在面臨全球能源危機及氣候變化挑戰的背景下，電力消耗的管理與控制顯得更加重要。捷運系統作為現代城市交通的重要組成部分，其節能措施的推行不僅能降低運營成本，亦能對環境保護產生積極影響。捷運車站的節能可藉由透過高效率技術導入與能源使用策略之最佳化，可有效降低車站各子系統於營運過程中的能源消耗，並提升整體能源使用效率。 本研究的目的係針對智慧捷運車站資訊技術整合，探討於捷運車站運轉階段的節能績效並建置節電管理模式，俾以提升整體節能之績效。研究中係以桃園地區機場捷運系統為例，選定捷運系統中運量高之A2、A3、A8、A9、A18五座高架車站及A12、A13兩座地下車站，運用量測 (Measure)、管理 (Manage)、節電 (Reduce) 系統，車站用電系統及建築資訊模型 (Building Information Modeling)之空間管理資訊，彙整出各類空間及設施系統之詳細耗電資料，並以時間序列及影響耗電因子進行統計分析，運用線性回歸及機器學習、深度學習技術，提前準確地預測能源消耗，不僅有助於最佳化用電量執行，還能提高電力供應的穩定性。透過資料分析及機器學習，可以精確的預測不同時間段內的用電需求，進而訂定一套各空間及設施系統的電力配置及負載分配。建構一套高績效節電管理模式，以發揮智慧捷運車站較佳節能效益。;In the context of the global energy crisis and the challenges of climate change， the management and control of electricity consumption has become increasingly important. As a crucial component of modern urban transportation systems， metro stations play an essential role in improving energy efficiency and implementing comprehensive energy-saving strategies to reduce overall electricity consumption not only reduces operating costs but also has a positive impact on environmental protection. Energy-saving technologies in rapid transit stations include high-efficiency electric drive systems， advanced lighting technologies， and intelligent energy management systems. These technologies reduce overall electricity demand by optimizing energy use. The purpose of this study is to explore the integration of information technology into smart MRT stations， investigate whether energy-saving performance still exists during the operation of MRT s， and establish an energy-saving management model to improve overall energy efficiency. Using the Taoyuan Airport MRT system as a case study， this study selected five elevated stations (A2， A3， A8， A9， and A18) and two underground stations (A12 and A13) with high ridership. Using measurement， management， and energy-saving systems， along with spatial management information from the station′s power system and Building Information Modeling (BIM)， detailed power consumption data for various spaces and facilities were compiled. Statistical analysis was conducted using time series data and influencing factors. machine learning， and deep learning techniques were employed to accurately predict energy consumption in advance， which not only helps optimize power consumption but also improves the stability of power supply. Through data analysis and machine learning， power demand at different times can be accurately predicted， thereby establishing a power generation and distribution system for each space and facility system. A high-performance energy-saving management model is constructed to maximize the energy-saving benefits of smart MRT.
