著書

Structural health monitoring (SHM), which evaluates building damage based on seismic response data, is useful for rapidly assessing damage due to earthquakes. If a reliable damage assessment system for houses can be developed, the loss of human lives caused by the aftershocks of major earthquakes can be reduced. When a target building is damaged by strong seismic ground motions, the shape of the restoring force–story drift angle curve (hysteresis loop) changes. Therefore, based on the hysteresis loop images of wooden buildings, this study proposes a method to classify the risk of collapse because of future seismic ground motions using a convolutional neural network (CNN). To develop training data, structural response analysis is conducted using “wallstat”, which is a numerical analysis software for wooden structures. SHM has more severe consequences when it misclassifies an unsafe condition as safe (miss) than when it misclassifies a safe one as unsafe (false alarm). Therefore, we propose the usage of a dataset that does not contain safe model data near the threshold value of major damage for training to decrease the possibility of misses, and we verify the effect of this. In creating the hysteresis loop images, we examine the range of the axes using two methods: fixing it at specific values and matching it to the range of the minimum and maximum values. The investigation results suggest that the trained CNN learns the changes in structural characteristics, such as stiffness reduction, damping increase, and residual displacement.

There are few published studies on torsional response of a symmetric structure, possibly owing to a low perceived risk. However, a symmetric structure could experience torsional vibration due to the torque generated by the inertial force (Q) perpendicular to the horizontal displacement (∆), because of geometric nonlinearity. The authors’ research group has called this the “Q–∆ effect,” and there are multiple resonance conditions of this torsional vibration. If a building satisfies one of these resonance conditions, a strong ground motion in two translational directions could induce the torsional response. In this study, we conducted an experiment to validate this Q–∆ resonance using a symmetric single-layer specimen with no eccentricity. First, resonance conditions were derived for the setup of the validation experiment. The equations of motion and vibration were based on our previous studies. In this study, we considered the input ground motion of two superposed sine waves, with frequencies equal to the two translational modes’ natural frequencies. With respect to the torsional response, we derived four resonance conditions for the torsional mode’s natural frequency. A specimen consisting of a slab, four columns, and two weights on the slab was fabricated focusing on one of the four resonance conditions in the torsional mode. The natural frequency in the torsional mode could be changed by moving the two weights on the slab. Adjusting the moment of inertia around the vertical axis passing through the center of gravity of the slab, ten patterns of natural frequencies were obtained from the one specimen. To understand the specimen’s structural and vibration characteristics, we conducted horizontal loading and free vibration experiments. All the experiments were conducted in two translational and one torsional directions. The vibration characteristics of the specimen were identified from the results of the free vibration experiments. Using a uniaxial shaking table, we conducted shaking table experiments, in which the input acceleration was two superposed sine waves. We first verified that the torsional response was very small when the specimen was excited in translational mode directions. In addition, we verified that the vibration characteristics of the specimens did not change after shaking table tests for each pattern of the specimen configuration. The validation experiments were performed by setting the specimen in the 45-degree direction from the principal axis (translational mode) direction on the shaking table. The translational response even when the natural period in the torsional mode was changed, the. The amplitude of the torsional response, however, did change, and it increased further when the torsional mode’s natural frequency was closer to the theoretically predicted resonance frequency. This result confirms that the torsional vibration of a symmetric structure is excited by an inertial force perpendicular to the horizontal displacement. However, the experimental result was larger than the simulated response based on a theoretical model. Future studies will use a specimen more suitable for this analysis and revise the prediction models to account for the second-order mode.

