• International Journal of Reliability and Applications
• /
• v.4 no.1
• /
• pp.13-26
• /
• 2003

This paper presents a comparative study of a few commonly used maintenance scheduling methods for small utilities that consists solely of thermal generating plants. Two deterministic methods and a stochastic method are examined. The deterministic methods employ the leveling of reserve capacity criterion, of which one uses a heuristic rule to level the deterministic equivalent load obtained by using the product of the unit capacity and its corresponding forced outage rate. The stochastic method simulates the leveling of risk criterion by using the peak load carry capacity of available units. The results indicate that for the size and type of the maintenance scheduling problem described In this study, the stochastic method does not produce a schedule which is significantly better than the deterministic methods.

• Steel and Composite Structures
• /
• v.2 no.3
• /
• pp.171-194
• /
• 2002

In this paper, the behaviour of axially loaded partially encased composite columns made with light welded H steel shapes is examined using ABAQUS finite element modelling. The results of the numerical simulations are compared to the response observed in previous experimental studies on that column system. The steel shape of the specimens has transverse links attached to the flanges to improve its local buckling capacity and concrete is poured between the flanges only. The test specimens included 14 stubcolumns with a square cross section ranging from 300 mm to 600 mm in depth. The transverse link spacing varied from 0.5 to 1 times the depth and the width-to-thickness ratio of the flanges ranged from 23 to 35. The numerical model accounted for nonlinear stress-strain behaviour of materials, residual stresses in the steel shape, initial local imperfections of the flanges, and allowed for large rotations in the solution. A Riks displacement controlled strategy was used to carry out the analysis. Plastic analyses on the composite models reproduced accurately the capacity of the specimens, the failure mode, the axial strain at peak load, the transverse stresses in the web, and the axial stresses in the transverse links. The influence of applying a typical construction loading sequence could also be reproduced numerically. A design equation is proposed to determine the axial capacity of this type of column.

• Computers and Concrete
• /
• v.33 no.5
• /
• pp.545-557
• /
• 2024

A real-case incident occurred where a 9-meter-high segment of a pre-fabricated concrete separation wall unexpectedly collapsed. This collapse was triggered by improperly depositing excavated soil against the wall's back, a condition for which the wall segments were not designed to withstand lateral earth pressure, leading to a flexural failure. The event's analysis, integrating technical data and observational insights, revealed that internal forces at the time of failure significantly exceeded the wall's capacity per standard design. The Lattice Discrete Particle Model (LDPM) further replicates the collapse mechanism. Our approach involved defining various parameter sets to replicate the concrete's mechanical response, consistent with the tested compressive strength. Subsequent stages included calibrating these parameters across different scales and conducting full-scale simulations. These simulations carried out with various parameter sets, were thoroughly analyzed to identify the most representative failure mechanism. We developed an equation from this analysis that quickly correlates the parameters to the wall's load-carry capacity, aligned with the simulation. Additionally, our study examined the wall's post-peak behavior, extending up to the point of collapse. This aspect of the analysis was essential for preventing failure, providing crucial time for intervention, and potentially averting a disaster. However, the reinforced concrete residual state is far from being fully understood. While it's impractical for engineers to depend on the residual state of structural elements during the design phase, comprehending this state is essential for effective response and mitigation strategies after initial failure occurs.

• Journal of Advanced Marine Engineering and Technology
• /
• v.14 no.4
• /
• pp.57-66
• /
• 1990

In the design of an electric power plant, the capacity to meet the peak load demand is one of the important factors to be considered. This peak load usually occurs when the most of the cooling air conditioning systems are being operated during daytime in summer season, which inevitably entails the construction of an additional electric power plant. This study is aimed to carry out a basic experiment for the development of a cooling air conditioning system using the ice energy by the surplus electric power during the night-time. The experimental apparatus consists of four major parts; (1) the heating section consisting of the air duct and I.D. fan, (2) the cold section with the ice chamber, (3) the bundle of heat pipes made in a form of the staggered arrangement with ${C_y}/{d_o}$=2.0 and ${C_x}/{d_o}$=1.73, (4) the refrigerator system to cool down the ice chamber. This study involves an intensive experiment concerning the convective heat transfer of the air flow surrounding the bundle of heat pipes. This major experimental parameters are the amount of working fluid, the velocity of air and the working temperature. The major findings of the present study are as follows; (1) The optimum amount of the working fluid necessary for the horizontal heat pipes is much more than that for the vertical type. (2) The convective heat transfer coefficients of the air are coincided with the empirical equations of Grimson and ${\breve{Z}ukauskas}$. (3) The equation of the mean heat transfer coefficient obtained in the present study is ${N_um}=0.32 {Re_max^{0.63}}$.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6C
• /
• pp.395-406
• /
• 2006

Drilled shafts are a common foundation solution for large concentrated loads. Such piles are generally constructed by drilling through softer soils into rock and the section of the shaft which is drilled through rock contributes most of the load bearing capacity. Drilled shafts derive their bearing capacity from both shaft and base resistance components. The length and diameter of the rock socket must be sufficient to carry the loads imposed on the pile safely without excessive settlements. The base resistance component can contribute significantly to the ultimate capacity of the pile. However, the shaft resistance is typically mobilized at considerably smaller pile movements than that of the base. In addition, the base response can be adversely affected by any debris that is left in the bottom of the socket. The reliability of base response therefore depends on the use of a construction and inspection technique which leaves the socket free of debris. This may be difficult and costly to achieve, particularly in deep sockets, which are often drilled under water or drilling slurry. As a consequence of these factors, shaft resistance generally dominates pile performance at working loads. The efforts to improve the prediction of drilled shaft performance are therefore primarily concerned with the complex mechanisms of shaft resistance development. The shaft resistance only is concerned in this study. The nature of the interface between the concrete pile shaft and the surrounding rock is critically important to the performance of the pile, and is heavily influenced by the construction practices. In this study, the influences of asperity characteristics such as the heights and angles, the strength characteristics and elastic constants of surrounding rock masses and the depth and length of rock socket, et. al. on the shaft resistance of drilled shafts are investigated from elasto-plastic analyses( FLAC). Through the parametric studies, among the parameters, the vertical stress on the top layer of socket, the height of asperity and cohesion and poison's ratio of rock masses are major influence factors on the unit peak shaft resistance.