Browsing by Type "text"
Now showing 1 - 7 of 7
Results Per Page
Sort Options
- PublicationA human detection framework based on hog and cnnf features using visual and far infrared images(2020-07-01)Chee, Kok WeiHuman detection has become an essential feature in many current technological devices. For example, the ability to detect human is useful in a surveillance or safety alarm systems. However, human detection from images is a very challenging task. The main challenges include low accuracy, complex scene with cluttered background or illumination issue such as low lighting condition or overexposure problem. In this research, a framework to detect human by using both visual and far-infrared image is proposed. The proposed method used a single channel image as input and Power Law transformation is applied on the input image to reduce the illumination problem. Image gradient magnitude is used in selecting the Region of Interest (ROI) with the purpose of narrowing down the investigation area within the complex scene. The proposed method used the fusion of Histogram of Oriented Gradient (HOG) and Convolutional Neural Network Filters Features (CNNF) as classification feature. The features are concatenated in both visual and far-infrared image to further improve the detection accuracy under various lighting condition such as daylight or nighttime. Performance evaluation using the visual images from Caltech benchmark dataset showed that the proposed method achieved 60.11% miss rate at 0.1 False Positive per Image (FPPI). Besides, this research has also evaluated the fusion of visual and far-infrared images using a manually collected dataset. The test results showed that the proposed method can achieve 56.95% miss rate at 0.1 FPPI using both visual and far-infrared images and 73.75% miss rate at 0.1 FPPI when using visual image only. This proved that the fusion of both visual and far-infrared images can achieve better performance compared to use visual image alone for human detection.
- PublicationAnalysis of the staggered crossbeam of pier designed based on deep beam design concept(2021-07-01)Ko, Jiunn XuanThe deep beam concept has different load transferring behaviour from the simple beam concept. Deep beam develops diagonal tension zone extends from loading point to support point, causing the shear cracking and leading to failure of the structure. The staggered crossbeam in the pier in an existing completed project had occurred the deep shear cracking due to lack of consideration of deep beam concept during the design stage, resulting in the heavy and complicated remedial works. Therefore, this study focuses on analysing the staggered crossbeam of pier designed based on established design philosophy. In this study, two staggered crossbeams of the pier with similar geometrical dimensions were designed based on the STM and simple beam concept, respectively. The reinforcements provided by these two methods were compared and evaluated. The result showed that the reinforcement provided by the STM is more than that designed based on simple beam concept. Moreover, the finite element analysis was conducted for staggered crossbeam pier with varied crossbeam elevation differences to assess the behaviour of reinforced concrete and the adequacy of reinforcement provided when subjected to concentrated vertical loads. The models were divided into five categories with a different elevation of right crossbeam and provided six reinforcement groups for each model. The models were analysed with ABAQUS finite element analysis software. The analysis result indicated that the adequacy of reinforcement for staggered crossbeam of the pier is reduced when the variation of crossbeam elevation increases. The adequacy of reinforcement for the staggered crossbeam of the pier was obtained based on the model with the least tension zone in the stress distribution pattern.
