手輪馬達電動輪椅之混合質子交換膜燃料電池電力鏈整合與實現

Abstract

本論文研究一個串聯式混合燃料電池電力鏈系統，實際應用整合於一台手輪馬達電動輪椅上。這個混合燃料電池電力鏈為一個質子交換膜燃料電池與兩組磷酸鋰鐵二次電池之組合。本研究的目的是透過電力管理的控制策略，使燃料電池具有穩定的輸出功率，驗證此混合電力鏈及電力管理策略搭載至輪椅的可行性及實用性，以提升整車行駛續航力，並維持磷酸鋰鐵電池整體的殘電量。。實驗證實本研究所提出的電力管理策略，使燃料電池在充電功率略高於平均負載功率的前提下，確實可以減少了二次電池組充電交換次數。在殘電量估測方面，本研究發現：與開路電壓量測法搭配庫倫積分法比較，使用EEPORM memory搭配庫倫積分法，在誤差累積修正後，能夠有效控制殘電量估測誤差在3 %以內。在使用者介面上，使用LCD顯示器顯示兩電池組的即時殘電量值，讓使用者能夠了解電池組的狀態而不會造成電力匱乏的情況發生；並有氫氣壓力異常警示裝置，提醒使用者氫氣即將耗盡，必須關閉氣體球閥，以確保使用者的行駛安全。

This paper studies a series hybrid fuel cell powertrain system which is applied and integrated practically on an electrical wheelchair power by rim motors. The hybrid fuel cell powertrain is a combination of one proton exchange membrane fuel cell and two lithium iron phosphate battery packs. The propose of this study makes the fuel cell working on a stable output power thought the power management strategy and verifies the feasibility of hybrid powertrain and management strategy on the wheelchair to upgrade the driving time and maintain the whole SOC of lithium iron phosphate batteries. Under the premise that the charging power of fuel cell is slightly higher than the average power of load, experiments verify the proposed power management strategy can indeed reduce the charging exchange number of secondary battery pack. In terms of estimating SOC compared with open circuit voltage measurement method with coulomb counting method, this study found that using the EEPROM memory with coulomb counting method can effectively control the estimation error of SOC less than 3% after the cumulative error correction. In the user interface, we used LCD display to show the immediate SOC of batteries and the user can know the status of the battery packs to avoid power shortage occurring. Besides, there is a hydrogen pressure abnormality warning device to remind users closing the gas valve when hydrogen run out.