Muroran-IT Academic Resource Archive
Not a member yet
5947 research outputs found
Sort by
Building Energy Efficient Semantic Segmentation in Intelligent Edge Computing
Semantic segmentation is a critical area in computer vision, which needs voluminous image data streaming from user devices. Usually, it is challenging to process semantic segmentation tasks in user devices due to the limited computation power and battery life. Intelligent edge computing effectively enhances the accuracy of semantic segmentation tasks by offloading computations to nearby devices, providing lower latency and improved responsiveness. However, inefficient offloading brings additional energy consumption due to the irregular relationship between task requirements and offloading settings. In this paper, we attempt to improve energy efficiency for processing semantic segmentation tasks in the edge environment by leveraging energy consumption and task requirements. We first investigate the power consumption with different offloading settings in a real intelligent edge environment. Based on the investigation, we formulate the offloading setting as a restricted multi-armed bandit problem and solve it by enhancing the upper confidence bound algorithm. Comprehensive simulation results show that the proposed solution significantly improves the energy efficiency for offloading semantic segmentation tasks in a given intelligent edge environment.journal articl
The Effect of Covid-19 pandemic in the aspects of Disaster Stress : Compare with the effect of 2018 Iburi East Area Earthquake
application/pdfAfter Covid-19 pandemic , a questionnaire was conducted on the psychophysiological situation and the implementation level of preventive method and stress level toward the student who affected by Iburi East Earthquake. It is indicated that the effect of psychophysiological situation is almost same level with immediately after earthquake and prevention method,especially mask and behavior restriction, are serious stressor. Covid-19 pandemic strikes people multiple stress as disaster stress.学術論文departmental bulletin pape
触覚の取得:粘土造形に関する研究
This paper examines student attitudes regarding the use of plasticine as a learning tool in the university language classroom. Plasticine clay was used as a tool where students had the opportunity to create models of vocabulary using their imagination and hands through pair work. The study in question introduced plasticine as a tactile learning tool in two English Communication EFL classes across three sessions. These sessions were then evaluated regarding the use of plasticine and its potential from a student perspective and using teacher reflections. This potential was rooted in factors such as memory, enjoyment, timing and future frequency. This research was undertaken to establish a baseline regarding the feasibility and levels of engagement around plasticine modelling which is rarely used in university language class settings. A survey questionnaire was utilized to garner the students’ attitudes. The results show that there was a positive endorsement of using plasticine and that there is potential to increase the frequency with which it could be deployed in the university classroom.journal articl
Self-generation of reward by logarithmic transformation of multiple sensor evaluations
Although the design of the reward function in reinforcement learning is important, it is difficult to design a system that can adapt to a variety of environments and tasks. Therefore, we propose a method to autonomously generate rewards from sensor values, enabling task- and environment-independent reward design. Under this approach, environmental hazards are recognized by evaluating sensor values. The evaluation used for learning is obtained by integrating all the sensor evaluations that indicate danger. Although prior studies have employed weighted averages to integrate sensor evaluations, this approach does not reflect the increased danger arising from a higher amount of more sensor evaluations indicating danger. Instead, we propose the integration of sensor evaluation using logarithmic transformation. Through a path learning experiment, the proposed method was evaluated by comparing its rewards to those gained from manual reward setting and prior approaches.journal articl
同期電動機のためのON/OFF法をもとにした位相最適化
室蘭工業大学Muroran Institute of Technology博士(工学)With the electrification trend of electric vehicles and industrial machines, more and more attention has been paid to electric motors. As a conversion device of electrical energy and mechanical energy, the performance of the motor will directly determine the competitiveness of the product. The performance of a motor is basically determined by the motor structure, which can be optimized by many different methods according to the way of shape expression. The topology optimization method is a kind of state-of-art method for structural optimization, in this method, the material can be freely distributed in the design domain. Compared with other geometry design methods such as the size design and shape design, topology optimization has the highest freedom to express the shape. Topology optimization was first used in the design of mechanical structures, and so far it has been extended to more fields, such as the design of electromagnetic devices and electric motors.
The purpose of this study is to broaden the application boundaries of topology optimization techniques in the design of electric motors. The author believes that there may be unexpected discoveries using topology optimization techniques to design iron cores in electric motors by topology optimization method as the optimized core can modulate the flux distribution. To this end, a typical topology optimization method - the ON/OFF method, is introduced for four interesting research topics, and the research findings are elaborated.
