分布式储能电池网络的特征提取关键技术研究

发布时间：2018-03-13 03:30

本文选题：分布式储能　切入点：铅酸蓄电池网络　出处：《广西大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着可再生能源的快速发展,与用户需求侧随机性的大幅增加,带来了微电网的能量波动与不确定性,而分布式储能电池充放电是消除上述能量波动与不确定因素的重要技术支撑。因此,充分了解储能电池的性能,建立准确的电池模型、提取电池网络特征、以便准确地评估电池状态对于提高储能电池的利用率和使用寿命具有重要意义。首先,本文分析了铅酸蓄电池本身自放电特性、内阻、电压、电流、温度等因素对电池性能的影响,以及电池组之间参数相互依赖、相互制约等因素的影响,建立一种精确的高阶电池模型。通过脉冲放电辨识方法和分段式充电方式辨识高阶模型的参数,并将所建立的高阶电池模型通过扩展卡尔曼滤波算法不断的预估、修正、优化使得测量出的SOC精度更高。仿真与实验测试验证了高阶模型的准确性和有效性。其次,在此基础上,研究了储能电池自适应降阶次优化求解方法,提出利用模糊优化策略对诸多参数进行筛选,在确保精度的情况下,降低了高阶模型的求解复杂度。并分别将其应用于串、并联电池组网络中,仿真与实验测试结果验证了降阶模型的有效性。最后,将高阶电池模型与自适应降阶优化算法应用于分布式储能电池管理系统,实现了信息实时共享,减少了系统的维护工作量,提高了系统的管理效率,可为分布式储能电池的实时监测与管理提供技术支撑。
[Abstract]:With the rapid development of renewable energy and the large increase of randomness on the user demand side, the energy fluctuation and uncertainty of microgrid are brought about. The charge and discharge of distributed energy storage battery is an important technical support to eliminate the energy fluctuation and uncertainty. Therefore, the performance of energy storage battery is fully understood, the accurate battery model is established, and the characteristics of battery network are extracted. In order to evaluate the battery status accurately, it is important to improve the utilization and service life of the battery. Firstly, the self-discharge characteristics, internal resistance, voltage and current of the lead-acid battery are analyzed in this paper. The influence of temperature and other factors on the performance of the battery, as well as the influence of the parameters of the battery pack on each other, A precise high-order battery model is established. The parameters of the high-order model are identified by pulse discharge identification method and segmental charging method, and the established high-order battery model is constantly predicted and revised by extended Kalman filter algorithm. The simulation and experimental results show that the high order model is accurate and effective. Secondly, on the basis of this, the adaptive order reduction optimization method for energy storage battery is studied. A fuzzy optimization strategy is proposed to screen many parameters, which reduces the complexity of solving the higher order model under the condition of ensuring the accuracy, and applies it to the series-parallel battery network, respectively. The simulation and experimental results show the effectiveness of the reduced order model. Finally, the high-order battery model and the adaptive reduced-order optimization algorithm are applied to the distributed energy storage battery management system, and the information sharing is realized in real time. The maintenance workload of the system is reduced and the management efficiency of the system is improved. It can provide technical support for the real-time monitoring and management of distributed energy storage battery.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM912

【参考文献】