemporal locality of response to multimodal stimuli of oddball paradigm from magnetoencephalography (MEG) signals. The validity of proposed index is demonstrated by simulation. Discrimination task was conducted with different ratio of target/non-target stimuli with visual and auditory location. Response to visual target stimuli were strong and made no spatiotemporal defference, so visual dominance was implied.

A virtual-sound catch ball game system was developed to improve the quality of life of the visually impaired. Computer-generated virtual sound balls were represented by stereo sounds, comprising localization parameters such as the interaural time and level differences, echo, and head-related transfer function. Ball-catching success was determined by images of the hand captured by a USB camera. Two sessions of experiments with blind subjects demonstrated the importance of distance recognition for human localization capability. Moreover, the addition of an echo parameter enhanced the sound-source image position and sense of enjoyment. The results showed that a virtual sound game is a good active-entertainment option for the visually impaired.

Neuronal systems are dynamical. In the dynamical system, the externally-driven responses, called transient dynamics, are stimulus-specific but reproducible. Under the assumption that the neuronal system is deterministic, we here hypothesized that such reproducible transient activities could produce computational capability in a chaotic dynamical system. To test this hypothesis, we estimated the maximal Lyapunov exponent of neuronal activities in the primary visual cortex (V1) of mice, and quantified their information processing capacity in the transient dynamics. Consequently, V1 was characterized as a chaotic system, where almost identical input time-series led to different trajectories. We also demonstrated that, when mice were visually stimulated with drifting gratings, the trajectories contained the input time-series information for at least 5.2 s after stimulation. These results suggest that computational capability in V1 emerges from reproducible transient activities in the chaotic system. Yet, the estimate information processing capacities in V1 were much lower than those in theoretical studies. Further verification is still required to elucidate the discrepancy between theoretical and experimental results.

Heart failure is one of frequent diseases for the elderly, and it is required to unveil the pathogenic mechanism of reentry, a type of lethal arrhythmia. Although fibrosis can cause arrythmia, little is known about how fibrosis disrupts normal cardiac function, and then causes arrythmia. In this study, we aimed to model reentry-like activity in vitro for elucidating the mechanism of reentry. First, rat cardiomyocytes were purified with a glucose-free medium, and cultured on microelectrode arrays. Extracellular potential was measured with or without lidocaine, a proarrhythmic reagent, on day 6-7. Reentry-like activities were observed from three out of five samples in presence and absence of lidocaine. We evaluated transitions between reentry and normal propagation. Activity pattern gradually changed from normal propagation to reentry, but that did rapidly from reentry to normal propagation, suggesting different mechanisms between start and end of reentry. Furthermore, changes in propagation pattern during reentry suggested that delay prolongation in partial area can terminate reentry-like activity in vitro. Overall, our system is suitable for elucidating pathogenic mechanisms of reentry.

In recent years, brain-machine interface (BMI) is attracting attention. BMI is a technology that enables machine operation using biological signals such as EEG. For further advancement of BMI technology, there is a need for advanced BMI devices. Therefore, the purpose of this study is development of BMI hardware specialized for handling EEG as an interface for human adaptive mechatronics (HAM) that know human’s state and operate according to the state. As one of the examinations, we are constructing a pattern recognition processor for EEG in real time on Field Programmable Gate Array (FPGA), which is an LSI that can reconfigure the processor. This paper reports on the designed EEGNet processor and the result of logic circuit simulation and implementation.

Ultrasonography is noninvasive and easy to perform, so it is usually used for treatment of myocardial infarction. However, the inspection technology is difficult, and a system to replace the inspection has not been developed. We have developed a measurement environment that extracts inspection techniques by measuring subjects, probe position, respiration and the force that the operator applies to the probe. We evaluated the techniques of the sonographer and developed a system that non-medical personnel imitates ultrasonography performed by cardiologists. In the future, we aim to develop a diagnosis environment automated by robots using such information.

Ultrasonography is a popular method used for medical diagnosis since it is non-invasive and does not require extensive equipment. However, it requires appropriate control of the probe orientation, which is difficulty even for experts. As an approach to realize automatic ultrasonography, we analyze echocardiography video using dynamic mode decomposition (DMD) to obtain its feature value. The feature value can be calculated from the dominant modes extracted using DMD. This gives the dynamics on the structure which exhibits pulsating modes, and the entire heart structure can be extracted as well from the stationary mode.

