Scientia Iranica
Volume:18 Issue: 6, 2011

Intensity Measure (IM), characterizing the strength of an earthquake ground motion, is used to predict the seismic responses of a structure. Recent studies have demonstrated that using Sa(T1) as the seismic scaling index for near-fault ground motions may introduce large variability in the estimated seismic demands. To develop an improved IM, this paper investigates the effectiveness of a large number of combinations of spectral values in reducing scatter in the estimated maximum interstorey drift ratios. For this purpose, Incremental Dynamic Analyses (IDAs) are performed for five generic frames of short-to-relatively long periods, under 40 pulse-like earthquake records. Statistical evaluation of the IDA results reveals that the optimal combination of spectral displacements, compared to that of spectral velocities, leads to less variability for all ranges of the frames. It is concluded that the optimal combination of spectral displacements can be used to predict seismic demands with dispersions less than those of Sa(T1), and which are closer to constant for various intensity levels. Additionally, with respect to the importance of collapse capacity predictions in earthquake engineering, an efficient IM that can considerably improve the accuracy of estimating the probability of collapse is also developed.

The pseudo symmetric subspace iteration method is an efficient technique for computing mode shapes of coupled fluid-structure systems. This approach handles the related unsymmetric eigenvalue problem in a manner that is similar to solving a symmetric eigen-problem. Moreover, it employs a symmetric active column solver in the core of its routines, which makes it very efficient. In the present study, an enhanced and accelerated version of this technique is described, which is ideal especially for extracting a large number of eigen-triplets (e.g. greater than or equal to 40). The accelerated approach is implemented in a special purpose finite element program, and several practical numerical examples are presented herein. Based on these experiences, it is concluded that the accelerated pseudo symmetric subspace iteration method is an extremely efficient approach. This is estimated to be about 3 to 11 times more efficient than its original basic counterpart version.

The effect of response feedback on designing optimal controllers for nonlinear frames has been studied. Different combinations of response feedback have been used in the performance index. The Newmark based nonlinear instantaneous optimal control algorithm has been used as the control algorithm in controlling the response of an eight-story bilinear hysteretic frame subjected to white noise excitation and real earthquakes, and controlled by either eight actuators or a single actuator. While the objective has been to minimize the maximum control force for reducing the maximum drift to below the yielding level, the distributed genetic algorithm (DGA) has been used to determine the proper set of weighting matrices in the performance index. Results show that the performance of the active control system depends on the combination of response feedback, where the velocity feedback has been more effective than acceleration and displacement. Also, although using the full feedback of response in the performance index leads to the design of optimal controllers that require the smallest control force, it is costlier, because it requires more online measurements. Finally, it has been concluded that amongst all possibilities, using only velocity feedback can provide the best results regarding the maximum required control force and online measurement, simultaneously.

Shallow landslides in nearly saturated uncohesive to slightly cohesive soils are triggered by high intensity, short duration rainfall which infiltrates into soil and changes intergranular friction and effective stresses. For this, the especially developed Soil–Water Interaction Modelling System (SWIMS) was used with CL-ML type soils. For simplicity, rainfall intensity and duration were kept constant. Results showed that (1) All 35° slopes were failed by translational failure. For the other (15°,25°) slopes, no failures were observed; (2) For all slopes, FOS increased with increasing compaction degree and decreased with increasing slope angle; (3) Other parameters, such as soil density, porosity, saturation degree, water contents, and water permeability may also affect shear strength/slope stability, especially for low degrees of saturation (S<95%), compared to high degrees of saturation (S=,>95%). (4) A correlation of SWIMS tests observed that average wetting band depths, with the calculated wetting band depths from the Lump Equation (hLE), were poor, as values were much higher than hLE values. Differences increased for very low degrees of saturation (S), compared to S>95%. This meant that the Lump equation underestimated wetting band depths. Further, if the Lump equation is still considered valid, this would imply either water-permeability increases, porosity decreases or both occur towards full saturation; a process where the last possibility is the most probable occurrence.

This paper presents formulation of a Hybrid Frequency-Time Domain (HFTD) method for the solution of nonlinear ground response problem. Using non-recursive matrix approach, the displacements, caused by the bedrock acceleration, at the surface of layered soil are calculated while pseudo-forces due to nonlinear behavior of soil are obtained iteratively using an appropriate transformation scheme. The analysis process is continued until the pseudo-forces reach negligible values. Two different ground motions recorded at Gilroy 2 reference site during the Loma Prieta earthquake of 17 October 1989 and Coyote Lake earthquake of 6 August 1979 are analyzed using proposed method. The results are in good agreement compared to field observations. Furthermore, the two input motions are analyzed with SHAKE (Equivalent Linear Method) and NERA (Non-linear Method) programs and it is shown that the proposed method provides better predictions when compared with the field observations.

