PTA装置溶剂脱水过程实时故障诊断综合系统
|
摘要
本文针对目前化工流程工业故障诊断系统在诊断的过程中对缓变故障的低诊断率,对故障原因分析不足等问题进行研究和改进,提出采用符号有向图方法进行故障寻源,多种智能方法相结合的实时故障诊断系统结构,构建集预测、诊断、专家意见等功能相结合的故障诊断和处理框架。
首先以精对苯二甲酸(Pure terephthalia acid, PTA)溶剂脱水过程为背景,搭建符号有向图(Signed Directed Graph,SDG)模型,在传统的符号有向图模型的节点引入趋势基元思想,按时序过程顺序对化工生产过程数据进行实时趋势分析。将趋势明显、符合SDG方法处理的故障点进行寻源报警处理；将由生产调整、误差等引起的尖峰现象从故障检测中剔除；该系统在SDG故障检测溯源的同时对实时数据进行神经网络故障预测,当预测故障概率大于预报阈值时对生产过程发出故障预警；同时将实时数据运用绝对值差分累积和计算,对缓变故障和波动大、不稳定的变量有预防和警示作用,在故障发生事故前引起重视。该系统在故障诊断的同时根据故障和故障源的判断给出专家指导意见,用于故障排除。
经过某石化厂PTA生产线溶剂脱水过程实例分析,这种多种智能方法结合的诊断系统方案提高诊断框架完备性,使SDG的溯源范围更广,准确率更高,减少其多义性,专家意见及时提出建议,有效避免系统扰动等原因引起的误报,改进差分累积和计算对缓变故障和波动大、不稳定的变量有预防和警示作用。可见该方法一方面很好的利用了模型的因果关系,同时也有效地利用了数据历史趋势。
This paper research and solve the problem that the diagnosis is difficult for slow variation fault of the process industry fault diagnosis system, and the analysis for the reason of fault is insufficient. The system combines strengths of lots of artificial intelligence method, based on SDG method for tracing to the fault source. This fault diagnosis and dispose framework contains forecast, diagnosis, expert opinion etc. function.
First, this paper present the traditional SDG method model against the process of PTA solvent dehydration, and introduce the trend element ideology on some node of the SDG model, and the chemical industry process data is used to instantly analyzing automatically in the time series order. The fault point with obvious trend and according with the SDG method will be searched the source of fault, spike phenomenon because of production adjustment and error will be removed from fault detection. Meanwhile, the system will forecast fault based on real time data utilizing neural network, furthermore absolute difference cumulation is used for prevention and reminding of the slow change fault and major fluctuations variable, so the slow fault and the unstable variable can be forecasted and alerted before the accident occurred. This system can provide expert guidance based on the source of trouble to get rid of the fault.
Based on the practice for the pure terephthalic acid solvent dehydration in some petrochemical factory, this method with a variety of intelligence method improve the completeness of the framework, makes the process of SDG tracing to the fault source more availability and accuracy, decrease the ambiguity, avoid misinformation which brought by system disturbance. One hand this method utilize the cause-and-effect relationship of the model, on the other hand, it can make use of the history data information effectively.
首先以精对苯二甲酸(Pure terephthalia acid, PTA)溶剂脱水过程为背景,搭建符号有向图(Signed Directed Graph,SDG)模型,在传统的符号有向图模型的节点引入趋势基元思想,按时序过程顺序对化工生产过程数据进行实时趋势分析。将趋势明显、符合SDG方法处理的故障点进行寻源报警处理；将由生产调整、误差等引起的尖峰现象从故障检测中剔除；该系统在SDG故障检测溯源的同时对实时数据进行神经网络故障预测,当预测故障概率大于预报阈值时对生产过程发出故障预警；同时将实时数据运用绝对值差分累积和计算,对缓变故障和波动大、不稳定的变量有预防和警示作用,在故障发生事故前引起重视。该系统在故障诊断的同时根据故障和故障源的判断给出专家指导意见,用于故障排除。
经过某石化厂PTA生产线溶剂脱水过程实例分析,这种多种智能方法结合的诊断系统方案提高诊断框架完备性,使SDG的溯源范围更广,准确率更高,减少其多义性,专家意见及时提出建议,有效避免系统扰动等原因引起的误报,改进差分累积和计算对缓变故障和波动大、不稳定的变量有预防和警示作用。可见该方法一方面很好的利用了模型的因果关系,同时也有效地利用了数据历史趋势。
This paper research and solve the problem that the diagnosis is difficult for slow variation fault of the process industry fault diagnosis system, and the analysis for the reason of fault is insufficient. The system combines strengths of lots of artificial intelligence method, based on SDG method for tracing to the fault source. This fault diagnosis and dispose framework contains forecast, diagnosis, expert opinion etc. function.
