Pusat Pengajian Kejuruteraan Awam - Tesis

Browse

Now showing 1 - 5 of 220

Streamflow prediction for sungai kulim, malaysia using random forest (rf) and support vector regression (svr) models
(2024-05-01)
Yeoh, Kai Lun
Short-term streamflow prediction is important for managing immediate risks associated with extreme and unpredictable weather events. Despite numerical models showing great capabilities in streamflow prediction, they require extensive data, fundamentals of hydrology, and calibration efforts. Conversely, data-driven models are relatively quick to model and can capture the non-linearity in the streamflow time series, without requiring knowledge about the natural catchment mechanism, thus gaining traction in recent years amidst digital evolutions. In this study, two machine learning (ML) models, namely random forest (RF) and support vector regression (SVR) were introduced for multi-step ahead streamflow predictions in the Sungai Kulim catchment which has undergone speedy urban development. The models with six different input combinations were developed and assessed, using 14 years of hydrological datasets. The results revealed that the performance of the non-parametric RF algorithm was high depending on the size of terminal nodes and input configurations. Increasing the terminal node’s size improved the accuracy of the RF model. The maximum relative improvements in RMSE and NSE were 36.9% and 60.6% respectively, considering lead time up to three hours. Introducing more correlated variables into the input makes the RF algorithm capture the dynamics in time series, resulting in higher generalization to new and unseen data. Conversely, the performance of the SVR algorithm was more dependent on the selection of kernel functions and its hyperparameters, rather than the input combinations. Next, the RF model (Nash-Sutcliffe Efficiency (NSE):0.392-0.963; root mean square error (RMSE):1.485 m3/s-5.720 m3/s) performed better than the SVR model (NSE:0.190-0.830; RMSE:3.020 m3/s-6.598 m3/s) during both validation and verification stages. Although the peak streamflow was underpredicted by both RF and SVR models at all lead times, the RF model still resulted in very good predictions (PBIAS < 10%). However, the predictions by the SVR models were unsatisfactory (PBIAS > 25%), except for the one-hour-ahead streamflow (PBIAS=5.28%). The hydrographs reproduced by the RF model had smoother crest segments, rising and recession limbs as well as closer peak values, compared with the SVR model. In brief, the overall accuracy of the ML models decreased with the increasing lead-time length. The findings of this research provide insight into the use of ML algorithms for short-term streamflow prediction in Malaysia. This supports the goals of the Sendai Framework and Sustainable Development Goals by issuing flood warnings, informing risk reduction strategies, and enhancing disaster preparedness. It fosters resilient communities, safeguards human health, promotes sustainable water management, and addresses climate change impacts, contributing to safer and more sustainable development.
Wildland fire risk classification of various tree species based on ignition characteristics, heat flux, and moisture content in China
(2024-07-01)
Yang, Wenxu
Firebrands are a major source of ignition for wildland fires. In the study of predicting firebrand behavior in forest fires, existing studies of firebrands formation have been performed on the effect of limited moisture content and heating condition resulting in debatable and insufficient data in laboratory-scale or computer simulation model. Aim of this study is to investigate pyrolysis and flammability of selected wildland fuels in typical regions of China, together with other necessary parameters. Based on previous findings, seven flammable wildland fuels with high fire potential were selected. The pyrolysis and flammability performance of selected wildland fuels at different heating condition and with different moisture content was determined using thermal gravimetric analyzer and cone calorimeter respectively. Result found that: 1. Fuel property: the thickness and area were strongly affected by moisture content level; 2. Pyrolysis property: the heat conductivity coefficient increased with moisture content; 3. Flammability property: peak of heat release rate increased with moisture content, time to ignition decreased with heat flux, mass loss rate increased with heat flux, heat flux had little effect on efficient heat of combustion value. Fire risk classification was established for wildland fire prediction, specimen A and specimen B possess high risk for wildland fire compared to specimen C, D, E, F, G.
Effective control measures to minimize cost overrun during construction phase of high-rise residential building projects in Chongqing, China
(2024-03-01)
Wang, Yue
