布穀鳥演算法應用於混合燃料電池電動機車之最佳能量管理

Abstract

相較於傳統燃油式車輛，發展電動車成為主要趨勢之一，然而電動車價格昂貴，有續航力比不上燃油車之缺點，且電池性能決定了電動車的最大行程與充電時間，故針對電池特性發展提升電動車續航力，本論文選用燃料電池混合電力機車進行實測，針對燃料電池與鋰電池雙電力提出最佳化能量分配策略並進行探討、優化與改善，利用基本規則庫、最小等效油耗法，以及布穀鳥搜索演算法，輸入目前的馬達需求功率、電池殘電量適時的對電池進行即時控制，以直流-直流轉換器與數位訊號控制器實現電動機車之硬體架構，探討能耗改善幅度以達到能量最佳化及行駛距離延長等目的。本論文選用測試過程中能觀察停車、高低車速以及加減速之行駛於市區的ECE-40行車型態。 為了比較最佳化能量管理策略與基本規則庫，首先透過數位訊號處理器下達控制命令於燃料電池車之電動機車整車控制器，透過整車控制器分配燃料電池之輸出功率於升壓轉換器並以機車實際測試，比較最佳化前後之數值其結果顯示行駛距離改善幅度約6.33%。 由於燃料電池混合電力機車使用單模組升壓轉換器，針對燃料電池耗盡時無法有效控制鋰電池電壓，造成控制效果不佳而提出並聯式直流-直流轉換器，將上述最佳化能量管理策略應用於此並聯式直流-直流轉換器系統，最後將整車系統進行實車測試比較，經實驗結果顯示並聯式系統之行駛距離能改善0.94%。Compared with traditional fuel-based vehicles, the development of electric vehicles has become one of the main trends. However, electric vehicles are expensive, have endurance comparable to the shortcomings of fuel vehicles, and battery performance determines the maximum travel and charging time of electric vehicles. The development of characteristics enhances the endurance of electric vehicles. This paper selects fuel cell hybrid electric locomotives for actual measurement, proposes optimal energy allocation strategies for fuel cells and lithium batteries, and discusses, optimizes and improves them, using basic rule bases and minimum equivalent fuel consumption. Method, and cuckoo search algorithm, input current motor demand power, battery residual power, timely control of the battery, DC-DC converter and digital signal controller to achieve the hardware structure of the electric motor, to explore energy consumption improvement The amplitude is to achieve the purpose of energy optimization and driving distance extension. This paper selects the ECE-40 line model that can observe the parking, high and low speed and acceleration and deceleration in the urban area during the test. In order to optimize the energy management strategy and the basic rule base, firstly, the digital signal processoris used to issue a control command to the motor vehicle vehicle controller of the fuel cell vehicle, and the output power of the fuel cell is distributed to the boost converter through the vehicle controller. And the actual test of the scooter, comparing the values before and after optimization, the results show that the driving distance improvement is about 6.33%. Since the fuel cell hybrid electric locomotive uses a single-module boost converter, the lithium-ion battery voltage cannot be effectively controlled when the fuel cell is exhausted, and the control effect is not good, and a parallel DC-DC converter is proposed to optimize the energy management described above. The strategy is applied to this parallel DC-DC converter system. Finally, the vehicle system is compared with the actual vehicle test. The experimental results show that the driving distance of the parallel system can be improved by 0.94%.