电工技术学报  2023, Vol. 38 Issue (9): 2310-2322    DOI: 10.19595/j.cnki.1000-6753.tces.212029
电工理论与新技术 |
电动汽车动态无线充电系统输出电流模型预测控制
田勇1, 冯华逸1, 田劲东1, 向利娟2
1.深圳大学物理与光电工程学院 深圳 518060;
2.深圳职业技术学院汽车与交通学院 深圳 518055
Model Predictive Control for Output Current of Electric Vehicle Dynamic Wireless Charging Systems
Tian Yong1, Feng Huayi1, Tian Jindong1, Xiang Lijuan2
1. College of physics and Optoelectronic Engineering Shenzhen University Shenzhen 518060 China;
2. School of Automotive and Transportation Engineering Shenzhen Polytechnic Shenzhen 518055 China
全文: PDF (2695 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 动态无线电能传输(DWPT)能够为行驶中的电动汽车提供能量补给,有助于缓解里程焦虑、节省充电时间。然而,车辆行驶过程中发生的随机性横向偏移以及分段导轨切换区域的耦合关系变化等,都会引起充电功率波动。该文以LCC-S型磁耦合谐振式DWPT系统为例,提出一种结合卡尔曼滤波和模型预测控制的恒流控制方法。首先建立电能接收端Buck降压电路的状态空间模型;然后基于该模型设计卡尔曼滤波状态估计器和预测控制器;最后通过仿真和实验,验证了所提出的卡尔曼滤波和模型预测融合控制算法的有效性,并与传统的PI算法进行比较。结果表明,该控制器显著提高了恒流控制速度,并且对互感变化具有很强的鲁棒性。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
田勇
冯华逸
田劲东
向利娟
关键词 电动汽车动态无线电能传输功率控制卡尔曼滤波模型预测控制    
Abstract:Dynamic wireless power transfer (DWPT) technique is helpful to reduce electric vehicle drivers' range anxiety and save charging time because it can charge an electric vehicle in motion. However, output power of the DWPT system fluctuates dramatically and frequently due to random coil misalignment and mutual inductance change in an adjacent coil region. Therefore, an accurate and rapid controller is highly expected to stabilize the output power of the system. Some control algorithms, such as PI control, μ control, model predictive control (MPC), have been introduced for this goal. Nevertheless, they are difficult to meet the requirements on rapid response and low complexity simultaneously. To this end, this paper proposes a constant current control strategy for an LCC-S compensated DWPT system by combining a Kalman filter (KF) and a MPC controller.
A Buck converter is introduced on the secondary side to regulate the output power. Firstly, the state-space model of the Buck converter is established by analyzing its work mode. Secondly, a MPC controller is developed for determining the duty ratio of the Buck converter. Thirdly, a Kalman filter is designed to estimate state variables of the Buck converter (i.e., capacitor voltage and inductor current), instead of measuring them using sensors, thus reducing hardware complexity and cost. Finally, simulation in Matlab/Simulink and experiments on STM32F334 are carried out to demonstrate the effectiveness of the proposed method. Also, the proposed method is compared with the traditional PI controller.
Simulation and experimental results show that the Kalman filter is able to estimate the state variables of the Buck converter accurately, which is the base for implementing the MPC controller. In the STM32F334 processor, the proposed KF-MPC requires a longer computation time (153 μs) than the PI controller (22 μs). However, the KF-MPC performs better significantly in respond speed because it needs much less total control cycles than the PI controller. Particularly, the KF-MPC just takes about 15 ms, while the PI controller takes about 2.2 s to track a new reference current. As the input voltage of the Buck converter changes due to coil misalignment, the KF-MPC can always keep the output current constant, while the PI controller takes about 3 s to recover the reference output current. When the load suddenly changes from 20 Ω to 15 Ω, and from 15 Ω to 25 Ω, the KF-MPC controller only takes 10 ms, while the PI controller takes about 1.4 s in average to recover the output current. In addition, both the KF-MPC and PI controllers do not influence the system efficiency obviously. The KF-MPC performs can work stably in a wide range of parameter value, which is valuable for practical applications.
Conclusions of the paper can be summarized as follows: (1) Although the KF-MPC controller requires a longer computation time than the PI controller, it performs a faster respond speed due to the significant reduction in the total control cycle. (2) The proposed KF-MPC only needs to measure the load current for implementing the MPC for the Buck converter, so it is more practical than traditional MPC controller, which needs more measurements. (3) The proposed KF-MPC controller performs extremely high robustness to mutual inductance change, and it does not depend on the communication between the primary side and the secondary side.
Key wordsElectric vehicles    dynamic wireless power transfer    power control    Kalman filter    model predictive control   
收稿日期: 2021-12-13     
PACS: TM724  
基金资助:国家自然科学基金(62001301)、广东省重点研发计划(2020B0404030004)和广东省普通高校特色创新(2021KTSCX276)资助项目
通讯作者: 向利娟 女, 1989年生,博士,讲师,研究方向为无线电能传输、电磁场建模与仿真等。E-mail:LXiang@szpt.edu.cn   
作者简介: 田 勇 男, 1985年生,博士,副教授,研究方向为无线电能传输、电池管理等。E-mail:ytian@szu.edu.cn
引用本文:   
田勇, 冯华逸, 田劲东, 向利娟. 电动汽车动态无线充电系统输出电流模型预测控制[J]. 电工技术学报, 2023, 38(9): 2310-2322. Tian Yong, Feng Huayi, Tian Jindong, Xiang Lijuan. Model Predictive Control for Output Current of Electric Vehicle Dynamic Wireless Charging Systems. Transactions of China Electrotechnical Society, 2023, 38(9): 2310-2322.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.212029          https://dgjsxb.ces-transaction.com/CN/Y2023/V38/I9/2310