Development of geo-electrical meter based on networking

Further development of earthquake equipments is closely associated with that of computer technology. Because Embedded PC104 module has the equivalent functions of PC,it has been widely used in recent years,and can provide a new and flexible hardware design environment,but its applications in observation instruments of earth-quake precursor are rare. The present paper introduces in detail the realization of a networked geo-electrical meter by applying the low price,high reliability embedded PC104 industrial computer.

Partitioning of Water Distribution Network into District Metered Areas Using Existing Valves

Water distribution network(WDN)leakage management has received increased attention in recent years.One of the most successful leakage-control strategies is to divide the network into District Metered Areas(DMAs).As a multi-staged technique,the generation of DMAs is a difficult task in design and implementation(i.e.,clustering,sectorization,and performance evaluation).Previous studies on DMAs implementation did not consider the potential use of existing valves in achieving the objective.In this work,a methodology is proposed for detecting clusters and reducing the cost of additional valves and DMA sectorization by considering existing valves as much as possible.The procedure of DMAs identification has been divided into three stages,i.e.,a)clusters identification;b)sectorization or boundaries optimization and c)performance evaluation of the partitioned network.The proposed methodology is evaluated on a simple network and a real-world water network with the findings provided and compared to the DMAs,established for a raw water network with no existing valves.It is found that there is an adequate difference in cost of strategy implementation in both the cases for the network under consideration and the existing valve system achieved better network performance in terms of resilience index.

Investigation of Bandwidth Requirement of Smart Meter Network Using OPNET Modeler

Smart meter networks are the backbone for smart electrical distribution grid. Smart meter network requires the bidirectional communications medium and interoperability capability. As thousands of meters are interconnected in the smart meter network, it is vital to select an appropriate communication bandwidth to facilitate real-time two-way information flows and this will also allow further uptake of greenhouse-friendly technology options and enhance energy security. Optimized Network Engineering Tools (OPNET) Modeler is one of most powerful simulation tools for the analysis of communication networks. In this paper, several models of different structured smart meter networks were developed with network parameters which were connected with different communication links such as 10 BaseT and 100 BaseT in order to measure propagation delay, throughput, and utilization of the network. It was found that the propagation delay decreases with higher bandwidth. The other network parameters, namely network utilization and network throughput were also analysed. Based on the investigation, it is recommended that the 100 BaseT communication link is suitable for the smart meter network. The outcome of this paper provided a guideline to the future smart meter network developer so as to avoid catastrophic challenges faced by some of the distribution companies.

Wet Gas Meter Development Based on Slotted Orifice Couple and Neural Network Techniques

A slotted orifice has many superiorities over a standard orifice.For single-phase flow measurement,its flow coefficient is insensitive to the upstream velocity profile.For two phase flow measurement,various character-istics of its differential pressure(DP)are stable and closely correlated with the mass flow rate of gas and liquid.The complex relationships between the signal features and the two-phase flow rate are established through the use of a back propagation(BP)neural network.Experiments were carried out in the horizontal tubes with 50mm inner di-ameter,operated with water flow rate in the range of 0.2m3·h-1 to 4m3·h-1,gas flow rate in the range of 100m3·h-1 to 1000m3·h-1,and pressure at 400kPa and 850kPa respectively,where the temperature is ambient temperature.This article includes the principle of wet gas meter development,the experimental matrix,the signal processing techniques and the achieved results.On the basis of the results it is suggested that the slotted orifice couple with a trained neural network may provide a simple but efficient solution to the wet gas meter development.

Secure Communication Networks in the Advanced Metering Infrastructure of Smart Grid

In this paper, a security protocol for the advanced metering infrastructure （AMI） in smart grid is proposed. Through the AMI, customers and the service provider achieve two-way communication. Real-time monitoring and demand response can be applied because of the information exchanged. Since the information contains much privacy of the customer, and the control messages need to be authenticated, security needs to be ensured for the communication in the AM1. Due to the complicated network structure of the AMI, the asymmetric communications, and various security requirements, existing security protocols for other networks can hardly be applied into the AMI directly. Therefore, a security protocol specifically for the AMI to meet the security requirements is proposed. Our proposed security protocol includes initial authentication, secure uplink data aggregation, secure downlink data transmission, and domain secrets update. Compared with existing researches in related areas, our proposed security protocol takes the asymmetric communications of the AMI and various security requirements in smart grid into consideration.