First, this paper present the traditional SDG method model against the process of PTA solvent dehydration, and introduce the trend element ideology on some node of the SDG model, and the chemical industry process data is used to instantly analyzing automatically in the time series order. The fault point with obvious trend and according with the SDG method will be searched the source of fault, spike phenomenon because of production adjustment and error will be removed from fault detection. Meanwhile, the system will forecast fault based on real time data utilizing neural network, furthermore absolute difference cumulation is used for prevention and reminding of the slow change fault and major fluctuations variable, so the slow fault and the unstable variable can be forecasted and alerted before the accident occurred. This system can provide expert guidance based on the source of trouble to get rid of the fault.
Based on the practice for the pure terephthalic acid solvent dehydration in some petrochemical factory, this method with a variety of intelligence method improve the completeness of the framework, makes the process of SDG tracing to the fault source more availability and accuracy, decrease the ambiguity, avoid misinformation which brought by system disturbance. One hand this method utilize the cause-and-effect relationship of the model, on the other hand, it can make use of the history data information effectively.
引文
[1]邵惠鹤.流程工业自动化发展趋向与先进控制技术.自动化博览,2002,19(2)：27-32
[2]黄启明,钱宇,林伟璐,等.化工过程故障诊断研究进展.化工自动化及仪表,2000,27(3)：1-5
[3]Evenson G, Befus S, Dolfi M, et al. LOPA as a global manufacturing API facility. Process Safety Progress,2009,28(4):312-316.
[4]Center for Chemical Process Safety, American Institute of Chemical Engineers. Layer of Protection Analysis:Simplified Process Risk Assessment. New York,2001.
[5]张其立,邱彤.LOPA被动型保护层有效性研究.2010中国过程系统工程年会会议论文集(PSE2010)363-367
[6]周华,李秀喜,钱宇.石油化工过程安全技术研究进展[J].化工进展.2008,27(10)
[7]罗云,樊运晓,马晓春.风险分析与安全评价[M].北京：化学工业出版社,2004
[8]Chang M K. Application of risk based inspection in refinery and processing piping[J]. Journal of Loss Prevention in the Process Industries,2005,18:397-402.
[9]Santosh. Reliability analysis of pipelines carrying H2S for risk based inspection of heavy water plants [J]. Reliability Engineering and System Safety,2006(91):163-170.
[10]吴重光.系统建模与仿真.清华大学出版社,2008,252-264
[11]L.N. de Castro. Immune, swarm, and evolutionary algorithms Part I:Basic models. Proceedings of the International Conference on Neural Information Processing, Workshop on Artificial Immune Systems.2002,3:1464-1468.
[12]American Petroleum Institute. Base resource documentation for risk based inspection (1st ed) [C].American petroleum Institute, Washington DC, USA:API Publication 581,2000.
[13]L. N. de Castro, F. J Von Zuben. The clonal selection algorithm with engineering applications. Proceedings of GECCO'00, Workshop Proceedings.2000:36-37
[14]Li X X, Qian Y, Wang J F. Process monitoring based on wavelet packet principal component analysis [J]. European Symposium on Computer Aided Process Engineering, 2004,13:455-460.
[15]王春利,张卫华,吴重光.故障诊断技术理论与发展趋势及其在石化领域的应用.石油化工自动化,2008,6：7-13
[16]Frank P M. New developments using AI in fault diagnosis[J]. Engineering Application of Artificial Intelligence,1997,10(1):3-14.
