基于遗传神经网络的柴油机故障诊断技术研究
|
摘要
本文在分析国内外智能诊断技术研究现状及柴油机故障诊断诸多方法的基础上,给出了一种基于遗传神经网络的柴油机故障智能诊断方法,并且对柴油机燃油系统故障诊断问题进行了深入研究。
首先,论文对故障诊断技术的研究现状及课题的研究内容进行了简要的论述,介绍了柴油机故障诊断的几种常见方法,并分析了神经网络故障诊断方法存在的问题。
其次,针对BP神经网络收敛速度慢和容易陷入局部极小值问题,将遗传算法和神经网络有机结合,利用遗传算法的全局搜索能力,优化网络的初始权值和阀值,解决其本身固有的两个缺陷,进而提高了BP神经网络诊断故障时的准确性和快速性。
最后,在MATLAB环境下进行仿真试验,把柴油机燃油系统典型故障数据作为遗传神经网络的训练样本,构建及训练网络,并对模拟故障进行诊断分析。仿真试验结果表明,基于遗传神经网络的故障诊断结果与实测值具有良好的一致性,只要选择足够典型的原始故障样本训练遗传神经网络,网络的稳定性就较好。基于遗传神经网络的故障模式识别方法能充分利用信息特征,实现输入与输出之间的映射关系,得出准确的诊断结果。
Based on the analysis of domestic and foreign research about intelligent diagnosis technology and fault diagnosis methods of diesel engine,a method of diesel engine fault diagnosis based on genetic neural network is given in this paper.The fault diagnosis of fuel system is deeply studied.
Firstly,the research status of fault diagnosis technology and contents of the topic have been discussed briefly in this paper;several popular methods of diesel engine fault diagnosis are introduced;the shortcoming of neural network in fault diagnosis is analysized.
Secondly,for shortcomings of slow convergence rate and falling into local minimum easily of BP neural network,genetic algorithm and neural network are combined.Using the global search ability of genetic algorithm,initial weights and thresholds of neural network are optimized and the two inherent shortcomings of network are solved,which has enhanced the accuracy and rapidity of neural network in fault diagnosis.
Lastly,simulation test is carried on by the MATLAB,the network is constructed and trained using typical fault data of fuel system as genetic neural network training samples and simulation fault is diagnosed and analysized.The simulation results show that the fault diagnosis result based on genetic neural network is well consistent with measured values.As long as we choose enough typical initial fault samples to train genetic neural network,the stability of network is better.The method of fault pattern recognition based on genetic neural network can fully use information characteristics, achieve mapping relation between input and output and get accurate result.
首先,论文对故障诊断技术的研究现状及课题的研究内容进行了简要的论述,介绍了柴油机故障诊断的几种常见方法,并分析了神经网络故障诊断方法存在的问题。
其次,针对BP神经网络收敛速度慢和容易陷入局部极小值问题,将遗传算法和神经网络有机结合,利用遗传算法的全局搜索能力,优化网络的初始权值和阀值,解决其本身固有的两个缺陷,进而提高了BP神经网络诊断故障时的准确性和快速性。
最后,在MATLAB环境下进行仿真试验,把柴油机燃油系统典型故障数据作为遗传神经网络的训练样本,构建及训练网络,并对模拟故障进行诊断分析。仿真试验结果表明,基于遗传神经网络的故障诊断结果与实测值具有良好的一致性,只要选择足够典型的原始故障样本训练遗传神经网络,网络的稳定性就较好。基于遗传神经网络的故障模式识别方法能充分利用信息特征,实现输入与输出之间的映射关系,得出准确的诊断结果。
Based on the analysis of domestic and foreign research about intelligent diagnosis technology and fault diagnosis methods of diesel engine,a method of diesel engine fault diagnosis based on genetic neural network is given in this paper.The fault diagnosis of fuel system is deeply studied.
Firstly,the research status of fault diagnosis technology and contents of the topic have been discussed briefly in this paper;several popular methods of diesel engine fault diagnosis are introduced;the shortcoming of neural network in fault diagnosis is analysized.
Secondly,for shortcomings of slow convergence rate and falling into local minimum easily of BP neural network,genetic algorithm and neural network are combined.Using the global search ability of genetic algorithm,initial weights and thresholds of neural network are optimized and the two inherent shortcomings of network are solved,which has enhanced the accuracy and rapidity of neural network in fault diagnosis.
Lastly,simulation test is carried on by the MATLAB,the network is constructed and trained using typical fault data of fuel system as genetic neural network training samples and simulation fault is diagnosed and analysized.The simulation results show that the fault diagnosis result based on genetic neural network is well consistent with measured values.As long as we choose enough typical initial fault samples to train genetic neural network,the stability of network is better.The method of fault pattern recognition based on genetic neural network can fully use information characteristics, achieve mapping relation between input and output and get accurate result.