Cost overrun during construction is a common issue in construction projects around the world, including in Chongqing, China. At present, there are few studies related to cost overrun at construction stage of high-rise residential building projects (HRBPs) in Chongqing. The aim of this research is to develop effective control measures from the contractor's perspective to minimize cost overrun during construction phase of HRBPs in Chongqing. Firstly, 65 cost overrun risk factors (CORFs) in construction projects were identified with literature reviews and semistructured interviews. All the risk factors have been prioritized through the analytic hierarchy process (AHP) based on their importance to project cost. Out of the 65 CORFs identified, 12 were classified as the critical ones that have a great potential to instigate a cost overrun during construction to take place in the real project. There were four risk factors which had the biggest impact on cost overruns, and their weights were 0.04 or above, including “low bid”, “force majeure”, “undetailed/inaccurate geological survey data”, and “increased loan interest rates”. Finally, control measures were developed for these four critical cost overrun risk factors (CCORFs). In order to control "low bid", firstly, strong comprehensive strength and a reasonable enterprise quota can improve the probability of winning the bid of the construction enterprise which is the construction enterprise needs to work hard all the time. Secondly, detailed records in the bidding process can help to develop measures after winning the bid. Finally, after winning the bid, the solution measures for the low-priced list items are essential. In view of “force majeure”, this study formulated measures to control largescale infectious diseases. In response to the epidemic, preventive measures and material guarantees need to be taken beforehand. If epidemic patients are discovered in the project, they must be managed and controlled as soon as possible. In addition, it is also important to retain relevant information and reduce the pressure on corporate funds. In order to control “undetailed/inaccurate geological survey data”, measures were formulated for the bidding stage and the construction stage. In view of "increased loan interest rates", the control measures were mainly formulated from both internal and external aspects of construction enterprises. The measures developed provided a guideline to control the risk of cost overruns and clear control key points to help contractors to minimize cost overruns in the construction phase of HRBPs.
Anti-vortex device efficiency for pump sump using hydraulic physical model
(2024-07-01)
Wahidah, Nur Rizzuan
Pump sump design is very crucial in order to optimize pump inflow and pump efficiency. The pump may become damaged as a result of the vortex that forms around the pump sump suction area. The vortex may change particle velocity and increase impeller erosion. A vortex is a special form of fluid motion, with unstable and nonlinear characteristics. There are two types of vortices that occur at a pump station inlet: a submerged vortex and a free surface vortex. The free surface vortex has a great impact on the project, which will reduce the effective water flow section at the inlet of the pump station and reduce the flow capacity. To increase pump efficiency, a phenomenon called vortex generation near the pump entrance in the sump during intake needs to be managed. In this study, a pump sump model is equipped with five variations of an anti-vortex device (AVD) types at five different water level in order to assess how the anti-vortex device design affects the efficiency of vortex management in the intake flow. According to the ANSI/HI standard, the sump that produced vortex must be eliminated and the swirl angle in the flow must not be greater than 5°. All AVD variations used in this study were successful in removing the free surface vortices (FSV) at high water level (853.5 mm, 638.5 mm and 463.5 mm) but the majority of them occurred at low water level (400 mm and 370 mm). This can be avoided by placing modification of an anti-vortex device (AVD) which are back wall fillet and Floating-Anti Vortex Device (F-AVD) near the intake pipe in the pump sump. The function of Floating Anti-Vortex Device (F-AVD) is to prevent and eliminate the formation of vortices and swirl in the water flow during low water level. Proved that after placing the back wall splitter and F-AVD, the swirl angle is reaching less than 5° which achieve the lowest swirl angle reading (2.3°). As opposed to a case where there is no AVD, the application of back wall splitter and F-AVD greatly improves performance in swirl angle reduction and vortex suppression.
Lignocellulosic degradation of efb and pome during thermophilic co-composting process with biochar and non-ionic surfactant as bulking agent
(2024-02-01)
Nursyafi Amila, Hilmy
EFB composting is a sustainable waste management practice that not only addresses the challenges of waste disposal but also contributes to soil health, reduces environmental impact, and supports sustainable agriculture. To overcome these difficulties some bulking agents was added to enhance the physiochemical process and microbial community. This research has two stages which is laboratory and batch reactor phase and also had two different mixtures which is mixture A and mixture B. The different between mixture A and B was mixture A contains more bulking agents while mixture B contains higher quantity of POME. In the laboratory phase, the results showed that Mixture B exhibited more favorable results compared to A, with optimal pH (7-9), moisture content (50-70%), and lower C/N ratio. The germination index increased more in mixture B (36.67% to 84.00% by week 7) than in A (22.22% to 80.00%). Next, for the batch reactor phase the results indicated varying pH levels (6.15-10.88), moisture content (50.00-68.20%), and C/N ratios (14.28-12.89). The study found that all mixtures underwent degradation, with significant microbial activity, particularly dominated by Actinobacteria. Combinations with these bulking agents resulted in higher reductions of lignin, cellulose, and hemicellulose. In the reactor phase, the mixture of EFB + POME + BC + SC exhibited the highest reduction in lignin at 43.59% by week 7 compared to other mixtures. In brief, the overall the results show that using a bulking agents help more in the degradation of EFB during the composting process.

Browse

Recent Submissions