【作者】 史常凯；

【导师】 肖峻；

【作者基本信息】 天津大学 ， 电力系统及其自动化， 2007， 硕士

【摘要】 随着地理信息系统(Geographic Information System,缩写为GIS)技术在配电网管理等各方面的广泛运用,配电GIS系统目前已经能为各种高级分析计算提供基础数据。如何有效利用配电GIS导出数据进行配电网规划,成为配电网规划研究中一个十分重要的研究方向。配电GIS系统所维护的数据量巨大且组成十分复杂,作为设备管理系统,数据的更新维护采用多用户分布方式,很难保证同一时间断面上数据的一致性,无法完全反映电网真实状态。另外,目前我国对基础数据管理水平较低,数据质量更是无法保证。本文针对配电GIS导出数据在应用于配电网评估规划工程中出现的数据质量问题,提出了一套基于规则的数据校验与修复方法,该方法包括数据表、拓扑结构和电气参数三部分的校验与修复。为解决配电GIS导出数据庞大、进行分析计算时间过长问题,在对规划评估中各项高级分析计算所需数据详细程度分析的基础上,提出了中压配电网的四层化简模型,可以将配电网数据分为按数据详细程度递减的原始网模式、全网模式、主干模式以及网架模式,并研究了各层间化简的方法。并编制程序对北京某区实际算例进行验证,结果表明本文方法能够显著降低GIS导入数据的规模,实例配电网化简到规划中最常用的主干模式后,数据规模降低到原始模式的4.2%,全网拓扑分析时间降低到1.1%,潮流计算时间降低到14.6%。显著化简同时还具有很高的保真性,例如与原始网比较的潮流计算误差小于0.8%。本文所提出方法已经开发软件实现,并在北京某供电公司的电网评估工作中得到实际应用。更多还原

【Abstract】 With the extensive use of GIS technology in distribution network management and other aspects, distribution GIS system has been able to supply basic data for various high-level analyses.How to effectively utilize exported data from distribution GIS to distribution network planning,has become a very important research direction of the distribution network planning study.Data maintained by distribution GIS is huge and very complex,and distribution GIS as a facility management system,it is very difficult to ensure the consistency of data at the same time,because of maintenance and updating of data using multi-user distribution methods,so this data is hard to reflect the real state of power network.Besides,basic data management in China is in a low level at present,this factor make the data quality more difficult to be ensured.For solving quality problems on utilizing exported data from distribution GIS to distribution network evaluation and planning, this paper proposes a set of rule-based data verification and correction (DVC) methods, including the DVC methods for data tables,topology structure and electrical parameters.For solving the long time problem of various advanced calculations which based on a huge amount of exported data from distribution GIS, based on the analysis for the detail degree of data required by these advanced calculations, this paper proposes a distribution model of four level which the detail degree of data is descending, distribution network can be described by original model, whole model, primary feeder model and structure model, and the simplification methods for each level are also deeply discussed. The practical data from part of Beijing distribution network show that the proposed mothod can reduce the amout of data significantly. For instance, data size in primary feeder model is 4.2% of that in original model, time for the whole network topology analysis is reduced to 1.1%, time to run power flow is reduced to 14.6% . At same time, analysis results are very close to the original data, e.g. power flow calculation error can be controlled within 0.8% .A software package of the presented methods are developed and successfully applied in power supply companies belonging to Beijing Electric Power Company.更多还原