[17]Venkatasubramanian V, Rengaswamy R, Kavuri S N. A review of process fault detection and diagnosis (Part II):Qualitative models and search strategies [J]. Computers and Chemical Engineering,2003(27):313-326
[18]Venkatasubramanian V, Rengaswamy R, Kavuri S N, et al. A review of process fault detection and diagnosis (Part Ⅲ):Process history based methods [J]. Computers and Chemical Engineering,2003(27):327-346.
[19]郭明.基于数据驱动的流程工业性能监控与故障诊断研究.2004年浙江大学博士学位论文.
[20]张贝克,张海洋,马昕.定性趋势分析在过程故障诊断中的应用研究[J].系统仿真学报.2008,20(10)
[21]Kourti T, MacGreor J F. Multivariate SPC methods for process and product monitoring [J]. Journal of Quality Technology,1996,28(4):409-428.
[22]何学文.基于支持向量机的故障智能诊断理论与方法研究[D].长沙：中南大学博士学位论文,2004.
[23]DASH S, VENKATASUBRAMANIAN V. Challenges in the industrial applications of fault diagnosis system[J]. Computer and Chemical Engineering,2000,24(2-7):785-791.
[24]郑小霞.大型PTA生产过程智能故障预报与优化操作研究.华东理工大学博士学位论文.
[25]邓晓刚,田学民.生产过程故障诊断方法研究现状及展望[J].石油大学学报(自然科学版),2005,29(6)：151-157.
[26]丁艳军,王培红,吕震中,等.生产过程早期故障检测与诊断的一种新方法[J].中国电机工程学报,2000,20(3)：61-70.
[27]Zhijie Zhou, Changhua Hu, Hongdong Fan, Jin Li. Fault prediction of the nonlinear systems with uncertainty. Simulation Modelling Practice and Theory.2008,16:690-703.
[28]K. S. Lu, R. Saeks. Failure prediction for an on-line maintenance system in a poisson shock environment, IEEE Transaction on Systems. Man and Cybernetics. 90(1979)356-362.
[29]Andre B. de Carvalho, Aurora Pozo. A symbolic fault-prediction model based on multi-objective particle swarm optimization. System and Software.2010,83 (5): 868-882.
[30]左宪章,康健,李浩.故障预测技术综述.火力与指挥控制.2010,35(1)：1-5.
[31]Zhijie Zhou, Changhua Hu, Hongdong Fan. Fault prediction of the nonlinear systems with uncertainty. Simulation Modelling Practice and Theory.2008,16(6):690-703.
[32]Cagatay Catal, Banu Diri. A systematic review of software fault prediction studies. Expert Systems with Application 2009,36(4):7346-7354.
[33]Cagatay Catal, Banu Diri. Investigating the effect of dataset size, metrics sets, and feature selection techniques on software fault prediction problem.2009 179(8): 1040-1048.
[34]Kim, Hoyong. Intelligence control of vehicle dynamic system by artificial neural network. Transportation Research Part A:Policy and Practice 1997 31(1):63-66
[35]Lo Shih chung B. Artificial convolution neural network for medical image pattern recognition. Neural Networks,1995 8(7-8):1201-1214
[36]王旭,王宏,王文辉.人工神经元网络原理与应用.沈阳：东北大学出版.2000：25-35
[37]孙明.基于神经网络的抽油井故障诊断系统.中国石油大学学位论文.
[38]魏勇.基于具有模糊输出BP网络的决策分类模型的研究.北京工业大学硕士学位论文.2002
[39]Aoki T, Ceravolo R, Stefano A Deetal.Seismic vulnerability assessment of chemical plants through probabilistic neural networks. Reliability Engineering and System Safety.2002,77(9):263-268.