引文
[1]李增芳.基于人工智能和虚拟仪器技术的发动机故障诊断专家系统研究:(博士学位论文).杭州:浙江大学,2005.7.
[2]Morgan.Predicting Future States With $n$ -Dimensional Markov Chains for Fault Diagnosis.Industrial Electronics,2009.5(56):1774-1881.
[3]吴欠欠,王直.故障树分析法在船舶柴油机故障诊断中的应用研究.机械设计与制造,2009.1(1):77-78.
[4]Yang P&Yam L.Theoretical and experimental study on fault diagnosis of gearbox.Integrating Dynamics,Condition Monitoring and Control for the 21st Century,1999,9:117-121.
[5]陈祥光,裴旭东.人工神经网络技术及应用.北京:中国电力出版社,2003.9.
[6]黄勇,郭晓平.基于改进BP神经网络的柴油机故障诊断研究.汽车科技,2009.3(2):55-58.
[7]吴仕勇.基于数值分析计算方法的BP神经网络及遗传算法的优化研究:(硕士学位论文).昆明:云南师范大学,2006.6.
[8]王洪元,史国栋.人工神经网络技术及其应用.北京:中国石化出版社,2002.12.
[9]王静伟.BP神经网络改进算法的研究.中国水运,2008.8(1):157-158.
[10]阎凡平,张长水.人工神经网络与模拟进化计算.北京:清华大学出版社,2000.
[11]Luo Zilai.Research on Fault Diagnosis Method of Diesel Engine Based on BP Network.Vehiche Engine,2008.6(176):119-123.
[12]袁新科.基于BP神经网络的发动机故障诊断研究:(硕士学位论文).济南:山东大学,2006.9.
[13]吴义虎.汽车发动机技术中的神经网络方法.北京:人民交通出版社,2000.6.
[14]Simon Haykin著,叶世伟,史忠植译.神经网络原理.北京:机械工业出版社,2004.
[15]吴微.神经网络计算.北京:高等教育出版社,2003.9.
[16]Ren Yunpeng.Approach to diesel engine fault diagnosis based on crankshaft angular acceleration measurement and its realization.Mechatronics and its realization IEEE International Conference,2005(3):1451-1454.
[17]胡守仁.神经网络导论.长沙:国防科技大学出版社,1993.
[18]杨世友,顾宏中,郭中朝.柴油机涡轮增压系统研究现状与进展.柴油机·Diesel Engine,2001(4):1-5.
[19]王小平,曹立明.遗传算法理论、应用与软件实现.西安:西安交通大学出版社,2002(1):7-78。
[20]Qin Huang.Research of motor fault diagnosis based on the improved genetic algorithm and BP network.Intelligent Control and Automation,2008.6(1)3131-3135
[21]崔玲丽,肖志权.实数编码遗传操作机制的研究.系统仿真学报,2003,15:11-15.
[22]Nakashima.Performance evaluation of combined cellular genetic algorithms for function optimization problems.Computational Interlligence in Robotics and Automation,2003.7(1):295-299.
[23]穆丽婉.柴油机常见故障诊断技术的智能化研究:(硕士学位论文).大连:大连铁道学院,2003.12.
[24]Murphy.L.A genetic algorithm tracking model for product deployment in telecom services.Circuits and System,2005.8(2):1729-1732.
[25]马守明.柴油机原理构造使用维修故障分析.北京:煤炭工业出版社,1981.
[26]曹龙汉.柴油机智能化故障诊断技术.北京:国防工业出版社,2005.
[27]Murphy.B.J;Lebold.Diagnostic fault detection and intelligent reconfiguration of fuel delivery systems.Aerospace Conference IEEE,2005.3:3514-3522.
[28]张宪波.基于虚拟仪器技术的柴油机燃油喷射系统故障诊断的研究:(硕士学位论文).杭州:浙江大学,2005.2.
[29]潜伟建,夏志澜.柴油机燃油系统故障诊断研究.柴油机·Diesel Engine,2002.1:12-17.
[30]杜伟肖.基于模糊理论对柴油机燃油系的故障诊断:(硕士学位论文).南宁:广西大学,2006.6.
[31]赵芳.基于模糊模式识别的船用柴油机状态监测和故障诊断:(硕士学位论文).济南:山东大学,2005.5.
[32]白雷石,刘玫.基于模式识别的柴油机故障诊断.控制工程,2002,9(6):87-80.
[33]杨文位.柴油机故障诊断的现代方法与展望.农机使用与维修,2007,1:78-79.
[34]Longhan Cao.The research of fault diagnosis for fuel injection system of diesel engine with ANN based on rough sets theory.Intelligent Control and Automation,2002.6(1):410-414.
[35]张铮.MATLAB程序设计与实例应用.北京:中国铁道出版社,2003.11.
