摘要
摘要：能源工业是国民经济发展的基础工业,凝汽器作为电厂的重要设备之一,发展凝汽器系统运行参数的监测,实施故障诊断,对发电机组安全经济运行具有重要意义。本文就某600MW电站N-32400-1型凝汽器系统的状态监测与故障诊断问题开展研究,主要研究成果总结如下：
针对该型号凝汽器,将冷却塔和凝汽器看作一个整体,建立了凝汽器压力应达值的计算模型；通过对凝汽器主要性能指标的分析,提炼出系统的18个运行参数,利用电站机组DCS系统的采集数据,建立了单参预测、双参预测和集成神经网络预测三种BP神经网络模型,进行对比研究,实现了对运行参数的有效预测；利用单参预测模型对凝汽器真空进行了短、中、长期预测。根据凝汽器的实时数据,结合BP神经网络的参数预测结果,开发了凝汽器状态评价模块,实现对凝汽器更为全面及时的评价。
利用专家系统的故障诊断方法对凝汽器系统的常见故障进行研究,通过知识库规则的提取、机制的推理以及机制的解释,开发了凝汽器的故障诊断模块,可针对不同的测点征兆,判断出故障的原因,并给出消除相应故障切实可行的措施,从而对凝汽器的正常运行起到了一定的指导作用。
对600MW电站凝汽器状态监测与故障诊断系统的开发与实现是通过两个子系统来实现的,一个是利用VB调用Matrix Vb的方法开发的凝汽器诊断子系统,另一个是利用力控组态软件开发的凝汽器监测子系统。利用该系统不仅可以对凝汽器进行实时的监测与诊断,还可根据电站DCS系统的采集数据,对凝汽器进行离线的状态评价和故障诊断。
本文所开发的凝汽器状态监测与故障诊断系统可对凝汽器进行全面的评价,从而提出详细的诊断结果,给出切实可行的监督指导措施,对提高凝汽器的安全性和可靠性有积极的作用。
Engrgy industry is the primary industry of economic development. Condenser is one of the important equipments in the power plant, from which the energy could be saved. Its state would seriously influence the security and economy of the turbine. So, it is important to develop the condition monitoring and fault diagnosis. Condition monitoring and fault diagnosis are studied on N-32400-1 condenser.
The calculation model of exhaust pressure is established. Eighteen operating parameters are epurated from the system after analyzing its main influencing factors. Three kinds of BP neural network model were established by using the historical data. A comparison among these three kinds prediction is also studied. The single parameter prediction is used for short-term, medium-term and long-term projections. According to the state of condenser, combine with the result of BP neural network prediction, a evaluation module of the condenser state is developed for realizing a more comprehensive and timely assessment of condenser.
By using the expert system, a diagnosis system of the common faults of the condenser is developed through the extraction of rules', the studying and explanning of mechanisms'. According to the different signs of the measuring points, the reasons for failure could be judged, and the feasible steps could be provided to eliminate the faults. The system could give a good guide for the conderser.
The condition monitoring and fault diagnosis system for condenser of 600MW power plant is achieved through two subsystems:one is by using vb language, and the other is by using the software of force control. This system can not only realize the real-time condition monitoring and fault diagnosis for the condenser, but also can realize the offline evaluation and diagnosis.
This condition monitoring and fault diagnosis system for the condenser could comprehensive analysis the condenser, then put forward detailed analysis and give practical measures. It has a positive effect for improving the security and reliability of the condenser.
引文
[1]郭创新,朱传柏,曹一家,吴新.电力系统故障诊断的研究现状及发展趋势[J].电力系统自动化,2006,30(8)：98-103
[2]P.A.Pialavachi.Power generation with gas turbine systems and combined heat and power[J].Applied Thermal Engineering,2000(20):1421-1429
[3]程道来,吴茜,吕庭彦,等.国内电站故障诊断系统的现状及发展方向[J].动力工程,1999,19(1)：53-57
[4]韩璞.火电厂计算机监控与监测.汽轮机性能在线监测及分析系统[M].中国水利水电出版社.2005
[5]张洪奎.汽轮机高压加热器的故障诊断与运行[J].电站辅机,1997,(3)：4-9
[6]李录平.汽轮机组故障诊断技术[M].北京：中国电力出版社.2002,196-217
[7]士一,庆贺庆等.汽轮机原理[M].北京：中国电力出版社.1992,50-58
[8]杨善让.汽轮机凝气设备及运行原理[M].北京：水利电力出版社.1993
[9]康松,杨建明等.汽轮机原理[M].北京：中国电力出版社.2000