[40]Ren Puli, Zheng Ouwang. Mining classification rules using rough sets and neural networks. European Journal of Operational Research,2004(157):439-448
[41]Foster W., Collopy F., Ungar L. H. Neural networks forecasting of short noisy time series. Computers & Chemical Engineering,1992,16(4):293-298
[42]吴宇.GSM-R越区切换的网络优化专家系统研究与开发.北京交通大学硕士学位论文.
[43]刘静.同步齿轮箱初始故障在线检测与诊断专家系统的研究.武汉科技大学硕士学位论文.
[44]刘长龙,马听,张贝克.基于SDG模型的故障诊断专家系统.计算机工程.2009,35(19)：233-235
[45]刘胜,王宇超,傅荟璇.船用计程仪故障诊断模糊专家系统.计算机工程.2009,35(18)：227-230.
[46]张广斌,葛运建,方凯.神经网络和专家系统在变速器检测中的应用.华中科技大学学报：自然科学版.2008,36(增刊Ⅰ)：53-56.
[47]马笑潇,黄席樾,柴毅.基于支持向量机的故障过程趋势预测研究[J].系统仿真学报,2002,14(11)：1548—1551
[48]Lapp S A, Powers G J. Computer-aided Synthesis of Fault trees. IEEE Transactions on Reliability.1977,26(2):2-12
[49]M. Iri, K Aoki, E.0'shima, H Matsuyama. An Algorithm for Diagnosis of System Failures in the Chemical Process. Computer & Chemical Engineering,1979,3:489-493.
[50]Shiozaki J, Matsuyama H, Tano K, O'shima E. Fault Diagnosis of Chemical Processes by the use of Signed, Directed Graphs. Extension to Five-range Patterns of Abnormality. International Chemical Engineering,1985,25(4):651-659.
[51]Yu C C, Lee C. Fault Diagnosis Based on Qualitative/Quantitative Process Knowledge. AIChE Journal,1991,37(4):617-628.
[52]Tarifa E, Scenna N. Fault Diagnosis, Direct Graphs, and Fuzzy Logic. Computers and Chemical Engineering,1997, (21):649-654.
[53]Vedam H, Venkatasubramanian V. PCA-SDG Based Process Monitoring and Fault Diagnosis. Control Engineering Practice,1999, (7):903-917.
[54]黄道平,龚婷婷,曾辉.基于非线性主元分析和符号有向图的故障诊断方法.化工学报,2009,60(12)：3058-3062.
[55]Vedam H, Venkatasubramanian V. Signed Digraph Based Multiple Fault Diagnosis. Computers and Chemical Engineering,1997,21:655-660.
[56]Maurya M R, Rengaswamy R, Venkatasubramaniana V. Application of Signed Digraphs-based Analysis for Fault Diagnosis of Chemical Process Flowsheets. Engineering Application of Artificial Intelligence,2004,17:501-518.
[57]Maurya M R, Rengaswamy R, Venkatasubramaniana V. Incipient fault diagnosis of ennessee Eastman flowsheet using signed directed graph and trend analysis. Proceedings of the 12th European Symposium on Computer Aided Process Engineering (ESCAPE-12),2002.
[58]Maurya M R, Rengaswamy R, Venkatasubramaniana V. A signed directed graph-based systematic framework for chemical engineering science,61,2006:1790-1810
[59]王春利,张卫华,吴重光.故障诊断技术理论与发展趋势及其在石化领域的应用.石油化工自动化,2008,6：7-13.
[60]吴重光,夏涛,张贝克等.基于符号定向图(SDG)深层知识模型的定性仿真.系统仿真学报,2003,15(10)：1351-1355.
[61]杨帆,肖德云.概率SDG模型及故障分析推理.控制与决策,2006,21(5)：487-491.
[62]杨帆,肖德云.结构残差在基于SDG故障分离中的应用.控制工程,2007, 14(3)：320-324.
[63]吴重光.系统建模与仿真.北京：清华大学出版社,2008.
[64]杨帆,萧德云.SDG建模及其应用的进展.控制理论与应用.2005,22(5)：767-774.
[65]UMEDA T, KURUYAMA T. A graphical approach to cause and effect analysis of chemical processing systems[J]. Chemical Engineering Science,1980,35(12): 2379-2388.
