應用電池健康度因子於複合電力電動車輛之最佳化尺寸與能源管理控制器設計
No Thumbnail Available
Date
2023
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本研究旨於開發具健康度因子之複合電力電動車輛最佳化尺寸與能源管理控制器設計，首先利用車輛模擬軟體Advisor(Advanced Vehicle Simulator)中數據資料庫建立複合電力電動車輛之MATLAB/Simulink控制導向物理模型，選用美國特斯拉公司(Tesla)生產之純電動車Tesla model 3作為參照；複合電力源參數利用現有之文獻搜索電力源類型，包含：鋰三元電池、磷酸鋰鐵電池、鈦酸鋰電池以及超級電容，對其不同電力源類型進行四項指標搜尋統整：比功率(kW/kg)、重量能量密度(kWh/kg)、體積能量密度(kWh/L)以及比價格(USD/kWh)。 基於參照車型之性能指標及目標函數，透過全域搜尋法(Global Search Algorithm, GSA)，求得最低出廠成本複合電力源尺寸以及DC/DC轉換器位置，再將相關參數導入上述複合電力電動車輛模型，進行基本規則庫控制(Rule Based Control, RBC)以及GSA之能量管理策略設計，為了進行整體控制策略驗證，將整車應用於不同電力源尺寸以及DC/DC轉換器位置，進而分析控制策略之能耗效益，控制策略(RBC以及GSA)於不同電力源尺寸運行兩次NEDC行車型態，GSA相較於RBC皆有14%以上之能源改善率；運行兩次WLTP行車型態，GSA相較於RBC皆有19%以上之能源改善率。 為驗證所設計之能量管理策略可於真實環境下運行，建置硬體嵌入式系統(Hardware-in-the-loop, HIL)，燒錄至硬體設備進行閉迴路即時(Real-Time)運算，透過最低出廠成本複合電力源尺寸進行HIL與SIL比較，相似度皆達到99%以上。 根據上述HIL與SIL比較結果可驗證本研究之複合電力電動車輛動態模型以及能量管理控制策略實屬合理，最終使用GSA控制策略於不同複合電力源尺寸以及不同行車型態效益分析。利用本研究設計之電池健康度因子，進行複合電力源於開發出廠至汰役階段測試，使複合電力源能平均達到汰役階段，並持續進行功率分配修正。再透過一體式尺寸/控制最佳化目標函數，計算於測試時間之總累積輸出功率與行駛里程，並分析不同複合電力源之尺寸由開發出廠至汰役階段之效益。最終結果為15kWh磷酸鋰鐵電池搭配60kWh鋰三元電池尺寸(DC/DC轉換器配置於磷酸鋰鐵電池)有最低之比壽命成本。雖12.06kWh磷酸鋰鐵電池搭配62.94kWh鋰三元電池尺寸(DC/DC轉換器配置於鋰三元電池)有最低之出廠成本，但加入本研究之電池健康度因子、時間總累積輸出功率總和以及公里數計算後，最低之出廠成本尺寸比壽命成本為9.809USD/km，一體式尺寸/控制最佳化之尺寸比壽命成本為9.066USD/km，與前者相比，比壽命成本改善率為7.6%；最終為驗證本研究之電池健康度因子之良窳性，利用一體式尺寸/控制最佳化之尺寸進行比較，加入電池健康度因子之比壽命成本為9.066USD/km，未加入電池健康度因子之比壽命成本為9.520USD/km，其前者相較於後者之比壽命成本改善率為4.8%。
The research focuses on designing optimal size and energy management for multiple-electric-energy vehicles with State-of-Health factor. First of all, we used a database from Advisor (Advanced Vehicle Simulator) to build the control-oriented physics equations of multiple-electric-energy vehicles based on Tesla model 3. Moreover, the literature offers the parameters of the multiple-electric-energy sources which include LFP battery, LTO battery, Li-3 battery, and supercapacitors. In addition, searching and summarizing four indices are specific power (kW/kg), gravimetric energy density (kWh/kg), volumetric energy density (kWh/L), and specific price (USD/kWh) for different types of hybrid energy storage system. By adopting the performance indices and the objective function, we used the global search method to calculate low-cost size of the system and the positions of the DC/DC converter for various vehicle types. These parameters will be input into the above models of multiple-electric-energy vehicles to calculate Rule-Based Control (RBC) and Global Search Algorithm (GSA).Additionally, in order to verify the overall control strategies, different sizes of electric-energy sources and the positions of the DC/DC converter will be applied in vehicles to conduct the analysis of control strategies, and then make a comparison between RBC and GSA. As RBC and GSA were used in different sizes of hybrid energy storage system under the two NEDC and two WLTP driving cycles, the control strategy of GSA energy improvement percentags are 14% and 19%, respectively, more than control strategy of RBC. In order to verify the designing control strategies that can be applied in the real environment, the control strategies should be input to HIL (Hardware-in-the-loop) to conduct the real-time calculation. Comparing HIL and SIL by lowest-cost size of hybrid energy storage system, the results show the similarity is over 99%. According to the results of comparing HIL and SIL, the reasonability of GSA control strategy has been verified with the best outcome of energy improvement percentage for the multiple-electric-energy vehicles model and energy management controller design. Thus, using GSA for the subsequent Integrated Optimization Approach (IOA) to analyze the specific life cost. In this study, the designed battery SOH factor has been adopted to test the hybrid energy storage system from begin of life to end of life, so that the hybrid energy storage can reach the end of life on average, and continuously modified power output. Then, we adopted IOA objective function to calculate the total cumulative output power and mileage during the test and analyze the specific life cost from begin of life to end of life. The final result is that the 15kWh LFP battery and the 60kWh Li-3 battery size (DC/DC converter is at the LFP battery side) have the lowest specific life cost. Although the 12.06kWh LFP battery and the 62.94kWh Li-3 battery size (the DC/DC converter is at the Li-3 battery side) have the lowest cost of begin of life, adding the battery SOH factor in this research and calculating the total cumulative output power and mileage, the specific life cost of the lowest cost of begin of life size is 9.809USD/km. However, the specific life cost size of IOA objective function is 9.066USD/km. Comparing with the former cost of begin of life size, the specific life cost improvement rate is 7.6%. In conclusion, the validation of the state-of-health factors in the present research was carried out through a comparative assessment utilizing size of IOA. The inclusion of the SOH factors in the specific life cost resulted in 9.066 USD/km. In contrast, when excluding consideration of the SOH factors, the corresponding cost is 9.520 USD/km. Evidently, the former exhibited a profound enhancement of 4.8% in terms of the specific life cost improvement compared to the latter.
