Filters
Availability
Publication Year
Resource Type
Faculty
School/Institute
Publication Search Results
Sort
Results per page
1 - 5 of 5
-
-
(2022)ThesisTime series models are used to model, simulate, and forecast the behaviour of a phenomenon over time based on data recorded over consistent intervals. The digital era has resulted in data being captured and archived in unprecedented amounts, such that vast amounts of information are available for analysis. Feature-rich time-series datasets are one of the data sets that have become available due to the expanding trend of data collection technologies worldwide. With the application of time series analysis to support financial and managerial decision-making, the development and advancement of time series models in the transportation domain are unavoidable. As a result, this thesis redefines time series models for transportation planning use with the following three aims: (1) To combine parametric and bootstrapping techniques within time series models; (2) to develop a time series model capable of modelling both temporal and spatial dependencies in time-series data; and (3) to leverage the hierarchical Bayesian modelling paradigm to accommodate flexible representations of heterogeneity in data. The first main chapter introduces an ensemble of ARIMA models. It compares its performance against conventional ARIMA (a parametric method) and LSTM models (a non-parametric method) for short-term traffic volume prediction. The second main chapter introduces a copula time series model that describes correlations between variables through time and space. Temporal correlations are modelled by an ARMA-GARCH model which enables a modeller to describe heteroscedastic data. The copula model has a flexible correlation structure and is used to model spatial correlations with the ability to model nonlinear, tailed and asymmetric correlations. The third main chapter provides a Bayesian modelling framework to raise awareness about using hierarchical Bayesian approaches for transport time series data. In addition, this chapter presents a Bayesian copula model. The combination of the two models provides a fully Bayesian approach to modelling both temporal and spatial correlations. Compared with frequentist models, the proposed modelling structures can incorporate prior knowledge. In the fourth main chapter, the fully Bayesian model is used to investigate mobility patterns before, during and after the COVID-19 pandemic using social media data. A more focused analysis is conducted on the mobility patterns of Twitter users from different zones and land use types.
-
(2022)ThesisOdours from the drying process at rubber processing plants have been identified as a major malodour contributor. The increasing number of complaints from these operations has resulted in the suspension and/or shutdown due to environmental impacts by local communities. Previous studies have indicated that packed-bed wet scrubber typically adopted as an odour abatement technology is ineffective at treating odours from these plants. Literature review reveals there is various chemical groups of volatile organic compounds emitted from the raw rubber processing. To date, no comprehensive analytical study was conducted on the effectiveness of wet scrubber in removing the VOCs. This research aims to understand the composition of VOCs emitted at full-scale drying processes in Malaysian rubber processing plants in terms of their chemical composition and sensorial profile. This led to the identification of key odorants responsible for the malodour issue. The wet scrubber performance to remove the VOCs was investigated as well as the operation optimisation to improve its performance. The adsorption by activated carbon was explored as a potential secondary treatment to improve the overall VOCs removal for emissions from rubber drying processes. Lastly, the sustainability of all options to upgrade the wet scrubber system was studied. The VOCs emissions samplings were performed at two typical commercial rubber processing plants in Malaysia. VOCs samplings were collected at both inlet and outlet of the wet scrubbers using a nalpohan bag attached to a vacuum drum and subsequently transferred to sorbent tubes at dry and wet seasons as well as at different operational times. The VOCs quantification was performed using a gas chromatography-mass spectrometry/olfactometry (GC-MS/O). Additionally, the presence of hydrogen sulfide (H2S) was determined using a H2S analyzer (Jerome 631-X) in one of the sampling periods. For the optimisation, the design and operating conditions of one of the plants were used in simulation and experimental studies. Meanwhile, the adsorption behaviour of selected VOCs was investigated on two types of virgin activated carbon (AC) namely coconut-based AC (CSAC) and palm kernel-based AC (PKSAC) manufactured in Malaysia. The ACs were characterised accordingly to determine their surface characteristics. The environmental impact of the proposed improved odour abatement technologies was compared by Life cycle Assessment (LCA). A total of 80 VOCs from various chemical functional groups was frequently detected by GC-MS/O with 11 new compounds not previously reported. 