Design of New Long-Distance Intelligent Watt-Hour Meter

The design scheme for a new kind of long-distance intelligent watt-hour meter is presented. In the system, the control computer communicates with the bank＇s host computer by the X. 25 network; the built-in embeded secure access module （ESAM） can verify data by commutation with the intelligent nodes that enhance the security of the data. IC TSS721 is adopted to collect data remotely, and the CPU IC card is used to ensure the the bargaining security and reliability. In addition, the software of the system is introduced.

Load Flow Analysis Framework for Active Distribution Networks Based on Smart Meter Reading System

With the expansion of distributed generation systems and demand response programs, the need to fully utilize distribution system capacity has increased. In addition, the potential bidirectional flow of power on distribution networks demands voltage visibility and control at all voltage levels. Distribution system state estimations, however, have traditionally been less prioritized due to the lack of enough measurement points while being the major role player in knowing the real-time system states of active distribution networks. The advent of smart meters at LV loads, on the other hand, is giving relief to this shortcoming. This study explores the potential of bottom up load flow analysis based on customer level Automatic Meter Reading (AMRs) to compute short time forecasts of demands and distribution network system states. A state estimation frame-work, which makes use of available AMR data, is proposed and discussed.

Traffic Scheduling Mechanism Based on Interference Avoidance for Meter Data Collection in Wireless Smart Grid Communication Networks

Meter Data Collection Building Area Network(MDCBAN) deployed in high rises is playing an increasingly important role in wireless multi-hop smart grid meter data collection. Recently, increasingly numerous application layer data traffic makes MDCBAN be facing serious communication pressure. In addition, large density of meter data collection devices scattered in the limited geographical space of high rises results in obvious communication interference. To solve these problems, a traffic scheduling mechanism based on interference avoidance for meter data collection in MDCBAN is proposed. Firstly, the characteristics of network topology are analyzed and the corresponding traffic distribution model is proposed. Next, a wireless multi-channel selection scheme for different Floor Gateways and a single-channel time unit assignment scheme for data collection devices in the same Floor Network are proposed to avoid interference. At last, a data balanced traffic scheduling algorithm is proposed. Simulation results show that balanced traffic distribution and highly efficient and reliable data transmission can be achieved on the basis of effective interference avoidance between data collection devices.

考虑压力均匀性的供水管网独立计量区域(DMA)分区优化研究

独立计量区域(district meteredareas,DMAs)是实现城市供水管网压力管理与漏损控制的重要手段。现有的DMA分区算法无法在分区时保障分区内部的压力均衡。为解决这一问题,本文提出一种基于迭代求解DMA分区方法,在保证分区均匀性的同时,实现分区内部压力均衡。该方法提出融合节点压力分布的节点相似矩阵,建立分区合理性评价指标,用于确定独立计量区域划分方案;在隔离阶段采用改进的多目标遗传算法,对边界管段开展隔离阀与流量计的优化布置;最后通过对上述2个过程进行循环迭代,确保最终分区方案的压力均衡。通过对实例管网的分区研究,验证改进方法的有效性和可行性。

压差式流量计误差自动化修正算法研究

针对在多干扰源扰动下压差式流量计测量结果面临输出不稳、误差较大的问题,提出多源扰动下的压差式流量计误差自动化修正算法。考虑全补偿气体可膨胀性系数、压缩系数、密度系数和流出系数等因素,研究压差式流量计误差自动化修正算法。利用均值滤波滤除信号中的高斯噪声,结合一阶滞后滤波优化卡尔曼滤波算法,修正多源扰动误差。引入自组织算法和Volterra神经网络进一步改进卡尔曼滤波算法,并优化卡尔曼滤波算法的先验模型参数,以实现多源扰动误差的自动化修正。试验结果表明,经该算法控制后:当参考流量为900 m^(3)/h时,示值误差绝对值为0.203%;当参考流量为700 m^(3)/h时,流量计重复性为0.06%。该研究可以有效识别并修正由于多源扰动造成的流量异常值,且流量测量精度较高。

