FAST-NEPAL: Regionally Calibrated Spectral Method for Reinforced Concrete With Masonry Infills

Reinforced concrete (RC) with masonry infill is one of the most common structural typologies in Nepal, especially in the Kathmandu Valley. Masonry infills are typically made of solid clay bricks produced locally in Nepal. This study aims to calibrate the spectral-based analytical method, namely, FAST, for Nepalese RC-infilled buildings. The FAST method has been initially conceived for Southern European RC buildings with hollow clay brick infills. The calibration is achieved by reviewing code prescriptions and construction practices for RC masonry infills in Nepal and updating the FAST method. The variables of FAST method are calibrated using different information sources and a Bayesian updating procedure to consider the global and local material properties for solid clay bricks. The FAST-NEPAL method obtained is then verified, considering a single school design, for which a detailed state-of-the-art vulnerability assessment is available. Being particularly suitable for large-scale assessment, the method is further validated using data from Ward-35 of Kathmandu Metropolitan City (in the vicinity of Tribhuvan International Airport) obtained from photographic documentation included in a geo-referenced database of buildings collected after the 2015 Nepal earthquake and prepared for census purposes. The comparisons show that the FAST-NEPAL method can be conservative relative to the other data sources for vulnerability and is more accurate at capturing low-level damage. This makes the approach suitable for large-scale preliminary assessment of vulnerability for prioritisation purposes.

An EM Algorithm for Target Tracking with an Unknown Correlation Coefficient of Measurement Noise

In this paper, an expectation maximization based sequential modified unbiased converted measurement Kalman filter is proposed for target tracking with an unknown correlation coefficient of measurement noise between the range and the range rate. Firstly, a pseudo measurement is constructed by multiplying the range and the range rate to reduce the strong nonlinearity between the measurement and the target state. The mean and covariance of converted errors are subsequentlsubsequently derived by modified unbiased converted measurement to weaken the error caused by the linearization of the measurement equation, which is effectively to improve the dynamic accuracy of target tracking. Then, the converted errors of the position and the pseudo measurement are decorrelated by the Cholesky factorization and thus to obtain the posterior probability distribution of the state by using the sequential filtering in the Bayesian framework. Finally, the expectation maximization is introduced in the updating procedure of the pseudo measurement to jointly estimate the target state and the correlation coefficient. The target tracking scenario with an unknown correlation coefficient is built to demonstrate the validness and feasibility of the proposed algorithm. Simultaneously, the results of the normalized error squared validate the consistency of the modified unbiased converted measurement.

Joint Modelling Approaches to Survival Analysis via Likelihood-Based Boosting Techniques

Joint models are a powerful class of statistical models which apply to any data where event times are recorded alongside a longitudinal outcome by connecting longitudinal and time-to-event data within a joint likelihood allowing for quantification of the association between the two outcomes without possible bias. In order to make joint models feasible for regularization and variable selection, a statistical boosting algorithm has been proposed, which fits joint models using component-wise gradient boosting techniques. However, these methods have well-known limitations, i.e., they provide no balanced updating procedure for random effects in longitudinal analysis and tend to return biased effect estimation for time-dependent covariates in survival analysis. In this manuscript, we adapt likelihood-based boosting techniques to the framework of joint models and propose a novel algorithm in order to improve inference where gradient boosting has said limitations. The algorithm represents a novel boosting approach allowing for time-dependent covariates in survival analysis and in addition offers variable selection for joint models, which is evaluated via simulations and real world application modelling CD4 cell counts of patients infected with human immunodeficiency virus (HIV). Overall, the method stands out with respect to variable selection properties and represents an accessible way to boosting for time-dependent covariates in survival analysis, which lays a foundation for all kinds of possible extensions.

Data-Driven Model Updating of an Offshore Wind Jacket Substructure

The present paper provides a model updating application study concerning the jacket substructure of an o?shore wind turbine. Theupdating is resolved in a sensitivity-based parameter estimation setting, where a cost function expressing the discrepancy betweenexperimentally obtained modal parameters and model-predicted ones is minimized. The modal parameters of the physical systemare estimated through stochastic subspace identification (SSI) applied to vibration data captured for idling and operational states ofthe turbine. From a theoretical outset, the identification approach relies on the system being linear and time-invariant (LTI) and theinput white noise random processes; criteria which are violated in this application due to sources such as operational variability, theturbine controller, and non-linear damping. Consequently, particular attention is given to assess the feasibility of extracting modalparameters through SSI under the prevailing conditions and subsequently using these parameters for model updating. On this basis,it is deemed necessary to disregard the operational turbine states—which severely promote non-linear and time-variant structuralbehaviour and, as such, imprecise parameter estimation results—and conduct the model updating based on modal parametersextracted solely from the idling state. The uncertainties associated with the modal parameter estimates and the model parameters tobe updated are outlined and included in the updating procedure using weighting matrices in the sensitivity-based formulation. Byconducting the model updating based on in-situ data harvested from the jacket substructure during idling conditions, the maximumeigenfrequency deviation between the experimental estimates and the model-predicted ones is reduced from 30% to 1%.