[66]LAPP S A, POWERS G J. Computer-aided synthesis of fault-trees [J]. IEEE Trans on Reliability,1977,26(2):2-12.
[67]石宇,邱彤,陈丙珍.用于化工过程的SDG故障分析方法.化工进展.2006,25(12)：1484-1488.
[68李安峰,夏涛,张贝克等.化工过程建模SDG方法[J].系统仿真学报,2003,15(10)：1364-1368.
[69]Cheung J T, Stephanopoulos G.Representation of Process Trends-Part I:A Formal Representation Framework [J]. Computers and Chemical Engineering,1990,14(4-5): 495-510
[70]SHIOZKI J, SHIBATA B, MATSUYAMA H. Fault diagnosis of chemical processes utilizing signed directed graphs-improvement by using temporal information [J]. IEEE Transacations on Industrial Electronics,1989,36(4):469-474
[71]DASH S, RENGASWAMY R, VENKATASUBRAMANIAN V. A novel interval-halving framework for automated identification of process trends [J]. American Institute of Chemical Engineers,2004,50(1):149-162.
[72]J.T.Y Cheung, G Stephanopoulos. Representation of process trends-part I. A formal representation framework[J]. Computers and Chemical Engineering (S0098-1354), 1990,14(4):495-510.
[73]M. R. Maurya, R. Rengaswamy, V. Venkatasubramanian. A signed directed graph and qualitative trend.analysis-based framework for incipient fault diagnosis. Chemical Engineering Research and Design,2007,85(10):1407-1422
[74]Dong Gao, Chongguang Wu, Beike Zhang, Xin Ma. Signed directed graph and qualitative trend analysis based fault diagnosis in chemical industry. Chinese Journal of Chemical Engineering,2010,18(2):265-276
[75]李希,谢刚.PTA技术国产化中的主要化学工程问题及其研究思路.聚酯工业,2001,14(1)：1-7
[76]Lindahl H. A., Abrams K. J., Stark L. E., et al. Method and apparatus for controlling the manufacture of terephthalic acid to control the level and variability of the contaminant content and the optical density. US Patent.4835307
[77]周红军,郭曦虹.PTA市场分析与扬子石化的营销策略.当代石油石化,2005,13(6)：24-27
[78]王丽军,李希,谢刚,等.对二甲苯液相催化氧化动力学(Ⅰ)反应机理和动力学模型.化工学报,2003,54(7)：946-952
[79]罗云,樊运晓,马晓春.风险分析与安全评价[M].北京：化学工业出版社,2004
[80]吴海桥.现代大型客机故障预报专家系统的研究与开发.南京航空航天大学博士学位论文,2002
[81]郑小霞.大型PTA生产过程智能故障预报与优化操作研究.华东理工大学博士学位论文
[82]Nounou M. N., Bakshi B. R. On-line multiscale filtering of random and gross errors without process models. AIChE Journal,1999,45(5):1041-1054
[83]李秀喜,钱宇,宋光均,等.化工过程监测与故障预报系统的设计与开发.计算机工程与应用,2003(31)：211-215
[84]吴海桥.现代大型客机故障预报专家系统的研究与开发.南京航空航天大学博士学位论文,2002
[85]陈敏泽,周东华.动态系统的故障预报技术.控制理论与应用,2003,20(6)：819-824
[2]黄启明,钱宇,林伟璐,等.化工过程故障诊断研究进展.化工自动化及仪表,2000,27(3)：1-5
[3]Evenson G, Befus S, Dolfi M, et al. LOPA as a global manufacturing API facility. Process Safety Progress,2009,28(4):312-316.
[4]Center for Chemical Process Safety, American Institute of Chemical Engineers. Layer of Protection Analysis:Simplified Process Risk Assessment. New York,2001.
