“双碳”背景下,配合分布式光伏规划开展储能规划是目前地市供电公司规划部门的重点工作之一。文章提出了面向分布式光伏消纳的中压配电网储能选址定容实用化模型和求解方法。首先利用生产模拟仿真分析分布式光伏并网对配电网的影响;其次基于鲁棒思想提出了场景削减方法,选取分布式光伏对配电网影响严重的场景集;最后提出了中压配电网选址定容的线性优化模型,以储能每小时最大充放电功率(电量)最小为目标,利用灵敏度系数法将线路过载和节点电压越限约束表达成储能充放电功率的线性函数。对改进的IEEE 33节点系统进行案例分析,研究表明文章提出的求解方法将双层优化问题化简成为单层优化问题降低了问题的复杂度,场景削减减少了约束条件数量,线性优化能快速有效地确定储能的位置和容量。

Under the dual carbon background, a key task of the planning department of the local power supply company is to carry out energy storage planning in conjunction with distributed photovoltaic planning. This paper presents a practical model and solution method for energy storage location and capacity determination for distributed photovoltaic consumption in medium-voltage distribution networks. First, production simulation is used to analyze the impact of distributed photovoltaic grid connection on the distribution network. Second, a scenario reduction method is proposed based on a robust idea, whereby the scenario with the most serious impact on distributed photovoltaic consumption in a distribution network is selected. Finally, a linear optimization model is proposed for the location and capacity of the medium-voltage distribution network. With the goal of minimizing the maximum charging and discharging power (electricity) per hour of energy storage, the line overload and out-of-limit node voltage constraints are expressed as a linear function of the charging and discharging power of energy storage, using the sensitivity coefficient method. The case study of the improved IEEE33 bus system shows that the solution method proposed in this paper reduces the complexity of the problem by simplifying the two-layer optimization problem into a single-layer, thereby reducing the number of constraints by narrowing the scenarios for linear optimization to quickly and effectively determine the location and capacity of energy storage.

随着高渗透率分布式光伏的接入,配电网的过电压问题愈发严重,传统集中式的电压优化控制方法因为变量维数过多而无法满足控制时间的要求。基于此,该文提出一种主动配电网电压分区协调优化控制的方法。首先提出无功/有功分区质量函数作为分区指标,并以网络快速分区算法对配电网进行无功、有功分区。在无功分区内,以无功调节量最少为目标建立光伏逆变器无功优化二阶锥模型,在有功分区内,以光伏有功剪切最少为目标建立光伏逆变器有功剪切二阶锥优化模型,采用并行计算方式对各分区内二阶锥优化模型同时求解,然后将各分区优化结果作为其他分区优化约束再进行迭代优化,直至所有分区内目标函数值不再发生变化,达到各分区之间协调控制的目的。最后以某实际馈线系统为例,验证所提方法能对配电网过电压进行快速有效的控制。

The higher the penetration level of PVs in distribution networks, the greater the overvoltage issues would be. Due to the extremely high dimension of control variables, the traditional centralized control methods are getting too complex, which would barely satisfy the operational requirements of future active distribution networks. In order to address this crucial issue, this paper presents a voltage control method based on a network partitioning technique. A community detection algorithm based on a reactive/active quality function is introduced to divide a distribution network into reactive/active power sub-networks. For the reactive power sub-networks, a second-order cone programming （SOCP） based model for PV inverters is established with the objective of minimizing the regulated reactive power. For the active power sub-networks, a SOCP based model is also proposed for PV inverters with the objective of minimizing the curtailed active power of PVs. The proposed models in each sub-networks are solved by a parallel way. After the optimization, the results of the optimization are exchanged among each sub-network, and the constraints in each sub-network are then updated accordingly. A new optimization process is then started until the stopping criterion is met. Finally, an actual feeder is employed to verify the feasibility and effectiveness of the proposed approach.

针对传统配电网分布式光伏承载能力评估方法中未考虑光伏集群的问题,构建了一种以光伏集群为评估单元的承载能力双层评估模型。模型外层通过价格弹性系数描述负荷需求变化,以售电商收益最大为目标优化分时电价,采用粒子群优化(particle swarm optimization,PSO)算法进行求解。内层以不往主网反送电作为指标,考虑电压偏差和设备热稳定约束,提出灵敏度排序方法求取光伏承载能力。在我国某县域实际配电网中求取了配电网光伏集群的承载能力,验证了所提模型和算法的有效性。

Aiming at the problem that photovoltaic clusters are not considered in the traditional evaluation method of carrying capacity of distribution network for distributed photovoltaic power, a two-layer model of carrying capacity with photovoltaic clusters as the evaluation unit is constructed. In the outer layer of the model, PSO algorithm is used to describe the change of load demand through price elasticity coefficient, and the time-of-use price is optimized with the goal of maximizing the income of E-seller. In the inner layer, taking no reverse power transmission to the main network as the index, considering the voltage deviation and equipment thermal stability constraints, a sensitivity ranking method is proposed to calculate the carrying capacity for photovoltaic power. In the actual distribution network of a county in China, the impact of a single photovoltaic cluster on the carrying capacity of the distribution network is calculated, and the carrying capacity of the distribution network for the photovoltaic cluster is calculated to verify the effectiveness of the proposed model and algorithm.