Structural model updating based on metamodel using modal frequencies

Modal frequencies are often used in structural model updating based on the finite element model, and metamodel technique is often applied to the corresponding optimization process. In this work, the Kriging model is used as the metamodel. Firstly, the influence of different correlation functions of Kriging model is inspected, and then the approximate capability of Kriging model is investigated via inspecting the approximate accuracy of nonlinear functions. Secondly, a model updating procedure is proposed based on the Kriging model, and the samples for constructing Kriging model are generated via the method of Optimal Latin Hypercube. Finally, a typical frame structure is taken as a case study and demonstrates the feasibility and efficiency of the proposed approach. The results show the Kriging model can match the target functions very well, and the finite element model can achieve accurate frequencies and can reliably predict the frequencies after model updating.

Forced Response Variation of a Compressor Utilizing Blade Tip Timing, Strain Gages, and As-Manufactured Finite Element Models

The dynamic response of turbine engine components varies widely due to manufacturing deviations in the blades known as mistuning. This dynamic variation is investigated using a single stage compressor experimentally using both blade tip timing (BTT) and strain gage (SG) measurements and using as-manufactured finite element models (AMMs) on a 1st bend mode. Operational BTT and SG safety limits were generated using both averaged and AMM models via Goodman material properties. The predicted individual blade stress/deflection (S/D) ratios and strain gage ratios for this mode will be compared to the average finite element counterparts. Additionally, the correlation between BTT and SG's will be presented. This correlation will be performed using two approaches: blade maximum stress comparisons and measured response compared to the sensors safety limits. It will be shown that accounting for geometry with AMMs produce more accurate strain gage to BTT correlation compared to average models. An experimental model updating procedure is developed to increase the strain gage to BTT correlation by optimizing the location the BTT optical spot probes measure on the blade chord. Implementing this procedure using as-manufactured models are able to improve strain gage to BTT correlation.

Scheduled QR-BP Detector with Interference Cancellation and Candidate Constraints for MIMO Systems

In this paper, a QR-decomposition-based scheduled belief propagation (BP) detector with interference cancellation (IC) and candidate constraints is proposed for multiple-input multiple-output (MIMO) systems. Based on a bipartite graph generated from an upper triangular channel matrix following linear transformation using QR decomposition, the proposed detector performs a sequential message updating procedure between bit nodes. During this updating procedure, candidate constraints are imposed to restrict the number of possible candidate vectors for the calculation of observation-to-bit messages. In addition, after obtaining the soft message corresponding to the bit sequence in each transmit symbol, a hard-decision IC operation is performed to reduce the size of the bipartite graph and indirectly update the messages for the remaining symbols. Therefore, the proposed scheme provides a huge complexity reduction compared to conventional BP detectors that perform message updating by using all related messages directly. Simulation results confirm that the proposed detector can achieve suboptimum error performance with significantly improved convergence speed and reduced computational complexity compared to conventional BP detectors in MIMO systems.

Simulated Lamb wave propagation method for nondestructive monitoring of matrix cracking in laminated composites

The estimation of the damping coefficient may help to improve the damage detection in composite materials. The purpose of this study was to develop the simulated Lamb wave propagation method for nondestructive monitoring of matrix cracking in laminated composites via the accurate estimation of their damping coefficient. Cross-ply composite specimens with different crack densities were fabricated and tested by the Lamb wave propagation technique. The phase velocity of the Lamb wave and the damping coefficient of the specimens were measured. The finite element models were developed at micro-scale (representative volume elements) and macro-scale (laminated specimens) levels to simulate the Lamb wave propagation in composite specimens. An optimization process was performed through the model updating procedure to achieve finite element models that were in good agreement with experiments. The phase velocity and damping coefficient, obtained from the updated FE models for two crack densities other than those used in the model updating procedure, were successfully examined by experimental results. It was also revealed that the damping coefficient and the rate of increase in the damping coefficient in terms of the crack density were higher for the composite laminates with a higher number of 90° layers. The damping of the fiber–matrix interphase and crack regions were considered in the model and shown as a significant contribution to the overall damping of the composite specimens. The proposed simulated Lamb wave propagation method can be used as a virtual lab for in-situ monitoring of laminated composites with different material properties, stacking sequences, and crack densities.

A sequential multi-prior integration and updating method for complex multi-level system based on Bayesian melding method

In engineering, there exist multiple priors about system and subsystems uncertainties, which should be integrated properly to analyze the system reliability. In the past research, an iterative updating procedure based on Bayesian Melding Method (I-BMM) was developed to merge and update multiple priors for the double-level system. However, the in-depth study in this paper shows that the original iterative procedure has no effect on the prior updating. Thus it is proposed that only a single BMM iteration process is needed following the original prior integration and updating formulation. BMM involves the sampling procedure for the probability density function (PDF) updating, wherein it is generally difficult to define the sampling number properly for obtaining accurate priors. To address this problem, a sequential prior integration and updating framework based on the original single BMM iteration process (S-BMM) is developed in this paper. In each cycle of prior updating, the sample number is sequentially added, and the difference between prior distributions obtained in the two consecutive cycles is measured with the symmetric Kullback-Leibler Divergence (SKLD). The sequential procedure is continued until the convergence to the accurate updated prior. The S-BMM framework for double-level systems is further extended for multi-level systems. Situations with some missing subsystem or component priors are also discussed. Finally, two numerical examples and one satellite engineering case are used to demonstrate and verify the proposed algorithms.