基于信息融合的变压器健康状态综合评估系统研究

发布时间：2018-05-20 14:14

本文选题：状态评估 + 集对分析　；参考：《上海电力学院》2014年硕士论文

【摘要】：在国家电网公司大力推广“建设坚强智能电网”的今天，作为枢纽变电站的核心设备，电力变压器直接影响着整个电网供电的安全和可靠。加之电力行业设备检修制度的革新，状态检修技术不断发展进步，作为实现状态检修的前提——状态评估越来越受到研究学者的重视。目前，变压器状态评估方法大多仍局限于利用单一信息源信息分析判断变压器的健康状态。但由于单一信息源的不确定性，状态评估的准确率往往会不尽如人意。因此，本文在建立变压器健康状态评估体系与合理选取状态量的基础上，研究了信息融合技术在变压器健康状态评估中的应用，构建并实现了具体模型，并针对状态评估之后的故障诊断环节进行了深入研究，利用智能算法提高了故障诊断的准确性，最终在前两者的基础上开发了基于虚拟仪器LabVIEW的变压器健康状态评估系统。首先，通过研究变压器故障类型与机理以及状态评估系统的架构方式，确定了变压器各分部件状态量的选取，，基于实际构建了一套较为全面的变压器分部件健康状态评估体系。其次，将改进层次分析法与熵权法以组合的方式优化权重，并将引入集对分析方法分析变压器健康状态问题中，对其分析结果再利用D-S证据理论融合，建立起基于信息融合技术的变压器健康状态评估模型，并且用实例验证，最终得到较为全面的变压器分部件健康状态评估结果。之后深入故障诊断环节研究，提出一种模拟退火算法与粒子群优化算法相融合的改进算法，构建自适应概率神经网络故障诊断模型并以实例验证。最后，基于上述两种模型，开发了基于LabVEW与MATLAB的混合编程的变压器健康状态评估系统平台，该平台实现了状态评估、故障诊断以及数据库管理三大模块功能。
[Abstract]:As the core equipment of the hub substation, power transformer directly affects the security and reliability of the power supply of the whole power grid when the State Grid Company vigorously promotes "building a strong smart grid". In addition, with the innovation of equipment maintenance system in electric power industry and the continuous development and progress of condition-based maintenance technology, as a prerequisite to realize condition-condition maintenance, status assessment has been paid more and more attention by researchers. At present, most of the methods of transformer state evaluation are still limited to use single information source information analysis to judge the health status of transformer. However, due to the uncertainty of a single information source, the accuracy of state assessment is often unsatisfactory. Therefore, based on the establishment of transformer health assessment system and the reasonable selection of state quantity, the application of information fusion technology in transformer health state evaluation is studied, and the concrete model is constructed and realized. The fault diagnosis link after the state evaluation is deeply studied, and the accuracy of fault diagnosis is improved by using intelligent algorithm. Finally, a transformer health state evaluation system based on virtual instrument LabVIEW is developed. Firstly, by studying the type and mechanism of transformer faults and the structure of the state assessment system, the selection of the state quantities of each part of the transformer is determined. Based on the actual situation, a set of comprehensive assessment system for the health status of the transformer components is constructed. Secondly, the improved Analytic hierarchy process (AHP) and entropy weight method are combined to optimize the weight, and the set pair analysis method is introduced to analyze the health state of transformer. The results of the analysis are combined with D-S evidence theory. A transformer health state evaluation model based on information fusion technology is established and verified by an example. Finally, a more comprehensive assessment result of transformer health status is obtained. After further research on fault diagnosis, an improved algorithm combining simulated annealing algorithm and particle swarm optimization algorithm is proposed, and an adaptive probabilistic neural network fault diagnosis model is constructed and verified by an example. Finally, based on the above two models, a transformer health evaluation system platform based on LabVEW and MATLAB is developed. The platform realizes the functions of state assessment, fault diagnosis and database management.
【学位授予单位】：上海电力学院
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM41

【参考文献】