充電式電動公車排班模式之研究

Abstract

本研究之目的是針對充電式電動公車設計一個不限制充電必須充飽的排班模式。目前文獻上均限制每次充電皆須充飽，卻造成實務上業者的充電成本增加和班次間銜接的困難度，因此本研究除了考量車輛續航力、車輛數以及充電座個數和種類之外，也希望能考量日夜間電價的差異，並且不限制每次充電的充電時間，協助客運業者能夠在考量電動公車的充電特性之下，更有效的去做班次安排。本研究採用具有跳脫區域最佳解機制的禁忌搜尋法(Tabu Search)作為演算法，在產生一組起始解後藉由不斷的移動並短暫接受目標值較差的解，以便移往全域最佳值。 測試範例則分成三個部分，首先採用自行設計的小範例驗證模式是否有效，接著採用現有的日月潭環湖公車時刻表進行分析，最後在以環湖公車為基礎設計三種不同發車頻率的時刻表進行驗證。分析結果發現本模式在各部分均能在充電座數目不多於車輛數的狀況下，產生一可行之車輛排程表，然而從結果可以看出發車頻率越高則會逐漸趨向採用更多的車輛以及更頻繁的使用快速充電器，來滿足充電的需求。

This study focus on the scheduling problems for battery charging electric buses, which is no charging limit. Most of the literature decided every time electric bus needed to be charge to 100%. But it will increase the charging cost and let the trips be served harder. We use tabu search to solve the model , which can avoid being stuck in suboptimal. Accepting no improving move will bring it to global optimal. This study uses simulating testing problems and a real problem to evaluate the performance of the algorithm. We choose the electric buses in Sun Moon Lake to be the real problem. And also design three schedules, which are different frequency, to test the algorithm. The testing results shows the higher frequency the schedule is, will increase the number of electric buses and the higher frequency the fast charging will be use.