Tatsumasa Kaita

Prediction of residual strength of corroded tensile steel plates

Recently, the damage due to deterioration of materials in many old steel bridge structures, which causes unavoidable strength reduction, is becoming a serious problem in Japan and all over the world. Therefore the remaining load-carrying capacities must be carefully evaluated in order to understand the feasibility of those steel structures for the current usage and to evaluate the necessity of retrofitting of selected corroded members to strengthen the existing structure. This paper proposes a new method to calculate the remaining yield and tensile strengths by using a concept of representative effective thickness (teff) with the correlation of initial thickness (t0) and the maximum corroded depth (tc,max), based on the results of many tensile coupon tests of actual corroded plates. Further, the feasibility of establishing of an analytical methodology to predict the residual strength capacities of a corroded steel member with fewer number of measuring points is also discussed.

Structural Appraisal-based Different Approach to Estimate the Remaining Fatigue Life of Railway Bridges

The current fatigue life estimation methods of railway bridges are generally based on combinations of measured strain histories, Miner’s rule, and railway code provided fatigue curve. Even though the past measured strain histories are available for major bridges, most of the old bridges do not have past strain measurements. Furthermore, in case of existing railway bridges where the detailed loading history is known, Miner’s rule might provide incorrect results because of its omission of load sequence effect. These reasons hinder the usage of current methods to estimate the realistic remaining fatigue life of most of the existing railway bridges. Therefore, a new method is presented in this article to estimate remaining fatigue life of riveted railway bridges. The method mainly consists of predicted stress histories, recently developed sequential law and fully known Wöhler curve. Here, it is essential to use the fully known Wöhler curve as the related fatigue curve. Hence, the technique, which utilizes transfer of the partially known Wöhler curve to fully known curve, is also discussed in this article. Further, this method describes a reasonably accurate procedure to obtain the past stress histories from present day measured strains. Initially, the article describes the proposed method for remaining fatigue life estimation. Then the sequential law and associated fully known S-N curve determination procedure are verified by comparing theoretical fatigue life, damage behavior with real fatigue life and damage behavior of few materials. Finally, the remaining fatigue life of an existing railway bridge is estimated. Hence, validity and merits of the proposed method is confirmed by comparing the results with previous method-based fatigue lives.

Development of an Efficient Maintenance Strategy for Corroded Steel Bridge Infrastructures

The potential for structural capability degrading effects caused by corrosion is of profound importance and must be both fully understood and reflected in bridge inspection and maintenance programs. As the number of steel bridge infrastructures increases throughout the world, it is an exigent task to conduct regular and detailed corrosion surface investigations to evaluate their residual strength capacities and to develop analytical models to understand their current conditions and critical locations, as well as yield and ultimate behaviors. This paper presents a simple, accurate, and rapid assessment method and an effective maintenance management strategy developed by using the results of tensile coupon tests conducted on numerous corroded plates obtained from a steel plate girder used for about 100 years with severe corrosion and an finite element method (FEM) analytical approach proposed by measuring only the maximum corroded depth, which can be used to make reliable decisions affecting cost and safety.

Development of Analytical Method for Predicting Residual Mechanical Properties of Corroded Steel Plates

Bridge infrastructure maintenance and assurance of adequate safety is of paramount importance in transportation engineering and maintenance management industry. Corrosion causes strength deterioration, leading to impairment of its operation and progressive weakening of the structure. Since the actual corroded surfaces are different from each other, only experimental approach is not enough to estimate the remaining strength of corroded members. However, in modern practices, numerical simulation is being used to replace the time-consuming and expensive experimental work and to comprehend on the lack of knowledge on mechanical behavior, stress distribution, ultimate behavior, and so on. This paper presents the nonlinear FEM analyses results of many corroded steel plates and compares them with their respective tensile coupon tests. Further, the feasibility of establishing an accurate analytical methodology to predict the residual strength capacities of a corroded steel member with lesser number of measuring points is also discussed.

Condition assessment of multispan masonry arch bridges

A reliability-based condition assessment procedure of multispan masonry arch bridges is presented in this paper. Safety margins (limit state functions) are introduced for each arch of the bridge, with the authors considering axle load as assessment criterion. Two variables constitute the introduced safety margins: provisional axle load (PAL) which is estimated using MEXE methods, and actual axle load (AAL), which is estimated using weigh-in-motion measurements of the bridge. That both variables follow log normal distribution is assumed. Failure probabilities of each arch (estimated from statistical parameters of variables) then are combined to get the failure probability of the bridge using reliability bounds. The bridge reliability index is estimated from the failure probability. Bridge condition prediction is performed through comparison of the reliability index with its acceptable reliability index. Further, corresponding reliability index changes are discussed when the statistical parameters behave as interval numbers. The introduced assessment procedure is illustrated by a Sri Lankan four span brick masonry arch bridge.