In the first research topic, the ON/OFF method is introduced into the design of the consequent-pole permanent magnet motor. The consequent-pole permanent magnet motor is a kind of special motor, it has attracted much attention in recent years due to its advantages of saving the amount of permanent magnets. As the design findings, it has been confirmed that the cogging torque and torque ripple of the motor can be suppressed to a rather low level if the iron poles are carefully designed.
In the second research topic, the ON/OFF method is introduced into the design of permanent magnet synchronous linear motors, and it is found that the end effect, which was considered an inherent defect of permanent magnet synchronous linear motors, can be almost eliminated.
In the third research topic, the ON/OFF method co-works with the electromagnetic-structural Multiphysics simulation to reduce the stator vibration of a switched reluctance motor. It is found that opening a inner window on the rotor poles and adopting anchor-shaped rotor pole are two design techniques for reducing radial electromagnetic force and stator vibration.
In the last research topic, aiming at the problem that both topology optimization and parameter optimization are simultaneously needed in the design of permanent magnet synchronous motor, a technique of coupling topology optimization and parameter optimization is proposed. A multi-objective optimization design of a permanent magnet synchronous motor
is carried out using the proposed technique, and the numerical results show that the proposed method can be used in the early conceptual design of a permanent magnet synchronous motor.
Overall, the author has conducted a series of studies on the design of some special motors and topology optimization - a state of art structural design method, out of author’s interests and research bases, in the hope that this paper can provide a reference for researchers in academia and engineers in industry.電気自動車や産業機械の電動化に伴い、電気モーターへの注目度が高まっている。電気エネルギーと機械エネルギーの変換装置である電気モーターの性能は、生産品の競争力を直接左右することになる。電気モーターの性能は基本的にモーターの形状で決まる。構造最適化(structuraloptimization)手法は、形状表現手法によるサイズ最適化(sizeoptimization、形状最適化(shapeoptimization、トポロジー最適化(topologyoptimizationの3種類に分類できる。トポロジー最適化法は、構造最適化の最先端手法であり、この手法では、設計領域で材料を自由に分布させることができる。サイズ設計や形状設計といった他の形状設計手法と比較して、トポロジー最適化は形状表現の自由度が最も高い手法である。トポロジー最適化は、機械構造物の設計で初めて用いられ、現在では、電磁デバイスや電気モーターの設計など、より多くの分野へ拡大されている。本研究の目的は、電気モーターの設計におけるトポロジー最適化手法のプロモーションである。最適化された回転子は磁束分布を適切に分布させることができるため、トポロジー最適化手法を用いて電気モーターの鉄心を設計することで、思わぬ発見があるか期待できる。本研究では、4つの興味深い研究課題について、トポロジー最適化手法であるON/OFF法を導入した。一つ目の研究課題では、consequentpole型永久磁石モータの設計にON/OFF方式を導入した。consequentpole型永久磁石モータは特殊なモータの一種であり、永久磁石の使用量を節約できる利点から近年注目されている。トポロジー最適化により,ローター鉄心の最適化を行ったところ、モータのコギングトルクやトルクリップルをかなり低く抑えられることが確認できた。二つ目の研究課題では,永久磁石同期リニアモータの設計にON/OFF法を導入し,永久磁石同期リニアモータの固有の欠陥とされていたendeffectをほぼ解消できることを明らかにした。三つ目の研究課題では、スイッチドリラクタンスモータのステータ振動を低減するために、ON/OFF法と電磁構造マルチフィジックスシミュレーションを結合した。その結果、ローターポールに内窓を設けることと、アンカー型ローターポールを採用することが、半径方向の電磁力とステーター振動を低減する2つの設計手法であることがわかった。四つ目の研究課題では、永久磁石同期モータの設計において、トポロジー最適化とパラメータ最適化の両方が同時に必要であるという問題に着目し、トポロジー最適化とパラメータ最適化を連動させるハイブリッド最適化手法を提案するものである。提案手法を用いて永久磁石同期モータの多目的最適化設計を行い、数値解析の結果、提案手法が永久磁石同期モータの初期概念設計に利用可能であることを示した。本論文で示したた特殊モーターの設計事例と最先端の構造設計手法であるトポロジー最適化の適用研究が研究者やエンジニアの参考になることが期待できる。doctoral thesi