基于智能电表数据的低压配电网拓扑与线路参数联合辨识

在低压配电网中,用户频繁地扩建和改接导致台账存在信息更新滞后、数据缺失等问题,难以获取当前运行状态下的拓扑及线路参数。针对无相角量测信息且含未知数量零注入功率“隐节点”的低压配电网,提出一种基于智能电表数据的低压配电网拓扑与线路参数联合辨识方法。基于低压配电网的电气特性以及辐射状网络的结构特点,推导出一种功率-电压比形式的线性逆潮流模型。通过线性回归求解得到阻抗距离矩阵,再利用无判定阈值形式的改进分组递归算法实现拓扑与线路参数的联合辨识。最后,在IEEE欧洲低压测试馈线和中国南京市某地区的实际低压配电网中对所提算法进行数值仿真,验证了所提算法的有效性。

高比例户用光伏接入下低压配电网户变关系识别

近年来,随着分布式光伏的发展,用户可参与的户用光伏系统在低压配电网中的并网数量快速增长。针对高比例户用光伏下的低压配电网具备的基本特征,提出一种基于台区配变及用户功率平衡关系的光伏时序功率卷积模型,并在此基础上,采用基于高斯混合模型的电压聚类优化目标函数进行户变关系识别。首先,基于用户-配变功率平衡关系,建立了高比例户用光伏的低压户变关系的多元线性回归模型;其次,提出一种基于Bi LSTM算法的光伏时序识别方法,筛选出光伏瞬时渗透率最高的时序片段;根据已识别的光伏时序片段建立光伏权重矩阵,并引入该矩阵将多元线性回归模型进行加权回归后转化为光伏时序功率卷积模型求解,在平抑光伏波动影响的同时初步识别户变从属关系;然后,提出一种基于高斯混合模型进行电压聚类的优化目标函数作为验证步骤,能够进一步排除验证错误的用户,提高户变关系的准确率;最后,以某地区实际低压配电网为例,分别使用该地区推广户用光伏前后不同时期的历史数据集进行识别,通过识别结果对比,验证了所提方法的有效性和实用性。

基于整体退火遗传小波网络的计量终端可靠性预测

为了解决小波神经网络初值敏感性及收敛稳定性问题,以提高计量终端软件可靠性预测建模的效率及准确性。文章完善了整体退火遗传算法(WAGA),并验证了其具有极强的整体收敛和全局优化能力,利用其全局寻优能力,优化小波神经网络(WNN)的参数,提出基于整体退火遗传小波神经网络(WAGA-WNN)的建模方法;用该方法建立计量终端的软件可靠性预测模型。实验结果表明,该方法可以解决小波神经网络初值敏感性及收敛稳定性难题,建立的软件可靠性预测模型效率和准确度较高。

不平衡数据集下基于多粒度近邻图的智能电表故障分类方法

智能电表故障的准确预测对实现计量设备精准主动运维、保障电网稳定运行具有重要意义。电表各故障类型样本的出现频次不同,且不同故障类型样本在高维特征空间中的分布存在重叠,这极大增加了故障预测的难度。现有不平衡分类方法通过构建单一样本信息与其对应类别标签的映射关系来划分样本类型,导致对具有相似表征信息的重叠区样本难以准确判别,降低了整体分类精度。该文提出一种基于多粒度近邻图的智能电表故障分类方法。首先,选择原始数据集中样本作为目标样本,以目标样本及其近邻样本作为节点、目标样本与其近邻样本连线作为边构建近邻图。根据选择的近邻样本数量不同构建多粒度近邻图,实现目标样本的信息扩充和训练样本的数量扩增,更有利于模型稳定训练。构建编码器挖掘近邻图节点特征,利用图注意力机制,根据近邻图节点编码特征和节点邻接关系将近邻样本信息自适应地聚合到目标样本,实现对相似样本差异的有效挖掘。对于给定测试样本,通过集成测试样本多粒度近邻图的分类结果,得到更精准、更鲁棒的智能电表故障预测结果。在20个KEEL(knowledge extraction based on evolutionary learning)和UCI(UC Irvine machine learning repository)不平衡分类公开数据集和智能电表实际故障数据集上的大量实验结果表明,与17种典型方法相比,该文所提算法在处理智能电表故障分类问题上具有显著优势。

