Michiyuki Hirokane

An Integrated Fuzzy Control System for Structural Vibration

Recently, numerous studies of structural control systems of civil structures and infrastructure have been carried out. To develop structural control systems, it is necessary to consider their special features such as complexity, uncertainty, and size. To consider these features, fuzzy theory has been applied to structural control systems. This study proposes an integrated fuzzy active control system based on fuzzy ensemble learning. It combines several fuzzy active control systems and improves structural vibrations caused by earthquakes. The proposed method includes two fuzzy active control systems, a fuzzy ensemble system, and a gating network. In this study, two fuzzy active control systems are constructed by applying particle-swarm optimization. The fuzzy ensemble system assigns a performance grade to each fuzzy active control system according to control effects from input patterns. The gating network determines the final control force based on the weight of their performance grade. By introducing fuzzy ensemble learning, the structural response is reduced more than when the response is controlled by individual fuzzy active control systems.

Extraction of minimum decision algorithm using rough sets and genetic algorithms

In civil engineering, it is crucial to reuse knowledge which has been accumulated through the experience of engineers, etc. For this purpose, it is necessary to establish a method for knowledge acquisition and a method for explicit representation of the acquired knowledge. This paper applies the genetic algorithm to the process of deriving a decision algorithm from instances by using rough sets, and proposes a method of deriving a simple and useful decision algorithm with a relatively small amount of computation. A decision algorithm is actually derived from the data on accident instances at actual construction sites, and the recognition rate and other performance measures are investigated by the k-fold cross validation method.

Extraction Method Based on Rough Set Theory of Rule-Type Knowledge from Diagnostic Cases of Slope-Failure Danger Levels

The failure of slopes caused by heavy rains in the typhoon and baiu seasons has taken tolls of lives and done damages all over country [1] [2]. To prevent such disaster, it is necessary to take appropriate measures against slope failure at appropriate spots at appropriate times. Preventive works of such structures as are capable of preventing slope failure have to be chosen, slopes surveyed thoroughly, geology, soil properties, costs of civil engineering works, scenery, and so on taken into account [3]. Slopes requiring such works and their danger levels of failure should be evaluated synthetically, and appropriate spots and times of such works should be determined Accordingly, it is important to estimate the danger levels of slope failure with accuracy of a certain degree, which also helps to minimize the damages by slope failure.

Life-Cycle Cost Based Maintenance of Infrastructure Systems Considering Prediction Uncertainties

In order to establish a rational maintenance program, it is necessary to predict the future changes of structural characteristics and environments. Considering the prediction uncertanties of the structural capacity and environments, this paper attempts to make and develop a method that can provide us with several practical maintenance plans. Applying the genetic algorithm developed by Tamaki to the maintenance planning problem, it is possible to obtain practical solutions that are robust to the variety of structural and environmental conditions. Some numerical examples are presented to demonstrate the applicability of the method developed in this paper.

Rough‐Set‐Based Knowledge Acquisition from Cases for Integrity Assessment of Bridge Structures

Many existing structures present an ever-increasing demand for maintenance and repair. It is necessary in the first place to diagnose such structures correctly. The diagnoses have been dependent to date on experts of great experience. To meet the increasing demand for maintenance and repair and perform diagnoses efficiently and correctly, the development of expert systems is hoped for. This article discusses application of rough-set theory to the acquirement of experiential knowledge contained in diagnostic cases. Rough-set theory was applied to cases in which experts diagnosed the damage of bridges, and a minimal-decision algorithm was derived that could make diagnoses equivalent to those of the experts. In addition, a method of deriving rules from the minimal-decision algorithm for the construction of expert systems is discussed.

Adaptive vibration control with predicting and learning abilities based on soft computing

This study is intended to improve the adaptive and learning abilities of structural vibration control by introducing predictive functions that use structural information of external forces and relative response velocities. This proposed method can renew the fuzzy control rules one step ahead of ordinary methods because predicted information is used in self-tuning of fuzzy rules. Based upon the modified control rules, the proposed method more precisely generates control forces to the structure and reduces structural vibration. Finally, the applicability and accuracy of the proposed method are discussed through comparison of several control methods.

Experiment of Disaster Prevention Education and Evaluation Using a Torrential Rain Simulated Experience Application

In the field of transport safety as represented by automobiles, aircraft, etc., risk prediction training using CG or applications imagining scenarios linked to accidents is being actively implemented as training for the purpose of avoiding accidents. In this kind of education for improving sensitivity toward risk, simulated experience is said to be effective. Therefore, in this research, we developed an application using augmented reality to enable re-creation of a series of rainfall patterns existing before and after the occurrence of weather disasters that occurred in the past. We used the application developed to enable audiovisuals of rainfall during disasters, and elicited a response to the timing and state of damage when actual damage occurred, to propose a disaster education framework capable of enabling evacuation action at an earlier stage. Furthermore, in disaster prevention training held by Takatsuki city, we put into practice the proposed disaster prevention education framework targeting the citizens, and used the Human Cognitive Reliability model (HCR) to verify the relationship of sensitivity to risk and evacuation action.

Study on Monitoring Method Using a Wrist Watch Type Device for Unsafe Behavior

If its construction business, its being talked about that there are most dead accidents in all the industries, and establishment of effective safety training is wished for. Various devices such as the occasion in which steel fell actually at a construction site and the interactive safety training which assumed the case when a worker fell are taken in it. But there is no end to accident by worried whole behavior caused by oversight, an attentiveness lack, a dangerous slight and accustoming. So the mechanism to aim at concentration to workers work by this research and monitor the worried whole behavior which tends to happen at the time of non-concentration using a simple device was considered. Specifically, the state that a wrist type device with a sphygmograph was loaded It was so and I worked and acquired time series data of a pulse in real time. I calculated the feature quantity of the integration value and the finite difference value from acquired data, inspected about concentration to work and relationship with a pulse based on its value and considered the monitoring technique of the pulse pattern when not concentrating.

Diagnosis for Axial Force of High-Strength Bolts Using Pattern Recognition

Currently, the bridges which were built in period of high economic growth are accounting for about 40% of all bridges and the deterioration of the high-strength bolts has become the important problem. It is necessary to diagnose periodically in order to find the deterioration of the high-strength bolts due to corrosion and metal fatigue and so on. The impact acoustics method is one of the diagnosis methods of the axial force. However, it is required for the method that can diagnose more accurately and efficiently because the impact acoustics method has problems which depend on the engineer’s experience and intuition, require the high costs for the diagnosis and lack in well-skilled engineers. In this study, we collected many wave pattern data of the impact acoustics that were taken by hitting a body which was attached to the high-strength bolts with the tapping hammer and recognized the axial force by using the frequency data that were taken by converting the wave pattern data through the pattern recognition method.

Automatic detection of bicycle direction using RealAdaBoost and C4.5

In recent years, there has been a decrease in the number of traffic accidents and deaths due to the improved vehicle safety and legislation related to traffic violations. However, the number of injuries is still large, with injuries incurred by bicyclists and pedestrians accounting for approximately 25% of the total number of injuries. Measures must be taken to ensure the safety of pedestrians and bicyclists, which is an important issue. This study, which proposes a system for automatically detecting the direction in which a bicycle is moving, is a contribution to the development of an overall support system for safe automobile driving.