[5]张其立,邱彤.LOPA被动型保护层有效性研究.2010中国过程系统工程年会会议论文集(PSE2010)363-367
[6]周华,李秀喜,钱宇.石油化工过程安全技术研究进展[J].化工进展.2008,27(10)
[7]罗云,樊运晓,马晓春.风险分析与安全评价[M].北京：化学工业出版社,2004
[8]Chang M K. Application of risk based inspection in refinery and processing piping[J]. Journal of Loss Prevention in the Process Industries,2005,18:397-402.
[9]Santosh. Reliability analysis of pipelines carrying H2S for risk based inspection of heavy water plants [J]. Reliability Engineering and System Safety,2006(91):163-170.
[10]吴重光.系统建模与仿真.清华大学出版社,2008,252-264
[11]L.N. de Castro. Immune, swarm, and evolutionary algorithms Part I:Basic models. Proceedings of the International Conference on Neural Information Processing, Workshop on Artificial Immune Systems.2002,3:1464-1468.
[12]American Petroleum Institute. Base resource documentation for risk based inspection (1st ed) [C].American petroleum Institute, Washington DC, USA:API Publication 581,2000.
[13]L. N. de Castro, F. J Von Zuben. The clonal selection algorithm with engineering applications. Proceedings of GECCO'00, Workshop Proceedings.2000:36-37
[14]Li X X, Qian Y, Wang J F. Process monitoring based on wavelet packet principal component analysis [J]. European Symposium on Computer Aided Process Engineering, 2004,13:455-460.
[15]王春利,张卫华,吴重光.故障诊断技术理论与发展趋势及其在石化领域的应用.石油化工自动化,2008,6：7-13
[16]Frank P M. New developments using AI in fault diagnosis[J]. Engineering Application of Artificial Intelligence,1997,10(1):3-14.
[17]Venkatasubramanian V, Rengaswamy R, Kavuri S N. A review of process fault detection and diagnosis (Part II):Qualitative models and search strategies [J]. Computers and Chemical Engineering,2003(27):313-326
[18]Venkatasubramanian V, Rengaswamy R, Kavuri S N, et al. A review of process fault detection and diagnosis (Part Ⅲ):Process history based methods [J]. Computers and Chemical Engineering,2003(27):327-346.
[19]郭明.基于数据驱动的流程工业性能监控与故障诊断研究.2004年浙江大学博士学位论文.
[20]张贝克,张海洋,马昕.定性趋势分析在过程故障诊断中的应用研究[J].系统仿真学报.2008,20(10)
[21]Kourti T, MacGreor J F. Multivariate SPC methods for process and product monitoring [J]. Journal of Quality Technology,1996,28(4):409-428.
[22]何学文.基于支持向量机的故障智能诊断理论与方法研究[D].长沙：中南大学博士学位论文,2004.
[23]DASH S, VENKATASUBRAMANIAN V. Challenges in the industrial applications of fault diagnosis system[J]. Computer and Chemical Engineering,2000,24(2-7):785-791.
[24]郑小霞.大型PTA生产过程智能故障预报与优化操作研究.华东理工大学博士学位论文.
[25]邓晓刚,田学民.生产过程故障诊断方法研究现状及展望[J].石油大学学报(自然科学版),2005,29(6)：151-157.
[26]丁艳军,王培红,吕震中,等.生产过程早期故障检测与诊断的一种新方法[J].中国电机工程学报,2000,20(3)：61-70.
[27]Zhijie Zhou, Changhua Hu, Hongdong Fan, Jin Li. Fault prediction of the nonlinear systems with uncertainty. Simulation Modelling Practice and Theory.2008,16:690-703.
[28]K. S. Lu, R. Saeks. Failure prediction for an on-line maintenance system in a poisson shock environment, IEEE Transaction on Systems. Man and Cybernetics. 90(1979)356-362.
[29]Andre B. de Carvalho, Aurora Pozo. A symbolic fault-prediction model based on multi-objective particle swarm optimization. System and Software.2010,83 (5): 868-882.
[30]左宪章,康健,李浩.故障预测技术综述.火力与指挥控制.2010,35(1)：1-5.
[31]Zhijie Zhou, Changhua Hu, Hongdong Fan. Fault prediction of the nonlinear systems with uncertainty. Simulation Modelling Practice and Theory.2008,16(6):690-703.