We measured the auditory steady state responses (ASSR) in a magnetoenchephalogram to study the mechanism of neural processing in the auditory cortex with octave illusion proposed by Diana Deutsch. Octave illusion is induced using consecutive dichotic sounds, where two tones that are one octave apart, e.g., 400 and 800 Hz, are alternated between the ears at 500-ms intervals. We then applied the frequency tagging method to the ASSRs that were recorded during the octave illusion. We compared the ASSR amplitude between non-illusion and illusion groups. The results showed that there was difference in the ASSR amplitude of the auditory cortex between the illusion and the non-illusion groups. In non-illusion group, the ASSR was showed right hemisphere dominance and contralateral activation. There was no significant difference in the illusion group. These results suggest that right hemisphere dominance and contralateral activation are related to neural processing of the octave illusion.

Long-term and continuous vital sign monitoring is essential for the early detection of hypertension. Blood pressure sensing based on non-contact biological measurement indices such as photoplethysmograms and skin temperature that can be obtained using visible light and infrared images has been attempted in previous studies. Near-infrared light has high transmissivity to living tissues and has been used for measuring arterial oxygen saturation. This study aims to improve the accuracy of non-contact blood pressure sensing using near-infrared light.

In the electricity system reform currently being promoted in Japan, the transmission and distribution division in general electric power companies will be legally unbounded as a power transmission and distribution business operator (TSO) in 2020 to ensure the neutrality of the transmission and distribution business. TSOs procure the reserves through a public offering, but from 2021 they are going to procure the reserves through the reserve market. For the time being, TSOs will make a contract in a merit order of the ability of the reserves, called ΔkW, at the time of procurement. Then, TSOs will activate the reserves out of the procured ΔkW in a merit order of electric energy, called kWh, at the time of operation. However, to minimize the sum of procurement cost and operation cost, it is necessary to study the clearing problem considering ΔkW and kWh values. In this paper, we propose a clearing problem that considers ΔkW and kWh values simultaneously in the reserve market, then, using stochastic optimization we formulate the clearing problem as a 0-1 mixed integer programming problem.

In the deregulated electricity market, power producers are exposed to risks of generation imbalance and decreasing earnings. In order to stabilize earnings, they need to make generation plan of their own generators and bidding plan to the electricity market in consideration of these risks. These risks are due to uncertainties of forecasts of electricity price, renewable energy generation, electricity demand, and so on. Up to now, there are many studies about stochastic unit commitment methods considering these risks. However, conventional methods have problems that operators have difficulty in reflecting risk appetite intuitively. In this study, we present a stochastic unit commitment method which optimizes VaR (Value at Risk) of profit with a given confidence level. A confidence level means the percent of event which operators consider for risks to all events, so operators can reflect own risk appetite intuitively. The usefulness of the proposed method has been confirmed by numerical simulation. The numerical results show that when a power producer has low risk tolerance, the volume of forward contract increases. Studying efficient calculation methods and formulations and the evaluation using actual scenario are future tasks.

This paper introduces an advanced video sensing technology which captures vibration of structural objects, and some efforts to apply it to the deterioration state evaluation of social infrastructure. The technology firstly applies motion tracking to image sequence captured by monocular video camera, and then retrieves minute 3D motions of target objects and in-plane displacements from motion vector fields. The technology finally estimates internal deterioration states based on spatio-temporal behaviors of structures. We have tested the technology in real environments and confirmed that it can successfully retrieve some structural behaviors of hydro turbine bearings and road bridges under external forces.

