Interactions between multiple filaments and bacterial biofilms on the surface of an apple

In this paper, the interactions between two dielectric barrier discharge(DBD) filaments and three bacterial biofilms are simulated. The modeling of a DBD streamer is studied by means of 2D finite element calculation. The model is described by the proper governing equations of air DBD at atmospheric pressure and room temperature. The electric field in the computing domain and the self-consistent transportation of reactive species between a cathode and biofilms on the surface of an apple are realized by solving a Poisson equation and continuity equations. The electron temperature is solved by the electron energy conservation equation. The conductivity and permittivity of bacterial biofilms are considered, and the shapes of the bacterial biofilms are irregular in the uncertainty and randomness of colony growth. The distribution of the electrons suggests that two plasma channels divide into three plasma channels when the streamer are 1 mm from the biofilms. The toe-shapes of the biofilms and the simultaneous effect of two streamer heads result in a high electric field around the biofilms, therefore the stronger ionization facilitates the major part of two streamers combined into one streamer and three streamers arise.The distribution of the reactive oxygen species and the reactive nitrogen species captured by time fluences are non-uniform due to the toe-shaped bacterial biofilms. However, the plasma can intrude into the cavities in the adjacent biofilms due to the μm-scale mean free path. The two streamers case has a larger treatment area and realizes the simultaneous treatment of three biofilms compared with one streamer case.     

基于小波去噪与随机森林的配电网高阻接地故障 半监督识别方法

针对配电网高阻接地故障识别易受噪声干扰、无标签数据难以利用的难题,提出一种基于小波去噪与随机森林的高阻接地故障半监督识别方法。区别于监督式学习方法仅利用标签数据,基于协同训练方法能够充分利用有标签数据与无标签数据。首先,使用小波阈值去噪算法消除零序电流中的噪声。其次,采用波峰波谷故障启动算法判断线路是否发生故障或扰动事件。运用小波变换提取零序电流的小波系数作为故障特征。最后,基于小波系数故障特征构建两个随机森林作为半监督分类器进行协同训练,从而实现高阻接地故障的检测识别。仿真结果表明,所提配电网高阻接地故障半监督识别方法可以充分挖掘配电网既有的故障案例中无标注数据蕴含的关键特征,从而提高故障分类准确率,具有较强的准确性和灵敏性。     

计及新能源波动与相关性的配电网最大供电能力调度方法

配电网中分布式电源的出力由于波动具有不确定性与相关性,会对配电网的调度方案产生深远影响。为了研究新能源出力的波动对配电网在最大供电能力下调度方案的影响,采用数据驱动技术构建了新能源出力不确定集。随后构建了考虑新能源出力最差情况下的配电网双层鲁棒优化调度模型。下层模型以发电成本最低为目标,使得调度结果具有较好的经济性。上层模型考虑新能源出力不确定性最差情况,使得调度结果具有较强的鲁棒性。通过KKT条件将双层模型简化为单层,并以重复潮流法进行迭代,最终得到计及新能源波动的配电网最大供电能力调度方法。通过仿真验证了该算法的有效性与经济性,并得到了提高配电网最大供电能力的运行策略。     

配电网新型消弧技术综述

随着配电网中电力电子设备和非线性负载增多,单相接地故障电流中除了基波容性电流外,还包含大量有功分量和谐波分量,传统消弧线圈无法补偿有功分量和谐波分量,导致故障点残流增大,影响故障电弧的自行熄灭,难以满足系统可靠性和安全性需求。阐述了有源电压消弧、无源电压消弧和有源电流消弧等新型消弧技术的原理、技术特点以及技术关键点,探讨了新型消弧技术综合控制思路,为后续技术研究以及装置推广应用提供参考。     

基于CEEMD和NTEO的故障行波定位方法

针对传统行波检测方法中定位不准确、计算量大等问题,提出了一种基于补充总体平均模态分解(complementary ensemble empirical mode decomposition,?CEEMD)和改进型Teager能量算子(novel Teager energy operator, NTEO)的行波定位方法。该方法采用CEEMD算法来分解故障行波信号,滤除故障信号中的低频分量。然后采用NTEO能量算子对分解后的信号进行差分运算,增强行波波头的突变特征,实现故障行波波头的精确标定。仿真结果表明,所提方法能够精准地标定故障行波波头,并且有良好的去噪能力。与传统方法相比具有更好的检测效果和更小的定位误差,可以有效地提高配电网行波定位的精确性。