ABSTRACT
Nowadays, electricity has become an integral part of human lives. Most of our daily appliances, tools, and personal belongings are inseparable from electricity. To ensure a proper electricity distribution with an efficient transfer capability, Extra-High Voltage (EHV) transmission towers are needed. To design such a structure, it is of utmost importance to account for the cost of said tower. However, the process to estimate the cost of EHV transmission towers is both time-consuming and strenuous on human labor since a lot of consideration have to be taken. To overcome this, an imperative requirement exists for a prompt, precise, and automated tool to replace the existing manual cost estimation method. This research endeavor aims to craft a tool using support vector regression (SVR) with the capacity to prognosticate construction expenses for projects involving EHV transmission towers. The exploration of pertinent literature has enabled us to amass historical data and delineate the attributes essential for estimating costs linked to EHV transmission tower construction. The investigation delves into a comprehensive dataset spanning the past decade in Taiwan. Within this timeframe, 317 EHV transmission towers were erected between 2009 and 2019. However, 79 of these instances are excluded due to incomplete information, thereby yielding 238 viable datasets (comprising 75 % of the overall total) to underpin the development of the SVR model. By configuring the parameters to C = 0.2 and γ = 0.1, followed by 5-fold cross-validation, the resultant SVR model attains a remarkable prediction accuracy of 97.91 %, on average. As a result, the proposed SVR-based model can effectively and accurately predict the cost of constructing an EHV transmission tower project and reduce the time spent on estimation, thus contributing to the enhancement of the resilience and robustness of the transmission network system.
ABSTRACT
The COVID-19 pandemic has introduced opportunities for more research in resilience as globally cities experienced lock-down, causing change to conventional energy consumption pattern especially in the residential sector. This study aims to quantify the increased energy demand during work-from-home arrangement, using high-rise public residential buildings in Hong Kong, where its government announced work-from-home arrangement four times in 2020. Building energy modellings were conducted to compare the total energy demand of residential units during normal and work-from-home arrangements, followed by validation against peer models and empirical data. A 9% residential energy demand increase was demonstrated, hence additional energy supply became desirable for the sake of resilience. This study assesses the possibility to leverage photovoltaic rooftop to supplement the increased energy demand. The photovoltaics' potential contribution was estimated by solar energy simulation and evaluated in terms of the capability to utilize its generation output to supplement the additional energy demand. During the four work-from-home periods, it was shown that a photovoltaic system could have supplemented 6.8% - 11% of the increased energy demand, mainly subject to the air-conditioning operation and solar generation. These findings are valuable to safeguard energy resilience in upcoming grid planning and operation.
ABSTRACT
Construction workers' unsafe behaviors are closely related to construction safety performance. Most existing studies on construction workers' personality traits and safety behaviors have ignored the flexibility of worker mix at construction sites, the dynamics of workers' behaviors, and the complexity of environmental risks at construction sites. Based on the cognitive process of construction workers' safety behaviors and from the perspective of personality traits, this research establishes an agent-based model of steelworkers' mutual assistance behavior. The AnyLogic platform is adopted to show emerging phenomena in complex problems. Through simulation experiments, the optimized configuration of construction team members under different risk environments can be obtained. This research is conducive to project managers to understand the influence of construction workers' mutual assistance on team safety, assess workers' potential for safe work before recruitment, and carry out active safety management from the source instead of looking for the cause of the accident afterward, making safety management theory more realistic and dynamic.
