Faisal Shabbir

Particle Swarm Optimization with Sequential Niche Technique for Dynamic Finite Element Model Updating

Due to uncertainties associated with material properties, structural geometry, boundary conditions, and connectivity of structural parts as well as inherent simplifying assumptions in the development of finite element (FE) models, actual behavior of structures often differs from model predictions. FE model updating comprises a multitude of techniques that systematically calibrate FE models in order to match experimental results. Updating of structural models can be posed as an optimization problem where model parameters that minimize the errors between the responses of the model and actual structure are sought. However, due to limited number of experimental responses and measurement errors, the optimization problem may have multiple admissible solutions in the search domain. Global optimization algorithms (GOAs) are useful and efficient tools in such situations as they try to find the globally optimal solution out of many possible local minima, but are not totally immune to missing the right minimum in complex problems such as those encountered in updating. A methodology based on particle swarm optimization (PSO), a GOA, with sequential niche technique (SNT) for FE model updating is proposed and explored in this article. The combination of PSO and SNT enables a systematic search for multiple minima and considerably increases the confidence in finding the global minimum. The method is applied to FE model updating of a pedestrian cable-stayed bridge using modal data from full-scale dynamic testing.

Application of multi-objective optimization to structural damage estimation via model updating

This paper presents a novel damage detection method which simultaneously updates the undamaged as well as damaged structure model in a multi-objective optimization (MOO) process. Structural health monitoring via analysis of modal data and model updating has received considerable attention in the previous decade. Such damage detection methods typically require an updated baseline model of the undamaged structure and the associated errors can become aggregated when this baseline model is subsequently used for damage detection. The use of multi-objective model updating alleviates those issues. A beam structure with and without damage has been used as an example and different noise levels have been added to the identified mode shapes. The results have been compared with single-objective model updating and it has been found that the proposed method is more efficient for accurate estimation of damage severity.

Study of adhesion characteristics of different bitumen–aggregate combinations using bitumen bond strength test

ABSTRACT The resistance of asphalt pavements to traffic loading and environmental deterioration depends on the mechanical coupling of the selected bitumen–aggregate material, bitumen–aggregate interfacial properties, and cohesion of bitumen. Studies have shown that bitumen–aggregate bond strength strongly depends on the composition of the two components, and the extent of exposure to moisture. For satisfactory performance of bituminous pavements, there is a need to choose suitable combinations of bitumen and aggregate materials. This paper evaluates the adhesion of different bitumen–aggregate combinations in dry and wet conditions in the form of pull-off tensile strength using bitumen bond strength (BBS) test. Tests were carried out after conditioning the selected combinations for 24, 48, and 72 h. The results indicated that the bond strength reduces with the passage of time for the samples exposed to moisture while it increases if samples are left in the dry condition. Results show that the strength values are also affected by the aggregates’ nature. Samples containing basic aggregates showed good results in comparison to the acidic aggregates under moisture conditioning. Significance of aggregates type, bitumen type, curing condition, and curing time on pull-off tensile strength of bitumen–aggregate samples was also studied. Curing condition showed maximum significance on bond strength.

Application of firefly algorithm to the dynamic model updating problem

Model updating can be considered as a branch of optimization problems in which calibration of the finite element (FE) model is undertaken by comparing the modal properties of the actual structure with these of the FE predictions. The attainment of a global solution in a multi dimensional search space is a challenging problem. The nature-inspired algorithms have gained increasing attention in the previous decade for solving such complex optimization problems. This study applies the novel Firefly Algorithm (FA), a global optimization search technique, to a dynamic model updating problem. This is to the authors’ best knowledge the first time FA is applied to model updating. The working of FA is inspired by the flashing characteristics of fireflies. Each firefly represents a randomly generated solution which is assigned brightness according to the value of the objective function. The physical structure under consideration is a full scale cable stayed pedestrian bridge with composite bridge deck. Data from dynamic testing of the bridge was used to correlate and update the initial model by using FA. The algorithm aimed at minimizing the difference between the natural frequencies and mode shapes of the structure. The performance of the algorithm is analyzed in finding the optimal solution in a multi dimensional search space. The paper concludes with an investigation of the efficacy of the algorithm in obtaining a reference finite element model which correctly represents the as-built original structure.

A comparison of two global optimization algorithms with sequential niche technique for structural model updating

Much effort is devoted nowadays to derive accurate finite element (FE) models to be used for structural health monitoring, damage detection and assessment. However, formation of a FE model representative of the original structure is a difficult task. Model updating is a branch of optimization which calibrates the FE model by comparing the modal properties of the actual structure with these of the FE predictions. As the number of experimental measurements is usually much smaller than the number of uncertain parameters, and, consequently, not all uncertain parameters are selected for model updating, different local minima may exist in the solution space. Experimental noise further exacerbates the problem. The attainment of a global solution in a multi-dimensional search space is a challenging problem. Global optimization algorithms (GOAs) have received interest in the previous decade to solve this problem, but no GOA can ensure the detection of the global minimum either. To counter this problem, a combination of GOA with sequential niche technique (SNT) has been proposed in this research which systematically searches the whole solution space. A dynamically tested full scale pedestrian bridge is taken as a case study. Two different GOAs, namely particle swarm optimization (PSO) and genetic algorithm (GA), are investigated in combination with SNT. The results of these GOA are compared in terms of their efficiency in detecting global minima. The systematic search enables to find different solutions in the search space, thus increasing the confidence of finding the global minimum.

Damage estimation using multi objective genetic algorithms

It is common to estimate structural damage severity by updating a structural model against experimental responses at different damage states. When experimental results from the healthy and damaged states are available, the updated finite element models corresponding to the two states are compared. Updating of these two models occurs sequentially and independently. However, experimental errors, updating procedure errors, modelling errors and parametric errors may propagate and become aggregated in the damaged model in this approach. In this research, a multi-objective genetic algorithm has been proposed to update both the healthy and damaged models simultaneously in an effort to improve the performance of the damage estimation procedure. Numerical simulations of a simply supported beam damaged at multiple locations with noisy mode shapes were considered and improved model updating results were confirmed. It was found that the proposed method is more efficient in accurately estimating damage severity, less sensitive to discretization as well as experimental errors, and gives the analyst an increased confidence in the model updating and damage estimation results.

Structural damage estimation using dynamic data and multi-objective optimization

A novel damage estimation method is proposed using a multi-objective optimization technique which simultaneously updates the damaged as well as undamaged structural model. Contemporary damage detection and estimation methods based on model updating typically require a prior updated baseline finite element model of the undamaged structure for subsequent updating of the damaged model. However, multi-objective optimization algorithms utilize both the undamaged and damaged structure models concurrently to improve the performance of the damage estimation procedure. An updated solution is selected which gives a trade-off between updating of both models using the concept of Pareto front. The technique was applied to a numerically simulated structure with damage, and natural frequencies and mode shape information was used for model updating. Different noise levels were added to account for the experimental errors. It was found that the proposed method gives more accurate damage location and estimation than traditional updating, and is less sensitive to experimental errors.