This work analyzes impacts on distribution networks produced by the connection of diverse types of Plug-in Electric Vehicles (PEVs), into a probabilistic framework. Some of the uncertainty sources are related with technical parameters of PEVs. Therefore, a review of PEVs currently available in the market is reported. Other uncertain parameters are related with the behavior of the PEV owners, for instance, the arrival and departure times to home, and the state of charge of the PEV when it is plugged to the grid. These parameters are modeled by using probability density functions, to then generate random numbers and perform Monte Carlo simulations. Each Monte Carlo simulation corresponds to the calculation of a power-flow in the analyzed network. The proposed methodology is tested on the IEEE 33-bus test distribution network, with the purpose of quantifying the influence of diversity of PEVs. The analysis is performed, specifically, for identify those transformers and lines that could be overloaded. Two scenarios of PEV penetration by 2025 and 2030 were assessed, i.e., considering that 10 and 30 percent of the residential customers will have at least one PEV, respectively. Obtained results reveal the importance of considering diversity of PEV model to conduct in a suitable manner this type of studies. The proposed methodology is expected to be useful for network planning expansion and to support the design of time of use tariffs.

中文翻译：

---

插电式电动汽车多样化模型对配电网影响的概率分析

这项工作分析了将各种类型的插电式电动汽车（PEV）连接到概率框架对配电网络的影响。一些不确定性来源与电动汽车的技术参数有关。因此，据报道对当前市场上的PEV进行了综述。其他不确定的参数与PEV所有者的行为有关，例如，到家的到达和离开时间，以及PEV插入电网时的充电状态。这些参数通过使用概率密度函数进行建模，然后生成随机数并执行蒙特卡洛模拟。每个蒙特卡洛模拟都对应于所分析网络中的潮流计算。所提出的方法已在IEEE 33总线测试分配网络上进行了测试，目的是量化PEV多样性的影响。专门进行分析，以识别可能过载的变压器和线路。评估了到2025年和2030年PEV渗透的两种情况，即考虑到10％和30％的居民用户将分别拥有至少一个PEV。获得的结果揭示了考虑以适当的方式进行此类研究的PEV模型多样性的重要性。预期所提出的方法对于网络规划扩展和支持使用时间费率的设计将是有用的。考虑到10％和30％的居民客户将分别拥有至少一辆私家车。获得的结果揭示了考虑以适当的方式进行此类研究的PEV模型多样性的重要性。预期所提出的方法对于网络规划扩展和支持使用时间费率的设计将是有用的。考虑到10％和30％的住宅客户将分别拥有至少一辆私家车。获得的结果揭示了考虑以适当的方式进行此类研究的PEV模型多样性的重要性。预期所提出的方法对于网络规划扩展和支持使用时间费率的设计将是有用的。