50 % of the chemical concentration of the emissions was dominated by the volatile fatty acids (VFAs) and 16 critical VOCs were identified including key odorants. H2S was detected in the emissions and potentially contributes to the odour impact. The composition of the emissions was observed to be influenced by the seasonal variation in terms of the number of VOCs detected for each season. Odour wheels were developed for the first time based on the sensory analysis of full-scale rubber emission before and after treatment by the wet scrubber as a management tool for on-site plant operators and regulatory authorities to assess the malodour impact on surrounding communities. The performance analysis of the existing wet scrubber technology revealed that it was ineffective at removing VOCs, indicated by the high chemical concentration and odour activity value (OAV) detected at the outlet emissions of the wet scrubber except for single-chain VFAs. The comparable odour categories and the number of sensory-related VOCs in both inlet and outlet emissions further revealed the wet scrubber’s poor sensory removal. The wet season observed a higher wet scrubber performance because of the greater VFAs concentration detected in the emissions. However, the wet scrubber is not suitable to remove the H2S due to its poor and inconsistent removal. The study demonstrated that the combination of sensory and quantitative analysis improved the accuracy to identify the odorants from rubber emissions and investigated wet scrubber poor performance. The simulation study revealed that the performance of the wet scrubber can be optimised (> 80 % chemical concentration removal) by modification of some of the operating conditions, namely the application of a higher liquid/gas ratio and greater interfacial area of packing to remove water-soluble VOCs and subsequently, reduce the odour impact (> 90 % OAV reduction). Laboratory-scale optimisation experiments demonstrated that the VOCs solubility is highly correlated with their absorption efficiency. Furthermore, the absorption of the VOCs is best at higher gas temperature (> 45 C) and low liquid temperature (< 10 C). The experimental results show the condition in the gas phase has a greater influence on the removal efficiency compared to the liquid condition. Optimal removal of the key odorants and other critical VOCs at this stage is necessary to minimise the odour impact and organic loading before subsequent treatment. The adsorption by activated carbon has the potential to be incorporated as a secondary treatment to remove the remaining low water-soluble VOCs that are inefficient to be removed by the wet scrubber. The boiling point of the VOCs was found to be the primary factor that influences AC adsorption behaviour whereas polarity and molecular structure were secondary factors. The emissions consist of multi-component VOCs, an adsorption competition has been observed where higher boiling point VOCs have a stronger affinity with the AC and displace the weaker adsorbed lower boiling point VOCs. The presence of high relative humidity (RH >70 %) was found to shorten the breakthrough times greatly and adversely affect the AC adsorption performance. The breakthrough of VOCs categorised as odorant is more critical because of the greater odour impact contribution than its chemical concentration. CSAC demonstrated greater adsorption capacity (average 38 %) than PKSAC (average 11 %) due to the different surface characteristics. However, both ACs showed comparable adsorption behaviour. Therefore, AC adsorption could be employed as a polishing (or secondary) stage after full-scale wet scrubber abatement to improve the overall odour mitigation from rubber drying processes. Life cycle assessment (LCA) was performed on the existing wet scrubber (WS), optimised wet scrubber (OWS) and a hybrid of the optimised wet scrubber with activated carbon (OWS+AC) at full-scale operation. It was found that the impact of malodorous emissions was the greatest for the direct emissions (> 99 %) of WS due to its lower efficiency in removing the malodour. 3-methylbutanal and 3-methylbutanoic acid have the greatest contribution from the direct emission of WS whereas H2S has a greater contribution from additional components used in OWS and OWS+AC. Electricity was found to be a major contributor to global warming and either ozone formation or human health potential impact categories with the increasing trend in WS, OWS and hybrid OWS+AC, accordingly. Meanwhile, the upgrading of the wet scrubber has also impacted the operational cost. In general, the environmental impact contributed by WS primarily came from direct emissions whereas the whole supply chain of OWS and hybrid OWS+AC has a greater environmental impact than their direct emissions. Nevertheless, the application of renewable energy is a good option to reduce the environmental impact.
