基于数据分析的设备故障预警与诊断技术在燃气电厂的应用

doi:10.3969/j.issn.1002-302x.2023.01.009

摘要/Abstract

摘要： 大数据、人工智能、边缘计算等先进技术不断进步，在燃气电厂利用基于数据分析的设备故障预警与诊断模型辅助检维修的优势逐步显现。AI设备故障预警与诊断模型依托统一的云平台搭建，覆盖燃气电厂燃机、汽机、锅炉和电气系统等多类主辅机设备，利用大数据分析技术和人工智能算法，主要算法包括分段函数、多元回归、深度神经网络等。模型以燃气电厂各系统设备的实时和历史数据为基础，将业务关联较强的数据与电厂内部专家经验相结合，实现设备参数劣化预警。文章结合燃气电厂AI模型应用实例，对建模过程、输入输出参数设定、训练样本选择、预警信息及故障诊断等进行说明与分析。利用AI设备故障预警与诊断模型可实时监督机组设备运行情况，在设备出现异常或故障的早期阶段进行智能预警和诊断，为机组设备运行安全提供保障。

关键词: 燃气电厂, 故障预警与诊断, AI模型, 大数据, 人工智能

Abstract: Progress is continuously made in big data, artificial intelligence and edge computing. The advantages of using the data-based analysis model of early-warning and diagnosis of equipment failure for auxiliary maintenance are gradually appearing at the gas-fired power plant. The AI early-warning and diagnosis model for equipment failure depends on a unified cloud platform, which covers various kinds of main and auxiliary equipment in the systems of gas turbine, steam generator, boiler and electricity. Big data analysis technology and AI algorithm mainly include piecewise function, multiple regression and in-depth neural network. Based on the real-time and historical data of various systems and equipment at the gas-fired power plant, the model combines the business-related data with the experiences of experts inside the power plant to bring about early warning of deteriorating equipment parameters. Based on the application case of the AI model at the gas-fired power plant, this paper elaborates and analyzes the modelling process, design of the input and output parameters, selection of the training prototype, early-warning information and diagnosis of failures. The AI early warning and diagnosis model for equipment failure can be used to monitor operation of the units in real time and make warning and diagnosis at the early stage of equipment abnormality and failure, maintaining operational safety of the units.

中图分类号:

TM621
TP18

何洪　阮航. 基于数据分析的设备故障预警与诊断技术在燃气电厂的应用[J]. 石油科技论坛, 2023, 42(1): 61-66.

He Hong, Ruan Hang. Data Analysis-based Technology for Early-warning and Diagnosis of Equipment Failure Used in Gas-fired Power Plant[J]. Petroleum Science and Technology Forum, 2023, 42(1): 61-66.