Recently, SLAM (Simultaneous Localization and Mapping) becomes a hot research topic in computer vision due to high demand to reconstruct surrounding environment with camera mounted on drone, etc. In SLAM system, Direct SLAM system has reported that it can achieve high 3D reconstruction environment since it utilizes the information of all pixels in images. In addition, to create accurate 3D reconstruction environment, SLAM system requires high camera tracking performance. Fortunately, to achieve this, it is well-known that a camera with wider field of view can help the performance improvement. From this fact, in this paper, we propose a new Direct SLAM system with 360-degree camera, which has two fish-eye lens and can capture front and back view scene. The characteristic points of our proposed method is that 1) we utilize LSD-SLAM system, which is a Direct SLAM one, since it can achieve faster 3D reconstruction compared with other SLAM systems, 2) we integrate the information of two camera coordinates to achieve high camera tracking accuracy, 3) we cover the frame-in/frame-out information by seamlessly taking over the camera tracking information from a fish-eye camera system to another one. From the experimental results with CG simulation, we achieved higher camera tracking accuracy with 360-degree camera compared to that with a fish-eye camera.

Recently, deep learning has been studied as one of the most effective methods in the machine-learning field, and lots of results have been reported. However, the most effective way to construct neural networks has not yet been determined. Besides, the interpretation of an obtained network by a user is difficult. To solve this problem, we have proposed a neural network with a support vector machine (SVM) called “SVM-NN”. In this proposed method, support vectors in the SVM determine the number of neurons in the neural network and their weights and biases. Then, the hyperplane of the neural network is expected to behave similarly to that of the SVM before training. This method has an advantage in that users can understand the mechanism of the network based on the support vectors. However, there are several problems to apply SVM-NN to real problems. In this study, we proposed the SVM-NN with the genetic algorithm. To confirm the effectiveness of proposed methods, the computer simulations are carried out taking benchmark problems as examples.

Requirements for optimizations are not only to find an optimal solution in sufficiently long computation time but also to find a reasonably superior solution in shorter time, especially for real optimization problems. In this paper, we propose a real-coded genetic algorithm that finds the reasonably superior solution efficiently in short computation time. In genetic algorithms, the population size is an important parameter related to the population diversity and convergence speed. By switching the population size from a small value to a large value, the proposed method finds the reasonably superior solution rapidly.

CAN is widely used as an in-vehicle network. However, since security of CAN protocol is not enough, various attacks on in-vehicle system have been reported. Therefore, as a countermeasure against the attacks, message authentication using a Message Authentication Code (MAC) has been proposed. In message authentication that also prevents replay attack, a counter is used for MAC generation. However, if the counters are out of synchronization between sender and receiver, subsequent messages cannot be authenticated. Therefore, a counter synchronization function is required to continue communication using message authentication. This study proposes a new counter synchronization method for message authentication in CAN communication. The proposed method synchronizes the counters using resynchronization message, which is based on a newly introduced synchronization counter. In order to reduce the communication traffic, the proposed method sends only the difference between the introduced counter and the counter for MAC generation. Experiments using mock-up system prove the validity of the proposed method.

Home Energy Report (HER) is gathering attention as a method to promote energy conservation behavior in residential sector. Most of the existing efforts about HER assume the use of total monthly usage data. These usually compare customers' monthly usages with that of similar neighbors. In our study, we developed a HER that uses smart meter data to give graphical information on customers' monthly, weekly, daily and hourly electricity usages. In the smart meter-based HER, the graphs that show hourly usage patterns are applied the method of automatically generating a message commenting the time zone to be paid attention by each customer. The automatic generation method uses smart meter data and several types of baseline hourly usages to analyze which time zone each customer should pay attention. In this paper, we provide a detailed explanation about the structure of the automatic generation method, and show analysis results of how the method have been worked by using the HER generation log data of approximately forty thousand households for twelve months.

Electricity consumption forecasting plays an important role in establishing and maintaining electric supply management systems. Power companies need to keep a balance between the power demand and supply for customers; this requires an accurate forecast. However, electricity consumption forecasting is affected by various factors such as different weather conditions, season, or temperature. If we cannot predict electricity accurately, the balance between the demand and supply would be destroyed, which may cause huge penalties to power companies. Therefore, electricity consumption forecasting is an important task. The purpose of this study was to forecast the electricity consumption of a manufacturing company every half an hour in the next day to prevent a power supply company from running out of power. In our work, we proposed a short-term electricity consumption forecasting method based on the attentive encoder-decoder and several nonlinear multi-layer correctors. The proposed method is verified in several experiments by using the actual data on electricity consumption of the manufacturing company. The results show that the proposed method outperforms previous methods.