This paper presents the measured behavior of small-scale single rammed aggregate piers as a function of the piers’ slenderness ratio. For this purpose, loading tests at the site were carried out on two groups of single rammed aggregate piers with a constant diameter of 135 mm and variable lengths of 350–1000 mm, and two groups with a constant length of 1000 mm and various diameters of 105–185 mm. The testing area consisted of relatively uniform saturated soft alluvial clay overlain by a 1-m-thick wet soft-to-stiff silt layer. Results show that when length and diameter change, the pier load and top settlement variations at the design limit, in terms of the slenderness ratio, are not in the same direction while other design limit parameters’ variations are. The variations of design limit parameters in the two modes of change to pier length and diameter including the applied load, top settlement, stiffness modulus and pier modulus, in terms of the slenderness ratio, make a linear function while the variations of load and settlement ratio show an exponential function. Interpretations of the test results are particularly focused on the load-settlement behavior and variations of design limit parameters as a function of pier slenderness ratios.

This study introduces a new application of GMDH in the prediction of scour depth around a vertical pier. Two models of the GMDH network were developed using genetic programming and a back propagation algorithm. Genetic programming was performed in each neuron of the GMDH instead of performing the quadratic polynomial. In the second model of the GMDH, the quadratic polynomial was used in each neuron of the network as a transfer function, and a back propagation algorithm was used for training of the network. Six effective parameters including pier diameter, flow velocity, flow depth, medium diameter of bed material, standard deviation of bed grain size and fluid dynamic viscosity were considered for prediction of the scour depth. Results of two GMDH networks were compared with results of several traditional equations. From result performances, it was found that although GMDH-GP is very time-consuming and more complicated, this proposed method has provided a better prediction of scour depth than GMDH-BP in training and testing stages. A sensitivity analysis was performed for the GMDH-GP model and the results indicated that the ratio of pier diameter to flow depth is the most significant parameter regarding scour depth. In particular, the GMDH network proved very effective in comparison with traditional equations.

Conformal Prediction (CP) is a novel machine learning concept which uses past experience to determine precise levels of confidence in new predictions. Traditional machine learning algorithms, such as Neural Networks (NN), Support Vector Machines (SVM), etc. output simple, bare predictions without an indication (confidence) of how likely each prediction is of being correct. However, in real-world problem domains, it is highly desirable to have predictions complemented with a tolerance interval to assess their credibility and accuracy. CP can be thought of as a strategy built on top of a machine learning algorithm to hedge its predictions with valid confidence. In this paper, a NN Regression (NNR) model is used to derive a decision rule for the inverse prediction of non-linear pavement layer moduli from Non-Destructive Test (NDT) deflection data. A CP is then implemented for the NNR decision rule and tested on an independent data set to demonstrate its error calibration properties. It is shown that CP can be used to derive reliable pavement moduli predictions without compromising the accuracy of the NNR decision rule but with control of the risk of error.

Globally, the Road Safety Audit (RSA) concept has been recognized as an effective tool in examining the crash potential of in-service and future roadways in planning and design stages. As such, there is critical need for a practical tool that focuses on the safety of the existing, as-built, local road facilities. As requested by the World Bank, the RSA process has been developed for this purpose, giving specific recognition to the safety performance of roadside and median barriers for three existing, typical freeways in Jiangxi, China, and to the gathering of design experience for a new RGF project. On top of a routine road safety audit process, a total of 172 roadside crashes are collected, with 74 belonging to single vehicle Run-off-Road crashes, and crash records are analyzed to supplement the qualitative auditing suggestions. The structure and safety performance of the roadside and median barriers in the three freeways reviewed are evaluated and compared with their counterparts in the US barrier system. Several critical issues were identified and improvement suggestions were recommended, including less attention being paid to the roadside Clear Recovery Zone (CRZ), weak barrier structures, un-protected roadside obstacles (i.e. barrier ends, advertisement signs, drainage ditches, etc.), and poor connections or transitions of rails. Based on these observations, detailed pertinent countermeasures for each issue have been suggested in a roadside safety audit report for guiding roadside safety design in the RGF project.