光热电站容量效益评估及影响因素研究

光热发电作为一种清洁能源,配置储热后具有良好的调节性能,近年得到大力发展。依据自身经济性,光热电站往往配备较大的储热装置,能够平移所吸收的光热能,且具有跨日调节能力,使得光热机组有一定替代常规机组带负荷的能力,即容量效益。本文提出一种基于等可靠性的光热电站容量效益评估方法,考虑了调峰方式、储热时长、新能源规模、光热电站规模对光热电站容量效益发挥的影响,采用基于数学优化的生产模拟仿真程序,以周为尺度,计算全年8760 h的系统运行状态,计及了机组启停、水电跨日调节、抽蓄跨日调节、光热电站的跨日调节等因素。最后通过对西北地区实际电网的仿真,验证了所提方法的有效性。     

基于改进K-means聚类的配电网电压暂降频次估计法

传统考虑保护动作特性的电压暂降频次估计法需要获取详尽的保护配置信息,然而配电网保护配置多样,在不同因素如过渡电阻、运行方式、故障类型等的影响下,阶段式保护各级保护区可能产生较大变化,采用传统方法对电压暂降持续时间进行评估可能会产生较大误差。文中提出一种基于改进K-means聚类的配电网电压暂降频次估计方法,在未知线路保护配置基础上,基于电压暂降历史监测数据与保护动作信息,采用改进K-means聚类算法,对电压暂降幅值-持续时间进行聚类分析,推断线路保护配置情况,计算保护动作时间与保护动作电压。根据计算结果,在考虑不同故障类型、不同运行方式及不同过渡阻抗的情况下进行配电网电压暂降频次估计。在IEEE RBTS-6母线测试系统的母线5配电网中进行仿真,验证了文中方法的有效性和优越性。     

基于信息间隙决策理论的储能电站鲁棒优化配置

储能电站在转移电力需求、平抑可再生能源和负荷波动等方面发挥着重要的作用,储能电站容量优化配置是其在规划建设过程中所需要解决的关键问题。文中提出一种基于信息间隙决策理论（IGDT）的双层鲁棒优化配置模型,为解决系统在运行过程中的不确定参数带来的负面影响,模型下层搭建以经济性为目标的储能电站日前鲁棒优化运行模型;模型上层引入IGDT将经济性和鲁棒性相结合,在一定的预期目标下最大化系统能够承受的波动性,使用Benders分解法求解模型,并以IEEE 33节点配电网系统验证方案有效性,结果表明所提方案可以兼顾经济性与鲁棒性,为储能规划提供参考依据。     

基于可行空间并行的安全约束最优潮流

为了提高电网控制决策的实时性,文中将计及预想故障安全约束的经济性优化问题分解为以发电成本最小为目标的基态最优潮流主问题和预想故障安全校核子问题,并提出一种基于可调空间切片并行的分解协调算法进行求解,以避免交替迭代。首先,按不同比例对可调空间进行切片形成多个切片方案,并基于并行计算平台采用原始-对偶内点法对各切片方案进行基态最优潮流主问题求解;然后,对各优化后方式进行预想故障并行安全校核;接着从通过安全校核的切片方案中挑选出发电成本最小的方案;最后,以贵州省某市电网为算例分析验证了所提方法的有效性。     

基于碳化硅器件的无线充电系统电源设计

碳化硅(SiC)器件耐高频高温、散热性能好、导通电阻小,应用于无线充电系统可有效提高无线充电系统的效率。文中首先对比了SiC材料与Si材料的特性,在此基础上研制了一套基于SiC器件的无线充电系统电源装置。该装置由整流模块和逆变模块构成,输入端接市电,且装置的整流模块具有功率因数校正功能。文中详细给出了整流模块的整流桥选型策略、滤波电路参数设计方法、Boost电路功率器件和无源元件设计原则及开关管控制电路设计方法,还给出了逆变模块的开关管选型策略、开关管驱动电路和开关管保护电路的设计方法。最后,实验结果验证了方案的有效性和可行性,输入侧实现了高功率因数,逆变电路开关管电压振荡得到抑制,实验样机的效率峰值可达98.2%。     

基于三维特征向量的非侵入式电热负荷细分算法

非侵入式负荷监测分解（NILMD）技术是当前居民用能服务深化提升和电力供需互动的重要数据获取手段,然而当前工程上应用广泛的事件驱动型NILMD技术一直无法准确细化分解电热负荷。针对这一问题,文中提出了一种基于三维特征向量的典型电热负荷细化分解算法。首先,基于有功、无功功率和电流谐波等电气负荷特征采用事件检测方法提取电热事件,在有功功率的基础上,引入运行时长、频繁启停次数等非电气负荷特征共同构建三维特征向量电器模型。然后,采用序贯覆盖法设计典型电热负荷细化分解命题学习规则和细化分解算法。最后,基于实证实验数据进行分解验证,发现4种典型电热负荷的细化分解准确率超过85%。实验结果表明,文中所提典型电热负荷细化分解算法有效地提高了4种典型电热负荷分解的准确率。     

Interplay between Coumarin Accumulation,Iron Deficiency and Plant Resistance to Dickeya spp.

