Browsing by Author "Matongo, Kabani"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Open Access
    A deep learning-based approach towards automating visual reinforced concrete bridge inspections
    (2021) Dube, Bright N; Moyo, Pilate; Matongo, Kabani
    Visual inspections are fundamental to the maintenance of RC bridge infrastructure. However, their highly subjective nature often compromises the accuracy of inspection results and ultimately leads to inaccurate prioritisation of repair and rehabilitation activities. Visual inspections are also known to expose inspectors to height and trafficrelated hazards, and sometimes require the use of costly access equipment. Therefore, the present study investigated state-of-the-art Unmanned Aerial Vehicles (UAVs) and algorithms capable of automating visual RC bridge inspections in order to reduce inspector subjectivity, minimise inspection costs and enhance inspector safety. Convolutional neural network (CNN) algorithms are state-of-the-art in relation to the automatic detection of RC bridge defects. However, much of the prior research in this area focused on detecting the presence of defects and gave little to no attention to characterizing them according to defect type and degree (D) or extent (E) ratings. Four proof-of-concept CNN models were therefore developed, namely a defect-type detector, crack-type detector, exposed-rebar detector and a shrinkage crack D-rating model. Each model was built by first compiling defect images, labelling them according to defect/crack type and creating training and test sets at a 90-10% split. The training sets were then used to train the CNN models through transfer learning and fine-tuning using the fastai deep learning python library. The performance of each model was ultimately evaluated based on prediction accuracies on the test sets and their robustness to noise. Test accuracies ≥ 87% were attained by the trained models. This result shows that CNNs are capable of accurately identifying RC bridge corrosion, spalling, ASR, cracking and efflorescence, and assigning appropriate D ratings to shrinkage cracks. It was concluded that CNN models can be built to identify and allocate D and E ratings to any visible defect type, provided the requisite training data that sufficiently represents noisy real-world inspection conditions can be acquired. This formed the basis upon which a practical framework for UAV-enabled and deep learning-based RC bridge inspections was developed.
  • Thumbnail Image
    Item
    Open Access
    Behaviour of FRP strengthened RC Beams with concrete patch repairs subjected to impact loading
    (2017) Habimana, Philbert; Moyo, Pilate; Matongo, Kabani
    The acceptable performance levels and serviceability of reinforced concrete (RC) structures are always the priorities of asset managers, engineers and researchers in any country. RC structures in service may fail to adequately perform due to changes in functionality, corrosion attack on the reinforcing bars, lack of proper and timely maintenance, and loading and standards updating, among other reasons. Impact loading is an extreme form of loading that can damage RC structures such as bridges, interchanges and flyovers during their life span. The repair and strengthening of deteriorating RC structures in service, by using concrete patch repairs and fibre reinforced polymers (FRP) respectively, has attracted a lot of attention from researchers and engineers. Nevertheless, these rehabilitated RC structures in service are susceptible to future deterioration with adverse effects. Inspection and periodic maintenance of strategic RC structures in use are essential for their safe serviceability and to avoid or mitigate economic loss. This experimental study was conducted on fifteen RC beams with the size of 155 x 254 x 2000 mm, in order to study their behaviour under impact loading testing. Twelve out of these fifteen RC beams were intentionally damaged by uniformly reducing 14 % of the cross-section of their main reinforcing bars, as this simulated the effects of corrosion on RC structures. The drop test, with the impactor applied from varying drop heights, was selected from the different types of impact loading testing methods and used in this research. Each tested RC beam was subjected to eight consecutive drop tests. During this experimental study 120 tests were performed and, from these tests, dynamic responses were recorded for analysis. Two transducers, a load cell and high-speed camera (HSC), were used to record data. In general the captured and stored dynamic responses led to the extraction of contact forces and deflections results. In addition, the HSC recorded video footage of the impact scenarios of the RC beams. The combined use of software such as Photron FASTCAM Analysis (PFA) and Matlab R2014a enables the acquisition of deflection results and, on the basis of these results, residual deflection
  • Thumbnail Image
    Item
    Open Access
    Characterisation of bridge-track interaction of a multi-span viaduct subjected to heavy haul loading
    (2022) Mupwedi, Emilia Joyce; Moyo, Pilate; Matongo, Kabani
    In many countries, railway transportation has been the primary mode of transportation, and engineers have been pushing boundaries to increase productivity and reduce costs for decades. The rail is the most important component of the railway infrastructure because it serves as the driving surface, direction guidance, and force transmission. Continuously Welded Rail (CWR), which is defined as rails that have been welded together, are now used in modern railways. When the rail is built on a bridge, the bridge's and rolling stock's behavior adds additional forces to CWR rails. As a result of the coupling effects of tracks and deformed superstructures, additional rail stresses are superimposed on other forces. These extra stresses are caused primarily by the longitudinal elongation of the superstructure as a result of temperature, braking, traction, and deck movement. The interaction of these forces between the rail and the bridge is therefore known as Track-Bridge-Interaction (TBI). Therefore, the horizontal forces must be precisely managed to prevent rail failure. This research presents a characterization of TBI for heavy haul railways and management of longitudinal forces to minimize the possibility of failure due to superimposed longitudinal forces. The Olifants River Viaduct (ORV), a 1 km long bridge with CWR and two continuous spans of 11 spans at each end and a drop span in the middle, was used as a case study in the research. The ORV has been equipped with monitoring systems to help manage the tracks. Thus, data from these systems were used to categorize the interaction forces. The research focused on categorizing the trains crossing the ORV into six (A-F) categories; the categorization was based on the train length and the commodities being hauled. The research also studied the speed variations of each train crossing the bridge. The speeds were analyzed using python and statistical tools in excel. Lastly, the impact of crossing trains on rail forces, rail temperature, ambient temperature, and deck movement was analyzed using python and statistical tools in excel. The study showed that the most frequent train to cross the bridge are category D trains with six locomotives and 342 wagons, while the train speed is dependent on the train length and the commodities hauled. Thus, the short trains in categories (A, B, and E) cross the bridge at higher constant speeds while the long trains in categories (C and D) cross the bridge at reduced speeds than the short trains but exhibit speed variations and sometimes cross the bridge at speeds exceeding the 50 km/h limit. Therefore, higher dynamic forces should be expected from short trains crossing the bridge at high constant speeds, but no additional forces should be expected on the rails from these trains as they experience no speed variations. At the same time, the long trains experience significant speed variations of both acceleration and decelerations, which imposes additional forces on the rails due to traction and braking. The imposed forces on the rails are predominantly due to crossing trains with significant speed variations of acceleration and deceleration, the acceleration change ranges from 5-30 km/h, and deceleration change ranges from 5-20 km/h. The braking and acceleration effect causes a change in the rail forces, rail temperature, and deck deflection, which in turn imposes additional forces on the rails. Therefore, high speed variation induces additional longitudinal forces on the rails. However, the imposed acceleration forces are higher than the braking forces, but the braking imposed forces are the most critical one as they tend to cause an increase in the tensile and compression forces when the forces are at their peaks, and there is a train present on the bridge, while acceleration causes a decrease in the rail forces at those times. The deck movement forces imposed on the rails were predominantly due to ambient temperature, which showed a positive linear relationship between the two. The deck expands with increasing ambient temperature and contracts with a decrease in ambient temperature. In contrast, the compression forces were within the given limits of 1100 kN, while the tension forces exceeded the rail force limit of 1400 kN when the rail temperature was between 0 − 20℃, and the deck deflection above 83 mm in the negative direction, and a present train on the bridge, making the rail more susceptible to failure during winter