The development of smart grid brings the possibility of load classification control,and also puts forward higher requirements for load shedding control.Therefore,focusing on the decision problem of load shedding control,a multi-objective optimization strategy of load shedding control based on load classification was proposed from the aspects of both the system economy and the stability.A three-objective dispatch model was established that comprehensively considers economics and grid stability,and a hierarchical penalty method was used to optimize grid benefits.At the same time,in view of poor stability of the traditional non-dominated sorting genetic algorithm with elite strategy,the non-dominated sorting method based on sequential search strategy and the screening strategy based on reference point were introduced to improve the calculation speed and ensure the balanced distribution of the solution set.Finally,the IEEE 10-machine 39-node system was taken as an example for verification.The results show that the improved algorithm can effectively improve the operation rate,and can better ensure the balanced distribution of the solution set,and has a good industrial application prospect.The control strategy obtained by the algorithm can realize the optimal control of coordination economy and voltage stability.
大规模分布式电源和储能等柔性资源的接入给配电网无功电压控制带来了新的挑战.提出了一种考虑柔性可控资源的主动配电网动态无功优化控制方法,以分时电价为基础,通过协调分布式新能源、储能系统以及无功补偿设备的综合出力,构建考虑柔性可控资源的无功优化模型.以系统有功功率损耗和系统电压偏差最小为目标函数,并通过改进粒子群算法进行模型的优化求解.最后,通过仿真算例验证了所提方法的合理性和有效性.所提方法能够在保证系统安全稳定运行的前提下充分挖掘柔性可控资源的潜力,提高系统运行的经济性.
针对谐振接地网络发生高阻、小故障角度单相接地故障时故障特征不显著等问题,利用同步相量测量技术,提出了基于广义S变换时频分析的接地故障区段识别方法.利用广义S变换提取某一高频段内的频域特征量,计算故障后1/4工频周期各测量点在该时频窗内的零序电流时频特征量,计算不同测量点零序信号之间时频特征分布相似度,构建综合时频特征分布相关系数矩阵,结合模糊C均值聚类,实现接地故障区段的识别.结果表明,所提方法能准确定位出故障区段,不受故障距离、故障初相角等因素影响.
为识别交流输电线路上发生的不同暂态干扰源以及干扰变化,提出基于卡尔曼滤波的交流输电线路暂态干扰源识别方法.分析交流输电线路中的不同干扰源的特征,区分雷击暂态干扰与故障暂态干扰,将干扰源特征输入至卡尔曼滤波器中,通过递归滤波的方式判断滤波系统的最佳预测值.经试验验证:所提方法具有较低的识别误差,且能够精准识别电压均值与电流瞬时功率的变化,可精确识别不同类型的干扰源.
工业用户作为城市用电主体,具有负荷组成结构复杂,易受生产计划等因素影响产生冲击性负荷的特点.而传统的负荷预测方法难以处理工业负荷的非线性,且未充分利用负荷残差序列信息,导致负荷预测效果不佳.为此,在采用传统的变分模态分解-门控循环网络预测方法得到初始预测值的基础上,进一步考虑增加基于门控循环网络的误差修正方法对初始预测值进行修正,从而更好地适应工业负荷的多变和突变性.算例分析表明,所提出的短期负荷预测方法可有效提升的工业用户负荷预测准确率,为工业用户在生产运行和参与电力市场交易等提供强有力的决策支撑.