低速度衝撃を受ける四辺単純支持RC版の耐衝撃能力評価に関する数値解析的研究
室蘭工業大学Muroran Institute of Technology博士(工学)主要な構造部材の1つである鉄筋コンクリート(RC)版構造において,集中荷重が静的かつ局所的に作用すると押し抜きせん断破壊によって終局に至るとともに,その設計法は確立されている。一方,落石などの低速度衝撃が RC 版に作用する場合には,未だ合理的な耐衝撃設計法は確立されていない。したがって,衝撃荷重を受ける RC 版の耐衝撃設計法を確立し,実構造物への適用を早期に実現するためには,実験とともに数値解析的検討も併用して効率的に研究を推進することが重要である。
このような観点より,本研究では低速度衝撃を受ける RC 版の終局状態を数値解析的に推定可能な手法の確立を目的として,四辺単純支持矩形 RC 版を対象に,三次元弾塑性衝撃応答解析を実施した。得られた解析結果と実験結果を比較することにより解析手法の妥当性を検討するとともに,終局時には衝撃荷重の第2波が消失することや除荷後における支点反力およびたわみにおいて自由減衰振動が励起されないことを実験および数値解析結果から確認し,これらの状態に至る場合を RC 版の終局状態と推定する手法を提案した。さらに,提案の推定手法をコンクリート強度や版厚が異なる RC 版に適用し,実験結果と比較することによってその適用性を検証した。本論文の構成は,以下の通りである。
第1章は序論であり,本研究の背景や既往の研究,および目的について述べている。
第2章では,本研究で対象とした試験体や重錘落下衝撃荷重載荷実験および計測システムの概要について述べている。
第3章では,第2章で示した RC 版の重錘落下衝撃荷重載荷実験とその結果について述べている。
第4章では,数値解析モデルや材料構成則およびその他の解析条件について述べる。
第5章では,低速度衝撃を受ける RC 版の耐衝撃挙動に及ぼす要素分割,材料構成則および減衰定数などの影響を確認するために,第3章で示した代表的な実験結果を対象に弾塑性衝撃応答解析を実施している。得られた結果と実験結果の比較より,RC 版の耐衝撃挙動を適切に再現可能なパラメータを同定した。さらに,提案の解析手法を用いて,衝突速度の異なる4種類の RC 版に対して数値解析を実施し,RC 版の終局状態を数値解析的に推定評価する手法を提案している。
第6,7章では,コンクリート強度および版厚が異なる RC 版を対象に,提案の解析手法を用いた数値解析を実施し,実験結果との比較によって,提案の終局状態推定手法の適用性について検討を行っている。その結果,提案の終局状態推定法は,版厚の薄い場合を除き,コンクリート強度や版厚にかかわらず適用可能であることを示すとともに,RC 版の耐衝撃能力に相当する最大入力エネルギーを安全側で評価できることを明らかにしている。
第8章は本論文の結論であり,本研究で得られた知見を整理している。When a reinforced concrete (RC) slab structure, which is one of the main structural members, is subjected to static and localized concentrated loads, it is well known that the failure mode is the punching-shearing mode. And its design specification has been established. On the other hand, a rational impact-resistant design procedure for the slab under low-velocity impacts such as falling rocks has not yet been established. Therefore, the design specification needs to be smoothly established by efficiently conducting not only experimental but also numerical research studies.
From this point of view, in this study, to establish a numerical analysis method for predicting the ultimate state of the RC slabs under low-velocity impact loading, 3D elasto-plastic dynamic response analyses were conducted for simply supported rectangular RC slabs. First, the effects of the mesh size, constitutive law of concrete, rebar modeling, and damping factor on the dynamic behavior of the RC slab under impact loading were investigated by comparing with experimental results. Then, the appropriate parameters were specified in order to properly simulate the dynamic behavior of the slab. Next, by investigating the experimental and numerical results in detail, a new criterion was proposed to determine the ultimate state of the slab, that is, when the second wave of the impact load disappears and no free-damping vibration is excited in the reaction force and deflection after the unloading. Finally, the applicability and universality of the proposed criterion were investigated by comparing with experimental results for the slabs with various compressive strengths and slab thicknesses.
