- 學位論文
應用粒子群最佳化之複合動力電動車節能動力分配即時策略
Energy Economy Real-time Driving Strategy for an Electric ehicle with Multiple Traction Motors by Particle Swarm ptimization
摘要
複合動力電動車的動力鏈由多顆特性不同的馬達所組成，這些馬達存在不同的高效率區間。本研究利用粒子群最佳化法發展複合動力電動車的節能動力分配即時策略，使車輛於不同行車模式下皆能透過動力分配，操作這些馬達於高效率區間中，以提升整體行車效率，並透過Hardware-In-the-Loop(HIL)的實驗平台驗證本策略的節能性與即時性。接著針對複合動力電動車進行動態規劃分析，以反向式的計算獲得最佳化的動力分配情形，同時也分析使用單動力的傳統電動車於相同行車模式下的行駛效率。透過以上模擬結果的比較，本策略於諸多性能皆有突出的表現。 此外，本研究也針對複合動力電動車的馬達建立模型，此模型以有限元素分析的結果作為馬達參數的資料，並以考慮磁通飽和的增量電感做為馬達的電感資料。因增量電感隨不同電流下的數值不同，使得建立電感資料庫的過程繁雜且耗時。本研究也提出增量電感簡化公式，將繁雜的電感資料得以簡化，以加速有限元素分析的時間。最後透過馬達實測結果的比較，分別從相電流波形、馬達力矩-轉速曲線與馬達效率圖驗證本馬達模型的準確性。
並列摘要
The power train of the multi-powered electric vehicle consists of three motors that have different characteristics. This study aims to develop a real-time driving strategy to improve energy economy by particle swarm optimization. It aims to ascertain an optimal power distribution to operate these motors in a high efficiency area and improve driving efficiency. The economy and real-time performance were proven by hardware-in-the-loop platform. Then, we obtained the globe optimal power distribution based on dynamic programming, and we also analyzed the driving efficiency of a traditional electric vehicle to compare the benefits of our real-time strategy. This study also developed useful models for three different traction motors used in a multi-powered electric vehicle. The motor parameters in these models used the results of finite element analysis. The incremental inductance that considers the flux saturation was used in these models. Since the process of incremental inductance analysis is complicated and time-consuming, we proposed simplified incremental inductance formulas to reduce the time in finite element analysis. Finally, we compared the phase current, torque-speed curve and efficiency map to verify the model.