石油化工设备技术 ›› 2022, Vol. 43 ›› Issue (1): 52-58,62.doi: 10.3969/j.issn.1006-8805.2022.01.011

• 状态监测与分析 • 上一篇    下一篇

面向流体机械的智能故障诊断系统设计

赵 鹏,陈兆龙,陈志立   

  1. 中国石油化工股份有限公司胜利油田分公司油气集输总厂,山东 东营 257000
  • 收稿日期:2021-10-18 接受日期:2021-12-29 出版日期:2022-01-18 发布日期:2022-01-18
  • 通讯作者: 陈志立 E-mail:zhaopeng376.slyt@sinopec.com
  • 作者简介:赵鹏,男,2000年毕业于山东工业大学环境工程专业,工学硕士,主要从事石油化工信息自动化设计与管理工作,高级工程师,已发表论文20余篇。

Design of Intelligent Fault Diagnosis System for Fluid Machinery

Zhao Peng, Chen Zhaolong, Chen Zhili   

  1. SINOPEC Shengli Oilfield Company, Dongying, Shandong, 257000
  • Received:2021-10-18 Accepted:2021-12-29 Online:2022-01-18 Published:2022-01-18
  • Contact: Chen Zhili E-mail:zhaopeng376.slyt@sinopec.com

摘要: 流体机械广泛应用于石油机械类行业,对其进行状态监测、诊断维护具有重要的实用价值。以旋转式流体机械的典型故障特征为对象,利用LabVIEW软件和MATLAB软件,设计了一套适用于旋转式流体机械的智能故障诊断系统。该系统除了具有常见旋转式流体机械故障诊断中的信号采集与处理功能、动平衡人机交互功能外,为了提高系统的智能化水平,还设计了基于遗传算法优化的BP神经网络智能故障识别系统。基于多功能转子实验台的测试结果显示,该诊断系统在旋转式流体机械故障问题诊断中,具有良好的故障识别率和准确率。上述研究表明,该测试系统具有精度高、功能齐全、可移植性和拓展性强等特点,可以较好地适用于旋转式流体机械故障诊断科研实践工作。

关键词: 旋转式流体机械, 故障诊断, 信号处理, LabVIEW, MATLAB, 人工神经网络

Abstract: Fluid machinery is widely used in petroleum machinery industry, so the condition monitoring, diagnosis and maintenance of rotating machinery are of important practical value. Taking the typical fault characteristics of rotary fluid machinery as the object, a set of intelligent fault diagnosis system suitable for rotary fluid machinery is designed using LabVIEW software and MATLAB software. In addition to signal acquisition and processing and dynamic balance human-computer interaction system of common fault diagnosis of rotary fluid machinery, an intelligent fault identification system based on BP neural network optimized by genetic algorithm is designed in order to improve the intelligent level of the system. The test results based on the multi-functional rotor test-bed show that this diagnostic system has good fault recognition rate and accuracy rate in the fault diagnosis of rotary fluid machinery. The test system has the characteristics of high precision, complete functions, portability and expansibility, which can be better applied to the fault diagnosis of rotary fluid machinery.

Key words: rotary fluid machinery, fault diagnosis, signal processing, LabVIEW, MATLAB, artificial neural network