- PublicationDevelopment of alkali-activated binder utilizing silico-manganese fume and blast-furnace slag(2021-05-01)Nasir, MuhammadThe negative impacts of proliferation of silico-manganese fume (SiMnF) of about 100-150 kg per tonnage of SiMn alloy produced and increase in the carbon footprint due to production of ordinary Portland cement (OPC) premised the need for this study. This is necessary to enhance public health, minimize the solid waste generation, reduce global warming and develop alternative cost-efficient construction materials for civil engineering infrastructures. This thesis addresses the use of alkali-activated binding technology to mitigate the challenges associated with the concrete and other industries. This led to the development of novel and sustainable alkali-activated mortars (AAMs) using high level of silico-manganese fume (SiMnF) and ground granulated blast furnace slag (GGBFS) as precursor materials (PMs) together with NaOHaq (NH) and Na2SiO3aq (NS) as the alkaline activators (AAs). The optimization of mixes was achieved using L16 orthogonal array based on the Taguchi method (TM). The mix parameters studied were GGBFS/PMs (0-0.5), sand/PMs (1.5-2.4), NH concentration (0-16M), NS/NH ratio (0-3.5), silica modulus (0-3.4) and AAs/PMs (0.5-0.53). The influence of curing methods, namely room-, moist-, and heat-curing (for 3-24 h between 25-95 °C) and durability performance under the exposure to acid and sulphate environments were also studied. Fresh properties and mechanical strength were evaluated, while analytical studies, such as mass stability, bond characteristics, nature of the products formed and morphology of the microstructures were undertaken using thermogravimetric (TG) analysis, FT-IR analysis, X-ray diffraction (XRD) analysis, and scanning electron microscopy (SEM) plus energy dispersive spectroscopy (EDS), respectively. The optimum mortar mix consisted of SiMnF:GGBFS, sand/PMs, Na2SiO3aq/10M-NaOHaq and AAs/PMs ratios of 70:30 wt.%, 1.5, 2.5 and 0.5 such that the SiO2/Na2O, H2O/Na2O and H2O/SiO2 ratios were 1.61, 17.33 and 10.77, respectively. This combination yielded a 3-, 7- and 28-day compressive strength of 22.5, 29.7 and 44.5 MPa, respectively at room-curing, whereas the heat-curing for 6 h at 60 °C was beneficial for attaining the highest strength within 3-days. Among the prominent compounds that defined the microstructure of the developed AAMs were stratlingite/gehlenite hydrate (C-A-S-H), nchwaningite/glaucochroite (C-Mn-S-H), and potassium feldspar (K-A-S-H) phases. Exposing the product to acid attack caused faster deterioration by decalcification and formation of gypsum with S-O bonds and formation of carbonation as a result of reactivity of lime with atmospheric CO2. Exposure to MgSO4aq caused more deterioration leading to spalling of specimens due to formation of gypsum and brucite crystals in comparison with Na2SO4aq where the stability was aided by quartz-based compound. It is envisaged that the results obtained from the novel AAMs would be beneficial in understanding the behaviour and an initiative towards practical application of the materials beside attaining economic, ecological and technical advantages.
- PublicationEffect of glass on stiffness of the cable-net structure under the static and dynamic conditions(2020-08)Marzuki, Nur AshikinStructural glass facade or cable-net supported glass façade has been widely used as a modern building facade in engineering construction for a long-span structure of greater than 7 meters owing to its simple, light and transparent structures and these advantages make it dominant in airport, lobby and others. The contribution of glass to the stiffness on a cable-net structure normally neglected since the effect has small percentage under static load compared to a dynamic load. Nevertheless, glass panels prone to provide structural stiffness of a cable-net structure when the cables are damaged or pre-tension losses. This research aims to investigate the damages occurred within the components of a cable-net structure has determining the contribution of glass stiffness on the cable-net structure with and without glass panels under both static and dynamic behavior. An experimental carries three components of component failures, including cable-connector, cables and glass panels. The glass stiffness contribution on the performance of cable-net structure was determined from the deflection and natural frequencies of the cable-net with and without glass panels under static and dynamic behavior. Under different pre-stress level of cable forces (1000 N, 1500 N and 2000 N), the contribution of glass stiffness to the cable-net structure increased with increased cable forces within 7% to 24.5%. No major effect was readily seen in the deflection for each cable losses within edge cable to provide the huge impact upon the structure, similarly with the cable connector failure where the difference only found less than 5%. The major contribution from glass stiffness are more significant as the cable being totally damaged to contribute 7.2% to the cable-net structure. Nevertheless, under dynamic loading, when both direction of the cables subjected to cable losses, the natural frequency of the cable-net structure decreased with increasing pre-tension losses, but cable-net structure with glass panels have a small difference between in each losses, i.e 0.12 Hz from 30% to 60% losses and 0.18 Hz from 60% to 100% compared with the cable-net structure without glass panels where 0.33 Hz from 30% to 60% and 1.06 Hz from 60% to 100%. It showed that the glass panels has significant contribute the stiffness to the cable-net structure. Hence, it is important to consider the contribution of the glass to the stiffness of a cable-net structure where the design of the cable-net structure can be proposed by consider the glass panels during the design process.