参考文献

[1]　单彤文. 天然气发电在中国能源转型期的定位与发展路径建议[J]. 中国海上油气, 2021, 33(2): 205-214.
Shan Tongwen. Positioning and development path suggestions of natural gas power generation in China’s energy transition period[J]. China Offshore Oil and Gas, 2021, 33(2): 205-214.
[2]　曹闯明,王同良. 中国海油燃气电厂智慧化发展之路[J]. 石油科技论坛, 2022, 41(5): 24-32.
Cao Chuangming, Wang Tongliang. Intelligent development of CNOOC gas-fired power plant[J]. Petroleum Science and Technology Forum, 2022, 41(5): 24-32.
[3]　王仲. 燃气-蒸汽联合循环机组智能诊断与健康维护技术研究[D]. 北京: 华北电力大学, 2021.
Wang Zhong. Research on technologies of intelligent diagnosis and health maintenance for gas-steam combined cycle unit[D]. Beijing: North China Electric Power University, 2021.
[4]　吕玥婷, 陈海文, 黄月丽, 等. 基于多轴燃气-蒸汽联合循环机组启停过程运行评价系统的研究及应用[J]. 仪器仪表用户,
0.420.350.300.250.210.4030日16:4031日00:0031日12:001日00:001日12:001日16:40入口滤网实际压力/kPa时间2021, 28(8): 94-97.
Lv Yueting, Chen Haiwen, Huang Yueli, et al. Research and application of operation evaluation system based on start-stop process of multiaxial gas-steam combined cycle unit[J]. Instrumentation Customer, 2021, 28(8): 94-97.
[5]　黄腾达.基于大数据分析的故障诊断方法研究[D]. 北京: 北京工业大学, 2019.
Huang Tengda. Fault diagnosis method based on big data analysis[D]. Beijing: Beijing University of Technology, 2019.
[6]　贾志军, 白德龙, 宋燕杰, 等. 基于KPCA算法的电厂设备在线故障诊断研究[J]. 华电技术, 2021, 43(8): 48-53.
Jia Zhijun, Bai Delong, Song Yanjie, et al. Research on on-line fault diagnosis and treatment of power plant equipment based on KPCA[J]. Huadian Technology, 2021, 43(8): 48-53.
[7]　杨承佐, 李想, 柳泓羽, 等. 基于大数据技术的发电厂一体化管理平台的构建与应用[J]. 自动化博览, 2017, 34(11): 72-74.
Yang Chengzuo, Li Xiang, Liu Hongyu, et al. Construction and application of integrated management platform for power plant based on large data technology[J]. Automation Panorama, 2017, 34(11): 72-74.
[8]　赵立慧. 火电厂引风机故障预警与诊断系统研究[D]. 北京: 华北电力大学, 2020.
Zhao Lihui. Research on fault warning and diagnosis system of induced draft fan in thermal power plant[D]. Beijing: North China Electric Power University, 2020.
[9]　张哲文. 基于相关向量机的两种电力设备故障预警研究[D]. 济南: 山东大学, 2020.
Zhang Zhewen. Research on fault early warning method for two power equipments based on relevance vector machine[D]. Jinan: Shandong University, 2020.
[10] 王守鹏, 赵冬梅. 电网故障诊断的研究综述与前景展望[J]. 电力系统自动化, 2017, 41(19): 166-170.
Wang Shoupeng, Zhao Dongmei. Research review and prospects for power grid fault diagnosis[J]. Automation of Electric Power System, 2017, 41(19): 166-170

[1]	曹闯明　王同良. 中国海油燃气电厂智慧化发展之路[J]. 石油科技论坛, 2022, 41(5): 24-32.
[2]	胡贵　崔明月　陶冶　黄雪琴　张国辉　石李保. 油气井筒工程数据平台技术进展及数据深度应用思考[J]. 石油科技论坛, 2021, 40(5): 65-72.
[3]	闫娜　李庆. 贝克休斯远程作业发展历程及启示[J]. 石油科技论坛, 2021, 40(5): 79-84.
[4]	曹刚　杨清海　任义丽　赵瑞东. 五大信息技术在油气行业数字化转型中的应用探讨[J]. 石油科技论坛, 2020, 39(6): 27-36.
[5]	光新军　王敏生　耿黎东　闫娜　皮光林. 人工智能技术发展对石油工程领域的影响及建议[J]. 石油科技论坛, 2020, 39(5): 41-47.
[6]	刘卓　张宇　张宏洋. 国内外数字油田技术发展趋势及策略[J]. 石油科技论坛, 2020, 39(4): 62-67.
[7]	李剑峰. 智慧石化建设：从信息化到智能化[J]. 石油科技论坛, 2020, 39(1): 34-42.
[8]	贾鹿　牛志杰　石国伟　李嗣旭　林道寿. 油气上游领域智能化发展方向探析[J]. 石油科技论坛, 2019, 38(6): 34-42.
[9]	杨午阳　魏新建　何欣. 应用地球物理+AI 的智能化物探技术发展策略[J]. 石油科技论坛, 2019, 38(5): 40-47.
[10]	李　斌　刘　伟　毕永斌　李　力. 浅议油田开发战略管理[J]. 石油科技论坛, 2019, 38(4): 24-29.
[11]	闫　娜. 石油工程数字化发展现状与策略分析[J]. 石油科技论坛, 2019, 38(2): 64-70.
[12]	廖　晋　金云智　吴迅达　梁　豪. 海上油气勘探目标全生命周期管理创新与实践[J]. 石油科技论坛, 2018, 37(5): 9-13.
[13]	刘　伟　闫　娜. 人工智能在石油工程领域应用及影响[J]. 石油科技论坛, 2018, 37(4): 32-40.
[14]	耿玉广　张　萌　耿英杰　薛良　张战敏　顾龚杰. 油田生产大数据分析应用研究[J]. 石油科技论坛, 2018, 37(4): 41-48.
[15]	万　宏　陆如泉　单卫国　江　河　张　震　王　曦　胡　勇　蒋　平. 从“剑桥能源周” 看未来能源战略走势[J]. 石油科技论坛, 2018, 37(4): 55-61.