超高压直流阀厅套管拆装平台与技术研究  

Research on EHV DC valve hall bushing disassembly and installation platform and technology

作　　者：罗宗源 邓文斌 王伟 蔡潞 万施霖 LUO Zongyuan;DENG Wenbin;WANG Wei;CAI Lu;WAN Shilin(Guiyang Bureau of China Southern Power Grid Ultra High Voltage Transmission Company,Guiyang 550000,China;Huanan Industrial Technology Research Institute of Zhejiang University,GuangZhou 510000,China)

机构地区：[1]中国南方电网有限责任公司超高压输电公司贵阳局,贵阳550000 [2]浙江大学华南工业技术研究院,广州510000

出　　处：《自动化与仪器仪表》2022年第4期186-192,共7页Automation & Instrumentation

基　　金：南方电网贵阳局超高压输电公司提供的科研项目资助。

摘　　要：超高压直流阀厅套管具有设备长度长、重量大、拆装难度大、各项拆装工艺控制以及试验要求细等特点。传统拆装工艺需使用吊车,且经常凭借现场人员的经验与目测进行施工。提出了一种智能拆装系统,该系统包括一个能在阀厅内部操作的智能拆装平台,以及基于模拟退火(Simulated Annealing,SA)算法改进的粒子群优化(Particle Swarm Optimization,PSO)的三维拆装定位算法和基于机器学习的拆装故障分析技术,能大幅度提高拆装过程的程序化、规范化、精细化和智能化。The EHV DC valve hall bushing has the characteristics of long equipment length,heavy weight,difficult disassembly and installation,various disassembly and installation process control as well as fine test requirements.The traditional disassembly and installation process requires the use of cranes and often relies on the experience and visual inspection of the site personnel.An intelligent disassembly and installation system is proposed,which includes an intelligent disassembly and installation platform that can be operated inside the valve hall,as well as a three-dimensional disassembly and installation positioning algorithm based on particle swarm optimization(PSO) improved by simulated annealing(SA) algorithm and a machine learning-based disassembly and installation fault analysis technique.The proposed system can realize the procedure,standardization,refinement and intelligence of the disassembly and installation process.

关 键 词：超高压直流阀厅 套管拆装 智能拆装平台 三维定位 粒子群优化 机器学习 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置] TM723[自动化与计算机技术—控制科学与工程]