The results obtained from this study are: 1) when RC slabs before reaching the ultimate state, the circular cracks due to punching shear failure and diagonal cracks on the bottom surface, and 45-degree punching shear failure cracks on the central section surface can be better simulated by using proposed FE model; 2) when RC slabs are in a structurally undamaged state, the second wave of the impact force, the natural vibration state of the reaction force and the deflection, and residual deflection after unloading can be analyzed numerically. Therefore, if these dynamic characteristics cannot be analyzed numerically, then the slab can be evaluated upon reaching the ultimate state, and 3) for RC slab with the height-to-span ratio ≥ 0.1, the maximum impact energy capacity, which may be equivalent to the load-carrying capacity of the RC slabs under low-velocity impact loading, can be conservatively (on the safe side) evaluated by using the proposed criterion.doctoral thesi
鉄筋埋設とアラミド繊維シート巻き立てによる既設鉄筋コンクリート製橋脚の耐震補強法に関する実験的研究
室蘭工業大学Muroran Institute of Technology博士(工学)我が国では,兵庫県南部地震以降,橋梁の耐震補強が精力的に進められているものの,いまだに実施されていない橋梁も多数存在する。本研究では,このような現行の耐震性能を保有しない既設鉄筋コンクリート製(RC)橋脚を対象に,2つのフェーズに着目した耐震補強法を提案し,その妥当性を実験的側面から検討した。まず,地震による被災損傷を受ける以前を想定した耐震補強法として,補強後の橋脚柱部の断面寸法をできるだけ変えないことを念頭に,橋脚柱部のかぶりコンクリートに軸方向鉄筋を埋設し,さらに柱部をアラミド繊維(AFRP)シートで巻き立てる耐震補強法を提案した。軸方向鉄筋比が現行基準よりも小さい縮小模型供試体を対象に正負交番載荷実験を実施し,提案補強法の耐震性能向上効果と妥当性評価を行った。次に,地震によって柱基部に曲げ損傷を受けた橋脚を対象に,その復旧対策としてAFRPシート巻き立てを施した場合の耐震性能検証を行った。柱基部に限定的な損傷を与えたRC橋脚を対象にシート巻き立てを施した縮小模型供試体の正負交番載荷実験を実施し,その妥当性を検証した。本論文の構成は以下の通りである。
第1章では,本研究の背景および目的,既往の研究成果および本論文の構成について述べている。
第2章では,提案する補強法の概要を述べるとともに,本論文における耐震性能の評価方法について述べている。
第3章では,提案補強法を施した供試体の正負交番載荷実験を行い,RC橋脚の塑性変形能を大きく低下させることなく,曲げ耐力の向上が可能であることを明らかにしている。
第4章では,第3章の実験結果をもとに,地震時保有水平耐力法に基づく耐震性能および履歴吸収エネルギーを指標とした性能評価を実施している。検討の結果,提案補強法を適用することにより,保有水平耐力法による耐震性能照査を満たすとともにエネルギー吸収性能が向上することを明らかにしている。
第5章では,被災したRC橋脚の復旧対策として,AFRPシート巻き立てが行われた場合の耐震性能検証を目的として基部に損傷を与えた供試体を対象に正負交番載荷実験を実施した。その結果,塑性変形能が向上することを明らかにしている。
第6章では,第5章の実験結果をもとに,第4章と同様の評価方法を用いて耐震性能評価を行っている。その結果,基部に曲げ損傷を有する橋脚に対してシート巻立てを施すことによって,保有水平耐力法に基づいた耐震性能照査を満足するとともにエネルギー吸収性能が向上することを明らかにしている。
第7章では,各章で明らかになった事項を要約し,本論文を総括している。Seismic retrofitting of bridges that were designed and constructed using old design codes that do not satisfy the current seismic performance is being vigorously advanced. However, there are still many bridges that have not yet been retrofitted. In this study, a seismic retrofitting method focusing on two phases was proposed for such existing reinforced concrete (RC) bridge piers that do not have the current seismic performance, and its validity was investigated from an experimental aspect.