The research focuses on designing optimal size and energy management for multiple-electric-energy vehicles with State-of-Health factor. First of all, we used a database from Advisor (Advanced Vehicle Simulator) to build the control-oriented physics equations of multiple-electric-energy vehicles based on Tesla model 3. Moreover, the literature offers the parameters of the multiple-electric-energy sources which include LFP battery, LTO battery, Li-3 battery, and supercapacitors. In addition, searching and summarizing four indices are specific power (kW/kg), gravimetric energy density (kWh/kg), volumetric energy density (kWh/L), and specific price (USD/kWh) for different types of hybrid energy storage system. By adopting the performance indices and the objective function, we used the global search method to calculate low-cost size of the system and the positions of the DC/DC converter for various vehicle types. These parameters will be input into the above models of multiple-electric-energy vehicles to calculate Rule-Based Control (RBC) and Global Search Algorithm (GSA).Additionally, in order to verify the overall control strategies, different sizes of electric-energy sources and the positions of the DC/DC converter will be applied in vehicles to conduct the analysis of control strategies, and then make a comparison between RBC and GSA. As RBC and GSA were used in different sizes of hybrid energy storage system under the two NEDC and two WLTP driving cycles, the control strategy of GSA energy improvement percentags are 14% and 19%, respectively, more than control strategy of RBC. In order to verify the designing control strategies that can be applied in the real environment, the control strategies should be input to HIL (Hardware-in-the-loop) to conduct the real-time calculation. Comparing HIL and SIL by lowest-cost size of hybrid energy storage system, the results show the similarity is over 99%. According to the results of comparing HIL and SIL, the reasonability of GSA control strategy has been verified with the best outcome of energy improvement percentage for the multiple-electric-energy vehicles model and energy management controller design. Thus, using GSA for the subsequent Integrated Optimization Approach (IOA) to analyze the specific life cost. In this study, the designed battery SOH factor has been adopted to test the hybrid energy storage system from begin of life to end of life, so that the hybrid energy storage can reach the end of life on average, and continuously modified power output. Then, we adopted IOA objective function to calculate the total cumulative output power and mileage during the test and analyze the specific life cost from begin of life to end of life. The final result is that the 15kWh LFP battery and the 60kWh Li-3 battery size (DC/DC converter is at the LFP battery side) have the lowest specific life cost. Although the 12.06kWh LFP battery and the 62.94kWh Li-3 battery size (the DC/DC converter is at the Li-3 battery side) have the lowest cost of begin of life, adding the battery SOH factor in this research and calculating the total cumulative output power and mileage, the specific life cost of the lowest cost of begin of life size is 9.809USD/km. However, the specific life cost size of IOA objective function is 9.066USD/km. Comparing with the former cost of begin of life size, the specific life cost improvement rate is 7.6%. In conclusion, the validation of the state-of-health factors in the present research was carried out through a comparative assessment utilizing size of IOA. The inclusion of the SOH factors in the specific life cost resulted in 9.066 USD/km. In contrast, when excluding consideration of the SOH factors, the corresponding cost is 9.520 USD/km. Evidently, the former exhibited a profound enhancement of 4.8% in terms of the specific life cost improvement compared to the latter.
Description
Keywords
複合儲能系統, 能量管理策略, 全域搜尋法, 電池健康度, Hybrid Energy Storage System, Energy Management Strategy, Global Search Algorithm, State-of-Health