-
(2022)ThesisThe increasing focus on sustainability has driven manufacturers to develop greener cementitious products in order to reduce the associated carbon dioxide emissions. Supplementary cementitious materials (SCM) such as ground-granulated blast furnace slag (GGBFS) are commonly used in the development of low carbon, blended cement formulations. The reduced strength development due to the addition of slag can be improved by the addition of chemical activators such as sodium sulfate (Na2SO4). The reactivity and early strength of cement:slag binders is significantly enhanced by the addition of Na2SO4 though the mechanisms underlying the relationship between the enhanced hydration reactions and the strength behaviour remain unclear. This thesis explores the impact of Na2SO4 on the early age chemical reaction mechanism in blended systems containing a high percentage of blast furnace slag. Early hydration reactions and resultant compressive strength in a 50:50 slag:cement binder in the presence of Na2SO4 were investigated. Early strength increases in the presence of Na2SO4 were shown to be due to a combination of increased alite hydration and increased slag dissolution. Increased alite hydration was due to neither reduced dissolved Al concentration nor increased alite under-saturation, but related to increased ionic strength due to the presence of Na2SO4. Increased slag dissolution was associated with both increased pH and decreased Ca activity with the two being connected through the portlandite solubility limit. Na2SO4 was shown to substantially enhance slag dissolution at fixed pH 13 with this action attributed to greater under-saturation of slag as a result of ettringite formation. Na2SO4 was shown to be superior to alternate activators in a slag:cement binder. Microstructural development in the presence of Na2SO4 was investigated utilizing mercury intrusion porosimetry (MIP), NMR relaxometry, and XRD. Increased rates of early strength development and decreased rates of late strength development due to the presence of added Na2SO4 were linked to effects on capillary porosity refinement. While the degree of hydration at later age was shown to have been lower in the presence of Na2SO4, and may have been responsible for the higher capillary porosity, a clear alteration in the pathway of microstructural development had occurred with inhibition to hydration of the slag component due to earlier microstructural development. Subsequently, different cement types with varying clinker chemistry were used to investigate which clinker phase was primarily responsible for the acceleration reaction with Na2SO4. Early hydration reactions were examined and monitored utilizing mercury intrusion porosimetry (MIP), calorimetry, and XRD while the resultant compressive strength in a 50:50 slag:cement binder in the presence of Na2SO4 was investigated for up to90days. Forallcementsexamined,astronginversecorrelationbetweenearlystrength development and late age strength was apparent with results suggesting that the retardation of the early hydration reactions due to the presence of sodium sulfate resulted in a higher late age strength development. Previous studies indicate that Na2SO4 activation improves the early strength of blended cements due to enhanced ettringite formation. In this study, we examine whether triethanolamine (TEA) provides further early strength increase through additional ettringite formation by investigating the effect of TEA at 0 to 0.5% w/w of cement in a 2.5% w/w sodium sulfate activated 47.5:50 GGBFS:Portland cement blend. Results of XRD analyses indicate that TEA encouraged ettringite formation through enhanced calcium aluminoferrite (ferrite) and gypsum consumption, and, to a lesser extent, enhanced tricalcium aluminosilicate (C3A) consumption. Increased porewater Fe(III) concentrations provided further evidence of TEA-facilitated ferrite dissolution. Compressive strength results correlated with the degree of calcium silicate hydration and ettringite transformation, being highest in the 0.2% TEA blend but lowest in the 0.5% TEA blend. Finally, the effect of higher gypsum content in blended cement:slag can led to C3A dissolution retardation. Associated early age strength development was examined using calorimetry, pore solution chemistry, XRD, and MIP. Gypsum was found in these cements to exert no effect on alite hydration as it is consumed during C3A hydration and ettringite formation. Compressive strength measurements in the “gypsum studies” reinforced the inverse relationship between early and late strength development and the critical role of alite during early hydration towards late age strength development. This work hopes to contribute a more comprehensive understanding of the effect of Na2SO4 on the early hydration stages of GGBFS incorporated Portland cement blend systems from a geochemical approach in effort to achieve greater sustainability through higher cement replacement.