[10]Baozhuang Shi, Li Yang. An Investigation on the Influencing Factors on On-Line Insulation Monitoring of HV Apparatus[J].Conference Record of the IEEE International Symposium on Electrical Insulation. Anaheim. CA USA.2000,81-84
[11]任乐呜,李文清.机组状态监测与故障诊断系统在紧水滩水电厂的应用[J].水电能源科学,2006,24(2)：94-95
[12]高永昌,丁勇军,张鹏.基于规则推理的地空导弹智能故障诊断研究[J].装备制造技术,2007,(3)：12-14
[13]L.M. Saini, M.K. Soni. Artificial neural network based peak load forecasting using Levenberg-Marquardt and quasi-Newton methods [J]. IEEE Power Engineering Review.2002, 22(7):578-584
[14]骆贵兵,李崇祥.LM算法的回热系统故障诊断人工神经网络模型[J].热力发电,2004,33(10)： 15-18
[15]杨凡,赵建民,朱信忠.一种基于BP神经网络的车牌字符分类识别方法[J].计算机科学,2005,32(8)：192-195
[16]D.C. Park, M.A. EI-Sharkawi, R.J. Marks Ⅱ et al. Electric Load Forecasting Using An Artificial Neural Network[J]. IEEE Transactions on Power Systems.1991,6(2):442-449
[17]El-Sharkawi, M.A. Oh, S. Marks et al. Short term electric load forecasting using an adaptively trained layered perception[J]. Proceedings of the First International Forum on Applications of Neural Networks to Power Systems.1991:3-6
[18]Parisi, R. Di Claudio, E.D. Lucarelli et al. Car plate recognition by neural networks and image processing[J]. ISCAS98.1998:195-198
[19]祝晓燕,王继选,宋敏霞.基于模糊数学的内燃机故障诊断系统[J].煤矿机械.2007,28(1)：178-180
[20]许俊峰,白敏丽,吕继祖,周龙,刘书亮,许斯都.四气门柴油机缸内流场LDA实验数据的分析研究[J].内燃机学报,2007,25(3)：241-246
[21]武永锋.基于遗传算法的电力系统故障诊断[D].天津：天津大学.2003
[22]张燕,周志伟,董秀臣.核电厂实时故障诊断专家系统的设计与实现[J].原子能科学技术,2006,40(4)：420-423
[23]彭军,程卫平,刘济泉,王光辉.一种故障诊断方法在导弹测试中的应用[J].飞航导弹,2007(3)：19-22
[24]鄣晓橙,彭爱平,蔡伟,谭立龙.某武器系统电控设备的故障诊断专家系统[J].计算机工程,2006,32(14)：246-248
[25]L.Yi Hui. Evolutionary neural network modeling for forecasting the field failure data of repairable systems[J].Expert Systems with Applications.2007,33 (4):1090-1096
[26]贾淑洁.凝汽器及其相关设备的故障诊断[D].黑龙江：哈尔滨工程科大学.2004
[27]丁燕.凝汽器动态数学模型及故障诊断系统研究[D].湖北：武汉大学.2005
[28]岑小路.火电厂双进双出磨煤机故障诊断与维修决策研究[D].湖北：武汉大学.2004
[29]M.H. Fazel Zarand, B.Rezaee, B.Turksen, E.Neshat. A type-2 fuzzy rule-based expert system
model for stock price analysis[J]. doi:10.1016/j.eswa.2007.09.034
[30]Hei Chia Wang, Huei Sen Wang.A hybrid expert system for equipment failure analysis[J].
Expert Systems with Applications.2005.28(4).615-622
[31]沈振飞,郑美蓉.汽轮机辅助设备[M].北京：水力电力出版社.1987
[32]汽轮机运行规程[Z].福建永安火电厂
[33]盛德仁,任浩仁等.运行工况下汽轮机主要参数应达值的数值分析[J].热力发电.2000(3)：28-30
[34]于新颖.凝汽器运行参数对其性能的影响分析[J].国家电力公司西安热工研究院.71003
[35]盛德仁,任浩仁,李蔚等.运行工况下汽轮机主要参数应达值的数值分析[J].热力发电厂,2000(3)： 28-30
[36]金生祥,杜霄龙.汽轮机运行热耗的偏差算法[J].华北电力术,1997(12)：14-17
[37]史佑吉.冷却塔运行与试验[M].北京：水利电力出社.1990：52-72
[38]窦照英.凝汽器黄铜管内应力验收标准研究[J].华北电力技术.1994,4
[39]张卓澄.大型电站凝汽器[M].北京：机械工业出版社.1993,3
[40]Rovithakis G A, Gananis V.I, Perrakis S E.等Real-time control of manufacturing cells using dynamic neural networks[J].Automatica,1999,35(1):139-149
[41]舒怀林,李柱.基于PID神经元多层网络的多变量解耦控制系统[J].自动化仪表,1998,19(3)： 24-27
[42]韩力群.人工神经网络理论、设计及应用[M].北京：化学工业出版社.2007
[43]朱太奇,史慧.人工神经网络原理及应用[M].北京：科学出版社.2006
[44]阮晓钢.神经计算科学[M].北京：国防工业出版社.2006
[45]M T Hagan, M B Menhaj. Training feedforward networks with the marquardt algorithm[J], IEEE Trans Neural Net.1994.5(6):989-993
[46]M T Hagan, M B Menhaj. Training feedforward networks with the marquardt algorithm[J], IEEE Trans Neural Net.1994.5(6):989-993
[47]邵军力,张景,魏长华.人工智能基础[M].北京：电子工业出版社.2000
[48]力控用户手册.北京三维力控科技有限公司
[49]清源计算机工作室.Visual Basic 6.0开发宝典[M].北京：机械工业出版社.1999
[50]刘圣财,李春葆.Visual Basic 6程序设计导学[M].北京：清华大学出版社.2002