We developed a novel magnetoencephalogram (MEG) system that features both a superconducting magnetic self-shield and a zero boil-off system. We successfully detected somatosensory evoked magnetic fields in human subjects using our MEG system. These results indicate that brain activity is being captured correctly compared to a previous study. This MEG study further shows the effectiveness of the superconducting magnetic shield and zero boil-off helium recovery system for research applications.

Stochastic resonance phenomenon improves sensitivity of a signal is under a certain probability by adding noise to a weak signal. The phenomenon might be implemented in electronic circuits, and it was used various fields. For example, the phenomenon was used to detect digital image data in severe shooting conditions. Therefore, we considered if it is possible to construct a stochastic resonance circuit in digital circuits, it can be integrated into various digital devices. In this study, we construct a stochastic resonance circuit with FPGA (Field-Programmable Gate Array), and investigate the change in detection accuracy. As a result, the improvement of the correlation coefficient was confirmed by increasing the number of parallelization.

Lasers possessing high coherence and narrow spectral line width have low output power. Coherent beam combining (CBC) is a method for generating a high-power laser, whilst maintaining coherence. However, the technology is not easily available to non-professional users and a new approach is required. The interference signal generated due to the interference between two independent lasers was observed and a frequency stabilization system using the absorption of ammonia was constructed. Beat frequency was stabilized to 3.9 MHz which was less than the spectral linewidth of the laser light and suggests that the frequency condition for CBC is satisfied.

We report on investigation of photon-stimulated desorption processes of polymethylmethacrylate (PMMA) induced by vacuum ultraviolet (VUV) irradiations. Atoms or molecules on PMMA surfaces were to be desorbed and dissociated as a result of absorption of the wavelength-selected VUV photons. Desorbed and dissociated atoms or molecules were then detected by a mass spectrometer which were obtained as a function of the irradiation wavelength. Desorption species with mass number of 28, 29, and 30 were detected at center wavelength of around 160-170 nm, which were identified as CO, CHO, and C₂H₆, respectively. These photon-stimulated desorption processes of PMMA depend on bond energy or molecular structure.

With development of information and communication technology, condition monitoring system for train vehicles has been improved recently. By monitoring the vehicles and quickly informing the train crew of vehicle’s condition data, it is expected to develop safety and stability of the train operation. However, some existing vehicles cannot install networks for transmitting the monitoring data, because it is difficult to lay new wired lines due to insufficient space. On the other hand, it is relativity easy to secure space for a wireless terminal. To choose this alternative, a wireless network to monitor the vehicles has to be constructed. In this paper, we propose a method of construction of an inter-vehicle network. Moreover we consider the sequences of data transmission for regular and urgent. Then, we implemented a prototype system using the proposed method and performed the test of the prototype for confirming functionality and evaluating the method.

An immediate evacuation is essential for safety when a big earthquake and a tsunami occurred. For example, Great East Japan Earthquake in 2011 killed over 10 thousand people and over 90 percent of them drowned by the tsunami, that is, people who failed to escape from the tsunami after the earthquake occurred got losses by it. If they evacuated immediately after an earthquake, it would be possible that the losses could be limited. In this study, we analysis text-based information expression on a mobile device to induce immediate evacuation. We implemented an application which provides disaster information on a smartphone, and compare text expression about information of tsunami height, arrival time of tsunami, an expected number of evacuees, damage of buildings and so on. The results of the comparison verification show the sentence based on damage of building using fear-arousing communication is the most effective to induce evacuation in this study. However, in over seventies, the effectiveness of the sentence is less than half.

Questionnaire and online testing are very important for teachers and students to realize their effective education as well as to improve their environment. This paper describes a Web-based tool to support educational environment, which can provide suitable services to design easily, to carry out smoothly, to report visually and to analyze statistically questionnaire/testing. We have developed our Web-based tool with “Ruby on Rails”, applied it to real lectures and evaluated it based on feedbacks from teachers and students through practical educational situations. With such an application, teachers will be able to carry out Web-based questionnaire/online testing, to review or score the feedback from students and to visualize students' understanding levels easily and smoothly.