Coumarins belong to a group of secondary metabolites well known for their high biological activities including antibacterial and antifungal properties. Recently, an important role of coumarins in plant resistance to pathogens and their release into the rhizosphere upon pathogen infection was discovered. It is also well documented that coumarins play a crucial role in the Arabidopsis thaliana growth under Fe-limited conditions. However, the mechanisms underlying interplay between plant resistance, accumulation of coumarins and Fe status, remain largely unknown. In this work, we investigated the effect of both mentioned factors on the disease severity using the model system of Arabidopsis/Dickeya spp. molecular interactions. We evaluated the disease symptoms in Arabidopsis plants, wild-type Col-0 and its mutants defective in coumarin accumulation, grown in hydroponic cultures with contrasting Fe regimes and in soil mixes. Under all tested conditions, Arabidopsis plants inoculated with Dickeya solani IFB0099 strain developed more severe disease symptoms compared to lines inoculated with Dickeya dadantii 3937. We also showed that the expression of genes encoding plant stress markers were strongly affected by D. solani IFB0099 infection. Interestingly, the response of plants to D. dadantii 3937 infection was genotype-dependent in Fe-deficient hydroponic solution.     

Highly Uniform and Stable Transparent Electromagnetic Interference Shielding Film Based on Silver Nanowire–PEDOT:PSS Composite for High Power Microwave Shielding

Electromagnetic interference (EMI) shielding materials, especially ones with excellent shielding effectiveness (SE), high optical transmittance, long-term stability as well as high uniformity, are urgently desired to meet the requirements of many applications. Herein, an extremely transparent, stable and uniform silver nanowire (Ag NW)–poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) composite film as an EMI shielding material is prepared, which possess excellent shielding capability to both small signal and high power microwaves (HPM). The composite film exhibits SE of 30.5 dB in the frequency range of 1–12 GHz (small signal) and simultaneously has an optical transmittance of 91.0%. The SE continuously increases to 41.4 dB, while the optical transmittance still maintains at 81.1%. The composite film is very uniform, and its SE is almost unchanged even when exposed in air for a year. The SE of this composite film under the excitation of HPM is also thoroughly investigated. The HPM SE is much larger than that of small signal. As the power density of HPM is increased, the SE firstly remains unchanged, then continuously increases, and finally saturates. The SE exceeds 50 dB with the excitation power density of 40 W. More interestingly, the SE is saturated at a fixed HPM power density.     

Ultra-Stretchable Monofilament Flexible Sensor with Low Hysteresis and Linearity based on MWCNTs/Ecoflex Composite Materials

Wearable human–computer interaction equipment is a common technology, which can improve the comfort, convenience, and safety of the human body, and also can monitor human health. The flexible and wearable tensile sensor can be conveniently installed on clothes or directly connected to the body. This provides a convenient, timely, and portable solution for the detection of human motion. Therefore, wearable electronic equipment is gaining more attention. In this paper, a highly stretchable, flexible, and sensitive strain sensor which is based on multi-walled carbon nanotubes/Ecoflex nanocomposites is reported. A monofilament tensile sensor obtains good linearity (10.77%), low hysteresis (1.63%), good stability (6000 cycles under 100% strain), and ultra-high strain range (ε = 1300%). This ultra-stretchable sensor has potential applications in human motion monitoring, medical rehabilitation, health monitoring, human–computer interaction, and soft robots.     

电网故障处置预案文本中的命名实体识别研究

电网故障处置预案是电网故障处置的重要参考,对电网故障处置预案文本中各类电力设备、名称编号等细粒度的关键实体信息进行抽取,是实现计算机学习理解预案内容并进一步支撑故障处置智能化的重要基础。文中提出一种基于深度学习的电网故障处置预案文本命名实体识别技术,首先采用字向量表征预案文本,然后将注意力机制以及双向长短期记忆网络相结合,有所侧重地提取实体词深层字符特征,最后采用条件随机场求解最优序列化的标注。算例表明:文中所提预案文本命名实体识别模型不依赖人工特征,能够自动高效地提取文本特征,准确识别预案文本中细粒度的实体词,满足预案文本中关键实体信息精确定位和识别的要求。