电力碳排放计量网络溯源方法及计量分析

碳排放流理论在电力间接碳排放量计量中有效应用,电碳表等计量装置已试点部署,形成电力碳排放计量网络(简称电碳计量网络)。电碳计量网络具有数字化、网络化的特征,是一种新的计量方式,目前尚未有针对电碳计量网络的溯源研究及分析。该文在基于电能的碳排放流理论实现方法基础上,首先讨论用户碳排放量的计量属性,再通过对电碳计量网络中通信线路不确定度、电碳表不确定度、电碳计量算法不确定度的研究,提出电碳计量网络整体溯源方法,并对溯源方法进行分析,得出普遍贡献、电能有限影响、逐级递减3条结论,可为电碳计量网络的实际应用提供理论支撑,最后,利用实验室中部署的电碳表系统验证了文中提出的溯源方法及结论的有效性。

基于防盗算法的多功能智能电力计量系统

针对市场上传统数字电表的局限性,提出一种多功能智能电力计量系统。该系统通过集成物联网,实现智能电表的实时监控、波形谐波消除、防盗检测和负荷预测等功能。采用傅里叶分析对谐波进行分解,并使用电感器将失真降低30%,以满足IEEE 519标准。采用防盗算法,通过比较总用电量和单个电表用电量来识别异常情况。采用长短期记忆(LSTM)神经网络进行下一天的负荷预测,平均绝对百分误差为7.56%。实验结果表明,该系统在增强供电质量、安全性、实时监控以及合理的短期预测方面都较好地满足了市场的需求。

优化遗传算法在天然气管网流量分配过程模拟计算中的应用

在天然气管网流动过程中,多根管道连接处气体流动不均匀,流量分配的损失难以精确计算,对下游管道天然气参数带来计算误差。以管网实际数据为基础,构建了流量分配多目标优化模型,采用线性化方法处理控制方程,通过对边界条件的选取,将多目标优化模型简化为单目标优化模型,并将控制矩阵改写为可以使用TDMA方法求解的三点格式。使用遗传算法求解流量分配后各参数,通过监测计算误差对计算模型及时修正,对管网后续流量分配过程进行计算。结果表明,该方法可以准确模拟天然气管网流量分配过程,压力、流量的最大相对误差分别为0.15%、0.31%,温度误差为0.16 K,满足能量计量要求。

基于混合量测状态估计的配电网故障定位方法

微型相量测量单元(micro-phasor measurement unit,μPMU)为配电自动化的进一步升级提供了良好的量测基础,但现阶段电网中μPMU数量有限,难以满足传统配电网故障定位的需求。针对该问题,结合电网中μPMU与智能电表等量测设备,并基于虚拟节点的多重状态估计方法,提出了一种基于混合量测状态估计的故障定位方法。首先,通过等效变换将μPMU和智能电表的测量信息输入到故障状态估计器当中。然后,利用μPMU将网络划分为不同的区域。根据状态估计结果计算故障电流,缩小故障搜索区域以减少计算复杂度。为了识别区域内的故障位置,通过设置附加虚拟故障节点形成多种特定的故障拓扑结构并执行多重状态估计,计算出用于识别故障位置的加权测量残差指标,以确定故障位置。最后,在实时仿真系统(real-timedigitalsimulation, RTDS)中进行仿真测试,结果表明所提方法在不同故障场景下均能准确有效地定位故障,且对量测误差具有较好的鲁棒性。

基于滤波器相似性剪枝的声学场景分类

为解决场景分类系统模型复杂度高且现有场景分类剪枝算法容易丢失重要特征信息的问题,提出一种基于滤波器相似性剪枝的声学场景分类方法。以滤波器自身相似性为基础,采用余弦距离和欧氏距离度量距离融合的方法生成相似矩阵判断滤波器重要性并消除冗余滤波器。为提高分类器的泛化性,在网络输出端用决策森林后处理并构建网络。实验结果表明,所提方法能够有效在降低模型复杂度的基础上提升分类精度和系统泛化性,尤其是当消除参数较多时,所提方法特别有利。