In the first study, a seismic retrofitting method was proposed for the bridge piers before they were damaged by the seismic disaster, in which seismic retrofitting rebars were embedded into the pier surfaces and covered with Aramid Fiber Reinforced Plastics (AFRP) sheets, so that the cross-sectional dimensions of the retrofitted pier columns would not be changed as little as possible. Cyclic loading tests on the reinforced concrete bridge pier model, that has a smaller axial bar ratio than that determined following the current design code applying the proposed method were conducted to evaluate the effectiveness of the proposed method in improving the seismic performance and the validity of the method. From these tests, the following results were obtained: (1) the horizontal seismic strength of the model could be increased without significantly degrading the ductility by applying the proposed method; (2) the energy absorption ability of the model has been improved by applying the proposed method; (3) the model applying the proposed retrofitting method had improved seismic performance and satisfied the current seismic performance verification requirements.
The second study, the seismic performance of RC bridge piers damaged at the column bases by seismic disasters was verified when AFRP sheets were covered after the piers were damaged. Cyclic loading tests on the reinforced concrete bridge pier model, in which that have limited damage and covered with AFRP sheet after damage were conducted to evaluate the effectiveness of the method in improving the seismic performance. From these tests, the following results were obtained: (1) the ductility of the model was improved even when the sheets were covered after damage; (2) the energy absorption ability of the model has been improved; (3) the model that covered with AFRP sheet after damage had improved seismic performance and satisfied the current seismic performance verification requirements.doctoral thesi
Learning IoV in 6G: Intelligent Edge Computing for Internet of Vehicles in 6G Wireless Communications
In sixth-generation (6G) wireless communications,
deep learning will still be essential for the Internet of Vehicles
(IoV). 6G will bring new opportunities and challenges to current
deep learning systems applied in the IoV environment. This
article introduces a new framework named Learning IoV in 6G,
focusing on the deep learning applications for IoV in 6G. We also
apply the proposed framework in a flying base station (FBS) to
evaluate the performance for future IoV in 6G.journal articl
EVALUATION METHOD OF ROAD SURFACE DAMAGE IN WINTER USING DEEP LEARNING AND SALIENCY MAP
積雪寒冷地では,融雪期にポットホールが頻発するため,定期的な道路巡視や網羅的な路面状態把握に基づく計画的な補修が求められている.昨今,AI等の導入による道路巡視の効率化が注目されているが,融雪期の道路を考慮した手法は確立されていない.本研究では,画像認識AIと顕著性マップを用いて,ポットホール発生路面の損傷度を評価する手法を構築した.まず,冬期の路面画像を用いてAIの学習を行ったところ,ポットホールの検出率は85%以上となった.次に,顕著性マップとテクスチャ解析によって,ポットホール部の広がりや深さを考慮した路面損傷評価が可能になることを示した.最後に,定期的な道路巡視やネットワークレベル補修計画を試み,本手法を用いることで,冬期舗装維持管理の効率化および高度化に寄与できることを示した.In snowy and cold regions, potholes occur frequently during the snowmelt season, requiring regular road patrols and systematic repairs based on road surface conditions. In recent years, in-vehicle cameras and AI have been introduced to improve the efficiency of road patrols . However, it is difficult to find a method to evaluate complex and significant road surface damage during the snowmelt season. In this study, we developed a method to detect potholes and further evaluate their road surface damage using image recognition AI (VGG16) and saliency map (Score-CAM). First, our method can detect potholes with an accuracy of more than 85% using AI trained on road surface images taken during the winter season. Next, we showed that the saliency map can evaluate road surface damage by extracting the shape of pothole areas in detail. Finally, we conducted periodic road patrols and planned network-level repairs using the proposed method. As a result, our findings can contribute to the efficiency and improvement of winter pavement maintenance and management.journal articl
静的曲げ圧壊型AFRPシート接着RC梁に緩衝ゴムを設置した場合に関する載荷方法の異なる衝撃荷重載荷時におけるシート破断現象
本論文では,載荷点部に緩衝ゴムを設置した場合における静的に曲げ圧壊型で終局に至るAFRPシート接着曲げ補強RC梁を対象として,衝撃荷重載荷方法の違いが梁の耐荷性状やシートのひずみ応答に与える影響を検討することを目的に,重錘落下衝撃荷重載荷実験を実施した。検討の結果,単一載荷時に終局に至る程度以上の入力エネルギーが作用する場合には,ひび割れが載荷点部に集中してシート破断に至ること,シート破断に至らない程度の入力エネルギー一定繰り返し載荷の場合には,載荷回数にかかわらずシートひずみも一定値に漸近する傾向を示し,シート破断には至らないこと,等が明らかとなった。journal articl