双源无轨电车车载储能系统及新型供电网络优化控制

发布时间：2018-07-26 06:53

【摘要】：双源无轨电车由车载储能系统和地面供电网络提供双动力源,兼顾有尾气零排放、节约能源、行驶灵活、投资少、建设周期短等优势,成为大中城市公共交通发展的优先选择。随着双源无轨电车的大规模推广和应用,车辆的机动性、运行路况的复杂性和供电网络结构的不确定性,给双源无轨电车的安全、可靠和有效运行带来了极大的挑战。本文首先基于北京市双源无轨电车及供电网络系统的实际参数和运行数据,分析了双源无轨电车的能耗特性,建立了供电网络分段容量估计的模型,并分别利用电网络分析计算和软件仿真的方法,对供电网络分段和系统的容量进行了评估,得出了现有地面供电网络系统及运行管理技术不能满足双源无轨电车发展需求的结论。针对现有供电网络系统,建立了车载储能系统出力的优化控制模型,并利用最优控制算法在约束条件下对模型进行求解,同时,采用线性自回归的方法建立车辆的行驶功率需求预测模型,利用带遗忘因子的递归最小二乘法对预测模型参数进行辨识,并通过有限冲激响应低通滤波器来提高预测精度,融合最优控制和行驶功率需求预测,提出了基于预测的车载储能系统的分布式最优控制策略,控制车载储能系统的出力。应用车辆实际运行数据的仿真结果表明,利用该算法在不影响车辆运行的条件下,可以大大降低分段馈线的电流,从而能容纳更多数量双源无轨电车的行驶,提高供电基础设施资源的利用率。针对现有供电网络存在的分段馈线电流不均的问题,提出了新型的分布式供电网络技术,提出了约束和加权一致性控制算法,对新型供电网络进行能量管理,仅通过相邻分段的通信,实现整个供电网络分段馈线电流的均衡,理论分析和证明了提出的一致性控制算法的稳定性,并提出了大观测误差条件下一致性控制算法收敛性的改进方法,使各分段的馈线电流能较快地、平滑地收敛到全局平衡分配的目标。在各分段额定容量不一致时,还可使各个分段的馈线电流按照各自的容量成比例的输出,实现全局的加权平均。分析和研究了提出的约束和加权一致性控制算法的鲁棒性和可扩展性,并通过仿真实例验证了算法应对实际运行线路负荷突变、供电网络重组等复杂工况的可行性和有效性。最后对前文的单目标能量管理模型进行改进,提出了权衡馈线电流平衡和降低线路功率损耗的新型供电网络能量优化管理模型,分别给出了全局优化方法和分布式优化方法的解,证明了在分段额定容量一致情形下,只利用相邻分段信息的分布式优化方法可以实现全局最优解,并给出了基于一致性控制的分布式迭代算法,对多目标能量优化管理模型进行求解。通过选取不同的权重因子,可以实现不同的权衡目标,在实际应用时,可根据不同的应用场景、双源无轨电车不同的发展阶段,合理选择权重因子。
[Abstract]:Dual source trolley bus is provided with dual power source by vehicle energy storage system and ground power supply network. It has the advantages of zero exhaust emission, energy saving, flexible driving, less investment and short construction cycle. It has become the priority choice for the development of public transportation in large and medium cities. With the large-scale promotion and application of dual source trolley buses, the mobility of vehicles, the complexity of operation conditions and the uncertainty of power supply network structure bring great challenges to the safety, reliability and effective operation of dual source trolley buses. In this paper, based on the actual parameters and operation data of Beijing dual source trolley bus and power supply network system, the energy consumption characteristics of dual source trolley bus are analyzed, and the model of segmental capacity estimation of power supply network is established in this paper. Using the methods of electric network analysis and software simulation, the power supply network segment and the capacity of the system are evaluated. It is concluded that the existing terrestrial power supply network system and operation management technology can not meet the development needs of dual source trolleybus. According to the existing power supply network system, the optimal control model of vehicle energy storage system is established, and the optimal control algorithm is used to solve the model under the constraint conditions, at the same time, The linear autoregressive method is used to establish the vehicle driving power demand prediction model, the recursive least square method with forgetting factor is used to identify the parameters of the prediction model, and the prediction accuracy is improved by the low-pass filter with finite impulse response. Combining the optimal control with the prediction of driving power demand, the distributed optimal control strategy of vehicle energy storage system based on prediction is proposed to control the output of vehicle energy storage system. The simulation results of the actual vehicle operation data show that the proposed algorithm can greatly reduce the current of the segment feeder line without affecting the vehicle operation, so that it can accommodate more double-source trolley cars. Improve the utilization rate of power supply infrastructure resources. In order to solve the problem of uneven current in the existing power supply network, a new distributed power supply network technology is proposed, and a constraint and weighted consistency control algorithm is proposed to manage the energy of the new power supply network. Only through the communication of adjacent segments, the equalization of the feed current in the whole power supply network is realized. The stability of the proposed consistency control algorithm is theoretically analyzed and proved. An improved convergence method of the consistency control algorithm under the condition of large observation error is proposed, which makes the feed current of each segment converge quickly and smoothly to the goal of global equilibrium allocation. When the rated capacity of each segment is not consistent, the feed current of each segment can be outputted in proportion to its own capacity, and the global weighted average can be realized. The robustness and extensibility of the proposed constraint and weighted consistency control algorithm are analyzed and studied, and the feasibility and effectiveness of the algorithm in dealing with the complex operating conditions such as line load mutation and power supply network reconfiguration are verified by simulation examples. Finally, the single objective energy management model is improved, and a new energy management model of power supply network is proposed to balance the feeder current balance and reduce the line power loss. The solutions of global optimization method and distributed optimization method are given respectively. It is proved that the global optimal solution can be realized by using only adjacent piecewise information in the case of uniform segmental rated capacity. A distributed iterative algorithm based on consistency control is presented to solve the multi-objective energy optimization management model. By selecting different weight factors, different tradeoff objectives can be realized. In practical application, weight factors can be reasonably selected according to different application scenarios and different stages of development of dual source trolleybus.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U482.2

【参考文献】