-
(2022)ThesisUrban stormwater runoff possesses the properties of intermittent occurrence, unexpectable volume and variable pollution which lead to different environmental issues, including flooding and waterlogging, pollution transportation, damage to downstream and contamination of the receiving waters. On the other hand, the low-level contamination (relative to sewerage) and large volume supply of stormwater makes it suitable as an alternative water resource to relieve the water shortage in the urban areas. Stormwater harvesting is under the concept of Water Sensitive Urban Design (WSUD) trying to treat stormwater properly for the different end-uses (like irrigation, toilet flushing and even for the uses close to human contact). Several treatment technologies (e.g., biofilters, constructed wetlands) have already been implemented to purify the stormwater with effective performance prior to reuse. However, the refractory organic micropollutants (especially herbicides) presented resistance to these nature-based solutions by showing variable treatment outcomes. In order to provide harvested stormwater for end-uses with high quality requirement (e.g., close to human contact recreational waters), a reliable treatment technology for organic micropollutants is desired as a post-treatment method in the stormwater harvesting system. This thesis aims to develop advanced oxidation processes (AOPs), in particular, photoelectrochemical oxidation (PECO), as the post-WSUD treatment approach for stormwater using its oxidation capacity towards the refractory organic micropollutants. Following the technology development procedure, three steps have been conducted: (1) testing the feasibility of AOPs for stormwater herbicides treatment; (2) investigating the intrinsic mechanism in the stormwater herbicides degradation process; and (3) assessing the operation conditions impact towards PECO stormwater treatment system. Boron-doped diamond (BDD) anode was used in the preliminary lab-scale tests for the feasibility study of AOPs towards stormwater organic micropollutants (two representative herbicides, diuron and atrazine - selected as the target pollutants in the study). The results showed that the effective herbicides degradation could be achieved by PECO process under 5 V operation (which was regarded as the optimal voltage in the system). The positive impact coming from voltage increase has been found in the study. BDD showed a remarkably durable property with stable removal performance under challenging voltage application (9 V) without observed deterioration. The catalysts loading showed negligible effect in removal performance. While the thermal effect was observed as a supporting factor for the process (higher temperature supported the oxidation process). Since BDD is not a perfect choice for scaled-up implementation due to its high manufacture cost, carbon fiber anode was chosen in the following studies and operated under low voltage (2 V) to avoid the possible anode deterioration under high voltage application. In the mechanism investigation, the superoxide radicals were found to be the major reactive species in PECO process. Meanwhile, hydroxyl radicals and free chlorine also demonstrated supporting impact for the oxidation process. With the identified intermediate products, degradation pathways of diuron and atrazine were proposed for the first time for three AOPs (PECO, electrochemical oxidation (ECO) and photocatalytic oxidation (PCO)) in stormwater herbicides degradation process. PECO was certified to be the preferrable stormwater treatment technology with the ability for further oxidation reactions towards herbicides degradation compared with ECO and PCO. In the third study, a flow reactor was designed and used to test the impacts of operational conditions (flow rate, light intensity, and initial pollutant concentration) for PECO process. An obvious improvement was observed for flow rate towards removal performance, while the light intensity was found to influence atrazine removal only. The initial pollutant concentration study demonstrated the robust performance of PECO flow reactor towards herbicides removal under challenging (240 μg L-1) pollutant concentration condition. The real stormwater experiments suggested the possible impacts coming from the stormwater chemistry towards PECO process. Further based on the energy consumption analysis, high flow rate (610 mL min-1) and normal light intensity (100 mW cm-2) were regarded as the optimal operational conditions for flow reactor system. Also, the effective PECO degradation performance of herbicides under the real stormwater environment has been verified by using the stormwater collected from field as supporting electrolyte in the experiments. Overall, this thesis confirms PECO as a promising stormwater herbicides treatment technology (potentially for all organic micropollutants) to provide further purification for stormwater high-quality targets. It also discusses the implications for the practical implementation and points out the future research directions for the system optimization.