PLC硬件结构的水电站故障监测系统及故障识别
2022,30(10):17-21
摘要：针对现有水电站故障监测系统识别效率低、工作难度大等问题,提出了一种基于巡检机器人视觉识别的故障监测方法,并将其应用于水轮机调节系统(Hydro-Turbine Governing System,HTGS)故障诊断问题。通过可编程逻辑控制器(Programmable Logic Controller,PLC)保证巡检机器人的稳定工作运行,通过非线性输出频率响应函数(Nonlinear Output Frequency Response Functions,NOFRFs)分析故障参数的特性。利用方向梯度直方图(Histograms of Oriented Gradients,HOG)作为模板,采用可变形组件模型(Deformable Part Model,DPM)算法实现HTGS的故障识别。试验表明,本研究方法处理2GB故障数据所耗时间为40s。
关键词：视觉故障识别;可编程逻辑控制器;水轮机调节系统;模糊控制;可变性组件模型
PLC hardware structure hydropower station fault monitoring system and fault identification
Abstract：Aiming at the problems of low identification efficiency and high difficulty of the existing fault monitoring system of hydropower stations, a fault monitoring method based on the visual recognition of the inspection robot is proposed and applied to the fault of the Hydro-Turbine Governing System (HTGS) Diagnose the problem. The programmable logic controller (Programmable Logic Controller, PLC) ensures the stable operation of the inspection robot, and analyzes the characteristics of the fault parameters through the nonlinear output frequency response functions (Nonlinear Output Frequency Response Functions, NOFRFs). Using Histograms of Oriented Gradients (HOG) as a template, the Deformable Part Model (DPM) algorithm is used to realize the fault identification of HTGS. Experiments show that the method used in this study to process 2GB fault data takes 40s.
Key words：Visual fault identification; Programmable logic controller; Turbine governing system; Fuzzy control; Variability component model
收稿日期：2021-10-29
基金项目：