[36]薛定宇.基于MATLAB/Simulink的系统仿真技术与应用.北京:清华大学出版社,2002.
[2]Morgan.Predicting Future States With $n$ -Dimensional Markov Chains for Fault Diagnosis.Industrial Electronics,2009.5(56):1774-1881.
[3]吴欠欠,王直.故障树分析法在船舶柴油机故障诊断中的应用研究.机械设计与制造,2009.1(1):77-78.
[4]Yang P&Yam L.Theoretical and experimental study on fault diagnosis of gearbox.Integrating Dynamics,Condition Monitoring and Control for the 21st Century,1999,9:117-121.
[5]陈祥光,裴旭东.人工神经网络技术及应用.北京:中国电力出版社,2003.9.
[6]黄勇,郭晓平.基于改进BP神经网络的柴油机故障诊断研究.汽车科技,2009.3(2):55-58.
[7]吴仕勇.基于数值分析计算方法的BP神经网络及遗传算法的优化研究:(硕士学位论文).昆明:云南师范大学,2006.6.
[8]王洪元,史国栋.人工神经网络技术及其应用.北京:中国石化出版社,2002.12.
[9]王静伟.BP神经网络改进算法的研究.中国水运,2008.8(1):157-158.
[10]阎凡平,张长水.人工神经网络与模拟进化计算.北京:清华大学出版社,2000.
[11]Luo Zilai.Research on Fault Diagnosis Method of Diesel Engine Based on BP Network.Vehiche Engine,2008.6(176):119-123.
[12]袁新科.基于BP神经网络的发动机故障诊断研究:(硕士学位论文).济南:山东大学,2006.9.
[13]吴义虎.汽车发动机技术中的神经网络方法.北京:人民交通出版社,2000.6.
[14]Simon Haykin著,叶世伟,史忠植译.神经网络原理.北京:机械工业出版社,2004.
[15]吴微.神经网络计算.北京:高等教育出版社,2003.9.
[16]Ren Yunpeng.Approach to diesel engine fault diagnosis based on crankshaft angular acceleration measurement and its realization.Mechatronics and its realization IEEE International Conference,2005(3):1451-1454.
[17]胡守仁.神经网络导论.长沙:国防科技大学出版社,1993.
[18]杨世友,顾宏中,郭中朝.柴油机涡轮增压系统研究现状与进展.柴油机·Diesel Engine,2001(4):1-5.
[19]王小平,曹立明.遗传算法理论、应用与软件实现.西安:西安交通大学出版社,2002(1):7-78。
[20]Qin Huang.Research of motor fault diagnosis based on the improved genetic algorithm and BP network.Intelligent Control and Automation,2008.6(1)3131-3135
[21]崔玲丽,肖志权.实数编码遗传操作机制的研究.系统仿真学报,2003,15:11-15.
[22]Nakashima.Performance evaluation of combined cellular genetic algorithms for function optimization problems.Computational Interlligence in Robotics and Automation,2003.7(1):295-299.
[23]穆丽婉.柴油机常见故障诊断技术的智能化研究:(硕士学位论文).大连:大连铁道学院,2003.12.
[24]Murphy.L.A genetic algorithm tracking model for product deployment in telecom services.Circuits and System,2005.8(2):1729-1732.
[25]马守明.柴油机原理构造使用维修故障分析.北京:煤炭工业出版社,1981.
[26]曹龙汉.柴油机智能化故障诊断技术.北京:国防工业出版社,2005.
[27]Murphy.B.J;Lebold.Diagnostic fault detection and intelligent reconfiguration of fuel delivery systems.Aerospace Conference IEEE,2005.3:3514-3522.
[28]张宪波.基于虚拟仪器技术的柴油机燃油喷射系统故障诊断的研究:(硕士学位论文).杭州:浙江大学,2005.2.
[29]潜伟建,夏志澜.柴油机燃油系统故障诊断研究.柴油机·Diesel Engine,2002.1:12-17.
[30]杜伟肖.基于模糊理论对柴油机燃油系的故障诊断:(硕士学位论文).南宁:广西大学,2006.6.
[31]赵芳.基于模糊模式识别的船用柴油机状态监测和故障诊断:(硕士学位论文).济南:山东大学,2005.5.
[32]白雷石,刘玫.基于模式识别的柴油机故障诊断.控制工程,2002,9(6):87-80.
[33]杨文位.柴油机故障诊断的现代方法与展望.农机使用与维修,2007,1:78-79.
[34]Longhan Cao.The research of fault diagnosis for fuel injection system of diesel engine with ANN based on rough sets theory.Intelligent Control and Automation,2002.6(1):410-414.
[35]张铮.MATLAB程序设计与实例应用.北京:中国铁道出版社,2003.11.
[36]薛定宇.基于MATLAB/Simulink的系统仿真技术与应用.北京:清华大学出版社,2002.