- PublicationEvaluation of damaged reinforced concrete beams repaired with epoxy injection and strengthening with carbon fibre reinforced polymer using acoustic emission technique(2020-09-01)Saliah, Soffian Noor MatDamage due to increasing loading in reinforced concrete (RC) structures is a crucial phenomenon where it may cause structural failure and collapse. Rehabilitation of damaged RC structure is important to restore structural integrity. The most promising techniques are structural crack repair and strengthening. However, assessment on damaged RC structure repaired with epoxy injection and strengthening with carbon fibre reinforced polymer (CFRP) is extremely required. From the literature review, it was found that the damaged RC beam repaired with epoxy injection and strengthened with CFRP using acoustic emission (AE) signal parameters are still limited. This study aimed to investigate, correlate, evaluate and develop flexural crack behavior (crack mode) of RC beam repaired with epoxy injection and strengthened with CFRP supplemented with AE signal parameters namely signal strength, average frequency (AF), rise angle (RA) and intensity analysis (IA). Flexural static test was carried out in the laboratory of heavy structure on RC beam size of 200 mm x 300 mm x 1500 mm. The RC beam was intentionally damaged prior to be repaired with epoxy injection and strengthened with CFRP. The flexural static test was carried out until failure. All analyses were derived from the AE signal collected at sensors labeled CH6 and CH7. Both sensors were then compared, analysed and discussed. It is found that the AE signal parameter are dependent on the type of repaired and strengthened RC beams. At the same time, good relationships between load application, AE signal parameter, crack modes and crack patterns were inferred on the RC beams. From the AF vs RA value, clear trend between the relationship of the crack mode on identification of tensile crack and shear crack were found. The IA well corresponds to the repaired and strengthened technique for each crack mode by the plots of intensity zone at CH6 and CH7. In addition, the understanding on visual inspection supplemented with AE technique is useful in future monitoring and maintaining of damaged RC structure repaired with epoxy injection and strengthened with CFRP.
- PublicationInfuence of mass transfer towards pilot-scale semi-continuous cultivation of chlorella vulgaris(2020-06-01)Khoo, Choon GekMicroalgae, well-known for their prominent photosynthetic efficiency and rapid growth rate emerge as a great feedstock for bio-energy production of third-generation biofuel. In this study, Chlorella vulgaris was chosen as the subject of investigation. The aim was to maximize the biomass production by investigating both the kinetic and mass transfer phenomena in a pilot-scale bubble column photobioreactor (BC-PBR) cultivation system. To account for the maximum microalgal biomass accumulation, the microalgae growth condition was optimized in the semi-continuous cultivation mode. Each cultivation cycle was carried out with 15 days of batch cultivation mode, followed by 3 cycles of 5 days each during semi-continuous cultivation mode. One-factor-at-a-time (OFAT) method was employed to investigate the effects of inoculum concentration of microalgae cells, photoperiod, and aeration rate towards microalgal growth performance, in the range of 0.1 – 0.35 g L-1, 12 and 24 h, and 0.12 – 0.19 vvm, respectively. The underlying mass transfer mechanism between gaseous CO2 and the culture medium were investigated under the optimized growth conditions. In addition, the reusability of the recycled water from the harvesting process was evaluated. To convert the microalgae into application biofuel, the harvested microalgal biomass was then converted into hydrochar via hydrothermal carbonization (HTC) reaction. The effects of hydrothermal temperature and retention time and the properties of hydrochar were studied at the range of 180 – 250 oC and 0.5 – 4 h, respectively. The research results showed that the optimum biomass accumulation was at 0.9819 g L-1, with cultivation conditions of: inoculum concentration of 0.3 g L-1, exposed under continuous (24 h) illumination with light intensity 60 – 70 μmol m-2 s-1, and supplied with compressed air at aeration rate of 0.16 vvm. The cultivation system underwent a bubble breakup mechanism during the transportation of gaseous CO2 into the culture medium with gas-liquid mass transfer coefficient, kLaL(CO2) of 0.45 s-1. Higher CO2 concentration environment did not affect the biomass accumulation due to the solubility limitation of CO2 in the microalgae culture. Based on optimized growth conditions for microalgae, a mathematical model for microalgae growth was developed. By incorporating the mass transfer parameter into