里程焦慮影響電動車充電系統之吞吐量分析

Abstract

科技發展迅速，加上政府補貼政策與氣候變遷引發環保意識的提升，國民更換電動車的意願逐漸提高。然而，隨著電動車數量增加，「充電基礎設施」的規劃面臨挑戰且受到重視。若充電基礎設施規劃不佳，可能會導致交通問題，因此對決策者而言，充電基礎設施的規劃與統籌至關重要。 本研究著重於電動車駕駛人的「里程焦慮」(range anxiety)因素，並將該因素量化，結合過去的方法提出PBRR(a)充電策略(Partition-Based Random Routing with Range Anxiety)。本方法設計一個充電系統，並藉由最佳化模型求解出電動車的充電途程策略，模型中考慮充電需求分布、供電量、充電站位置及每座充電站的等候線穩定性以及電動車駕駛人的「里程焦慮」等因素。其目標是最大化充電系統的吞吐量(throughput)。本研究著重分析在不同供電量、充電站位置及充電需求分布下，里程焦慮如何影響充電系統的吞吐量。 實驗中本研究同時考量里程焦慮的隨機性，將其假設為一個隨機變數，並計算出吞吐量的期望值，提供決策者進行充電系統評估的參考。最後，本研究將原模型延伸至位置與資源分配問題(Location-Allocation Problem)，求解出最佳的充電站設立位置以及供電量分配，並統整出最佳解的系統配置傾向。

As the number of electric vehicles (EVs) increase, the planning of charging infrastructure has become a critical challenge. Suboptimal planning of these infrastructures may result in considerable traffic congestion and other logistical issues. Thus, it is crucial for policymakers to implement a strategic approach to the planning of these facilities to ensure their efficient and sustainable integration into the existing transportation framework. This study focuses on the impact of "range anxiety" on EV drivers and introduces a charging strategy, termed "Partition-Based Random Routing with Range Anxiety." This strategy involves the development of a charging system and an optimization model designed to determine the optimal routing policy for EVs. The model incorporates various factors, including the distribution of charging demand, power capacities, the location of charging stations, the stability of waiting lines at each station, and the range anxiety experienced by EV drivers. The objective is to maximize the throughput of the charging system. This study specifically explores how range anxiety influences the throughput of the charging system under varying conditions of power capacities, charging station locations, and demand distribution. In the experimental analysis, the stochastic nature of range anxiety is also modeled as a random variable, and the expected throughput is computed to provide a reference for policymakers. Furthermore, this study extends the original model to address the Location-Allocation Problem, determining optimal charging station locations and power capacity allocations.