The special investigation committee of Architectural Institute of Japan (AIJ) proposed a Resilience Performance Index (RPI) and a Business Continuity Plan Level Rating (BCPLR) of Buildings. RPI is a new approach to quantify the resilience performance of buildings. The proposed BCPLR classifies building resilience performance into four levels from the viewpoint of BCP. The committee also investigates how to disseminate the proposed approaches in order to promote BCP activities in various sectors. This paper addresses the investigation results by the AIJ committee. Firstly, the outline of the proposed RPI and BCPLR of buildings is introduced. RPI is defined as the degree of recovery of a building up to a certain period of days, which is expressed by a set ratio of usable floor area. RPI is an index for resilience performance. The assessment result is presented as, for example, 7 days to recover to 90% resilience performance. BCPLR is an indicator of the overall level of resilience performance of buildings based on the recovery time, which is a key factor in considering BCP, obtained from RPI. The result of assessment is presented as the number of stars such as three stars; 90% functional recovery within one week: two stars; 90% within one month: one star; 90% within six months: and no star; more than 6 months. Secondly, we examine existing systems that are expected to improve the seismic performance of buildings and/or provide incentives for disseminating BCP. We clarify the following typical incentives: 1) provision of risk information, 2) third-party certification, and 3) financial incentives, e.g., tax incentives, compensation, and penalties. Thirdly, we conduct interviews with relevant organizations and companies to identify the potential use of the proposed RPI and BCPLR. The results of the interviews confirm the importance to disseminate them to society more in order to utilize them. In particular, it is pointed out that concrete methods to improve the performance of buildings need to be explained in an easy-to-understand manner and their visibility in society need to be increased. We also investigate the applicability of RPI and BCPLR in the non-life insurance market in terms of financial incentives. The results show that, despite the challenges, lower premiums could be offered for buildings with higher resilience performance. Finally, based on the above investigation results, a leaflet is published to make better understanding for RPI and BCPLR in order to promote BCP activities.

日本建築学会建築にかかわる社会規範・法規範特別調査委員会建築物の安全性ガイドライン特別調査小委員会の活動成果を取りまとめ出版。多くの地震リスク評価手法を比較し解説するとともに地震リスクコミュニケーションの理論と方法を紹介。1.2節耐震安全性にかかわる指標と資料をどのように提示するかの執筆と１章全体の編集を担当。

日本建築学会荷重運営委員会信頼性工学利用小委員会が，建築工学における確率・統計の利用の普及活動の一環として出版。地震リスク評価について，いくつかの例題をExcel を使って解くことで，ひととおり学べる内容となっている。４章フラジリティと損傷度の評価の執筆を担当。

日本建築学会の2015年版「建築物荷重指針・同解説」を活用し性能設計を行うため、設計の流れや設計例、詳細な解説を記載。１章「性能設計の実現に向けて」の1.1節「目標性能水準設定の基本的な考え方と設計者の役割」の執筆を日本女子大学平田京子先生と共同で担当。

日本建築学会による，建築物の構造設計の際に用いる荷重について，基本理念や考え方を示した指針。１章総則の1.2節基本概念と1.3節用語，２章荷重の種類と組合せの2.1節荷重の種類，2.2節荷重の基本値，2.3節荷重の組合せと要求性能の執筆を担当。

現代制御理論は電気、機械など工学の様々な分野で応用されており、動的システムの振る舞いを状態方程式で表現することで、内部状態の理解と制御系の設計の見通しが格段に良くなるメリットがある。建築構造物の振動制御においても、制御系設計を行うに当たり非常に強力なツールとなる。一方、現代制御理論の理解には高度な数学的知識が必要であるため、本書はプログラムを動かして計算し、結果が理論どおりになることを確かめながら学ぶことができるよう作られた。プログラミング言語としてPythonの容易さとC・Fortran並みの処理速度を兼ね備え、AI、データサイエンス、科学技術計算分野で需要が高まっているJuliaを用いている。内容は、運動方程式を状態方程式に変換する手順から始まり、構造物の振動応答を理解する上で欠かせない複素固有値解析、周波数領域および時間領域の応答解析の理論等を解説している。そして、構造物の振動制御設計の事例も多く紹介し、理論以外の観点で考慮しなければならないポイントも詳しく解説している。 1章・4章・付録を東北大学五十子幸樹教授と共同で執筆。

日本建築学会住まいづくり支援建築会議情報事業部会が、住宅を建てたい、あるいは買いたいという人に、住宅に対する正確な知識と理解を深めてもらうため、情報普及活動の一環として出版。マンションについて、そして建物を建てる、選ぶ、住むことについて、建築にかかわる専門家の視点と学会という中立的な立場から、メリット・デメリットの双方を紹介。

In this chapter, we present an overview of the E-Simulator for application to seismic response analysis of building structures. Accuracy and computational performance of simulation using the E-Simulator are discussed through examples of seismic response analysis of a four-story steel frame, buckling analysis of a column, and simulations of static cyclic responses of a composite beam and an exterior wall. Results of seismic response analysis with fixed base as well as soil-structure interaction analysis are presented for a high-rise building frame. Computational performance using the K computer is also discussed.