the modified growth model, which was validated through an extended 120 days (21 cycles) of semi-continuous cultivation. In addition, the microalgae cells were proven to be able to grow in the recycled harvesting water. On the other hand, the highest energy yield of hydrochar was achieved at 76.59%, at the HTC under 210 oC for 0.5 h. Comparatively, higher heating value (HHV) of hydrochar produced was measured to be 24.51 kJ g-1, which is higher than that of raw biomass (12.58 kJ g-1). Moreover, the HTC process produced an aqueous phase that could be used as an alternative nutrient source for microalgae cultivation, yielding an average biomass accumulation of 0.8483 g L-1, demonstrating the feasibility of close loop cultivation. To conclude, mass transfer was a dominant factor affecting the kinetic growth of microalgae in pilot-scale semi-continuous BC-PBR cultivation system. It further affected the quality of produced biomass, and thus affected the downstream processing route chosen for optimal conversion of bioenergy.
- PublicationMulti-stage and multi-responses optimization of punching die performance using taguchi method(2020-04-01)Ja’afar, Noor AzamPunching process is one of the fundamental technologies in metal forming operations. The sharpness of the die edge in the punching process is very important to ensure the quality of the produced hole. Hole punching is a multi-stage and multi-response process, beginning with the fabrication of the die, where machining parameters need to be considered, then geometries of the die, i.e. the design parameters, and finally the punching process parameters. At each stage, the responses are different but would be implicitly related between stages. From the literature review, the evaluation and optimization of these stages are usually treated separately and explicitly to avoid complexity. Thus, the new approach to investigate the effect of punching die wear was taken by considering all stages in punching die process, starting from the fabrication, design and punching process utilisation. The purposes of this study were to improve the sharp edge of punching die from wear problem in order to enhance punching tool life and product quality. The study parameters at every stage of fabrication process, design process and punching operation were considered in a series of experiments and optimization processes using the Taguchi Method. Analysis of variance (ANOVA) was used to determine the significant factors that affect the responses. As the study considers the entire punching process, the results of the experiment from the previous stage has been utilized for the next stage of experiment. In the first stage, the effects of WEDM cutting parameters of pulse on time, pulse off time, servo voltage and ignition pulse current on the surface roughness and white layer thickness (WLT) of 2379 steel were evaluated. In the second stage, the effects of two die design parameters, i.e. die clearance angles and die opening sizes, were compared on the accuracy and quality of punched holes produced on 1.4-mm thick S275 mild steel. The die was fabricated using the optimal WEDM cutting parameters identified in the earlier stage. Burr formations and die weight losses were used as an indicator for determining the best die design parameters. Subsequent experimentation was conducted to optimize the die design parameters for die opening size, angular clearance angle and die clearance on die wear. Finally, based on these optimal die design parameters, several punching dies were fabricated for the optimization of the punching process parameters such as material types, sheet thickness, and lubrication on die wear. A non-linear mathematical model was developed for predicting the response of the die wear. For the first stage, it was found that pulse on time was the most significant factor affecting the surface roughness for both types of cutting modes (main cut and multiple trim cuts). Meanwhile, the results show that pulse on time and pulse off time were the most significant factors on WLT for both main cut and multiple trim cuts. The second stage results show that the design parameters combination of Ø20 mm die opening size, 0.50° angular clearance angle and 2% die clearance would give the optimal condition for the lowest wear rate. Final results show that the combination of punching process parameter of electro-galvanized (EG) plate of 1.4 mm thickness and no lubrication gave the optimal condition for the lowest wear. The confirmation experiment showed that the predicted results by the developed mathematical model differ by only 0.0007% with the actual value. The developed model helps appropriate selections of die design geometries, WEDM machining parameter during fabrication and punching parameters during operation to improve the tool life and product quality.