Azlan Abdul Rahman

A comparison of artificial neural network learning algorithms for vibration-based damage detection

This paper investigates the performance of Artificial Neural Network (ANN) learning algorithms for vibration-based damage detection. The capabilities of six different learning algorithms in detecting damage are studied and their performances are compared. The algorithms are Levenberg-Marquardt (LM), Resilient Backpropagation (RP), Scaled Conjugate Gradient (SCG), Conjugate Gradient with Powell-Beale Restarts (CGB), Polak-Ribiere Conjugate Gradient (CGP) and Fletcher-Reeves Conjugate Gradient (CGF) algorithms. The performances of these algorithms are assessed based on their generalisation capability in relating the vibration parameters (frequencies and mode shapes) with damage locations and severities under various numbers of input and output variables. The results show that Levenberg-Marquardt algorithm provides the best generalisation performance.

Application of Neural Network for Prediction of Unmeasured Mode Shape in Damage Detection

The major problem in the vibration-based damage detection field is still a limited number of sensors and the existence of uncertainties. In this paper, a new approach combines a multi-stage ANN model and statistical method to detect damage based on the limited number of sensors with consideration of uncertainties. The first stage of the ANN is used to predict the unmeasured mode shapes data based on limited measured modal data. The second stage ANN is devoted to predicting the damage location and severity using the complete modal data from the first-stage ANN. To incorporate the uncertainties in modal data, Gaussian noise is applied to the input variables and the probability of damage existence is calculated using Rosenblueths point estimate method. The feasibility of the proposed method is demonstrated using an analytical model of a continuous two-span reinforced concrete slab. The application of a multi-stage ANN showed results having a high potential of overcoming the issue of using a limited number of sensors in structural health monitoring.

Underground Corrosion Model of Steel Pipelines Using In Situ Parameters of Soil

A simple yet practical model to estimate the time dependence of metal loss (ML) in underground pipelines has been developed considering the in situ soil parameters. These parameters are soil resistivity, pH, moisture content, chloride content, and salinity. The time dependence of the ML was modeled as Pmax ¼ ktn, where t is the time exposure, k is ML constant, and n is the corrosion growth pattern. The results of ML and in situ parameters were analyzed using statistical methods such as data screening, linear correlation analysis, principal component analysis, and multiple linear regressions. The best model revealed that k is principally influenced by ressistivity, and n appears to be correlated with chloride content. Model optimization was carried out by introducing several observation criteria, namely, water access, soil color, and soil texture. The addition of these factors has improved the initial accuracy of model to an R2 score of 0.960. As a conclusion, the developed model can provide immediate assessment of corrosion growth experienced by underground structures.

Optimization of aluminium recovery from water treatment sludge using Response Surface Methodology

For decades, water treatment plants in Malaysia have widely employed aluminium-based coagulant for the removal of colloidal particles in surface water. This generates huge amount of by-product, known as sludge that is either reused for land applications or disposed to landfills. As sludge contains high concentration of aluminium, both can pose severe environmental issues. Therefore, this study explored the potential to recover aluminium from water treatment sludge using acid leaching process. The evaluation of aluminium recovery efficiency was conducted in two phases. The first phase used the one factor at a time (OFAT) approach to study the effects of acid concentration, solid to liquid ratio, temperature and heating time. Meanwhile, second phase emphasized on the optimization of aluminium recovery using Response Surface Methodology (RSM). OFAT results indicated that aluminium recovery increased with the rising temperature and heating time. Acid concentration and solid to liquid ratio, however, showed an initial increment followed by reduction of recovery with increasing concentration and ratio. Due to the solidification of sludge when acid concentration exceeded 4 M, this variable was fixed in the optimization study. RSM predicted that aluminium recovery can achieve 70.3% at optimal values of 4 M, 20.9%, 90 °C and 4.4 h of acid concentration, solid to liquid ratio, temperature and heating time, respectively. Experimental validation demonstrated a recovery of 68.8 ± 0.3%. The small discrepancy of 2.2 ± 0.4% between predicted and validated recovery suggests that RSM was a suitable tool in optimizing aluminium recovery conditions for water treatment sludge.

Altering fiber density of cockscomb-like fibrous silica–titania catalysts for enhanced photodegradation of ibuprofen

Fibrous silica-titania (FST) catalysts were synthesized by microemulsion followed by silica seed-crystal crystallization methods under various molar ratios of toluene to water (T/W). The catalysts were characterized by XRD, UV-DRS, FESEM, TEM, AFM, N2 adsorption-desorption, FTIR, and ESR. The results revealed that altering the T/W ratio affected the growth of the silica and titania and led to different size, fiber density, silica-titania structure, and number of hydroxyl groups, as well as oxygen vacancies in the FSTs, which altered their behavior toward subsequent application. Photodegradation of ibuprofen (IBP) are in the following order: FST(6:1) (90%) > FST(5:1) (84%) > FST(7:1) (79%) > commercial TiO2 (67%). A kinetics study using Langmuir-Hinshelwood model illustrated that the photodegradation followed the pseudo-first-order and adsorption was the rate-limiting step. Optimization by response surface methodology (RSM) showed that the pH, initial concentration, and catalyst dosage were the remarkable parameters in photodegradation of IBP. The FST (6:1) maintained its photocatalytic activities for up to five cycles reaction without serious catalyst deactivation, and was also able to degrade other endocrine-disrupting chemicals, indicating its potential use for the treatment of those chemicals in wastewater.