[32]Cagatay Catal, Banu Diri. A systematic review of software fault prediction studies. Expert Systems with Application 2009,36(4):7346-7354.
[33]Cagatay Catal, Banu Diri. Investigating the effect of dataset size, metrics sets, and feature selection techniques on software fault prediction problem.2009 179(8): 1040-1048.
[34]Kim, Hoyong. Intelligence control of vehicle dynamic system by artificial neural network. Transportation Research Part A:Policy and Practice 1997 31(1):63-66
[35]Lo Shih chung B. Artificial convolution neural network for medical image pattern recognition. Neural Networks,1995 8(7-8):1201-1214
[36]王旭,王宏,王文辉.人工神经元网络原理与应用.沈阳：东北大学出版.2000：25-35
[37]孙明.基于神经网络的抽油井故障诊断系统.中国石油大学学位论文.
[38]魏勇.基于具有模糊输出BP网络的决策分类模型的研究.北京工业大学硕士学位论文.2002
[39]Aoki T, Ceravolo R, Stefano A Deetal.Seismic vulnerability assessment of chemical plants through probabilistic neural networks. Reliability Engineering and System Safety.2002,77(9):263-268.
[40]Ren Puli, Zheng Ouwang. Mining classification rules using rough sets and neural networks. European Journal of Operational Research,2004(157):439-448
[41]Foster W., Collopy F., Ungar L. H. Neural networks forecasting of short noisy time series. Computers & Chemical Engineering,1992,16(4):293-298
[42]吴宇.GSM-R越区切换的网络优化专家系统研究与开发.北京交通大学硕士学位论文.
[43]刘静.同步齿轮箱初始故障在线检测与诊断专家系统的研究.武汉科技大学硕士学位论文.
[44]刘长龙,马听,张贝克.基于SDG模型的故障诊断专家系统.计算机工程.2009,35(19)：233-235
[45]刘胜,王宇超,傅荟璇.船用计程仪故障诊断模糊专家系统.计算机工程.2009,35(18)：227-230.
[46]张广斌,葛运建,方凯.神经网络和专家系统在变速器检测中的应用.华中科技大学学报：自然科学版.2008,36(增刊Ⅰ)：53-56.
[47]马笑潇,黄席樾,柴毅.基于支持向量机的故障过程趋势预测研究[J].系统仿真学报,2002,14(11)：1548—1551
[48]Lapp S A, Powers G J. Computer-aided Synthesis of Fault trees. IEEE Transactions on Reliability.1977,26(2):2-12
[49]M. Iri, K Aoki, E.0'shima, H Matsuyama. An Algorithm for Diagnosis of System Failures in the Chemical Process. Computer & Chemical Engineering,1979,3:489-493.
[50]Shiozaki J, Matsuyama H, Tano K, O'shima E. Fault Diagnosis of Chemical Processes by the use of Signed, Directed Graphs. Extension to Five-range Patterns of Abnormality. International Chemical Engineering,1985,25(4):651-659.
[51]Yu C C, Lee C. Fault Diagnosis Based on Qualitative/Quantitative Process Knowledge. AIChE Journal,1991,37(4):617-628.
[52]Tarifa E, Scenna N. Fault Diagnosis, Direct Graphs, and Fuzzy Logic. Computers and Chemical Engineering,1997, (21):649-654.
[53]Vedam H, Venkatasubramanian V. PCA-SDG Based Process Monitoring and Fault Diagnosis. Control Engineering Practice,1999, (7):903-917.
[54]黄道平,龚婷婷,曾辉.基于非线性主元分析和符号有向图的故障诊断方法.化工学报,2009,60(12)：3058-3062.
[55]Vedam H, Venkatasubramanian V. Signed Digraph Based Multiple Fault Diagnosis. Computers and Chemical Engineering,1997,21:655-660.
[56]Maurya M R, Rengaswamy R, Venkatasubramaniana V. Application of Signed Digraphs-based Analysis for Fault Diagnosis of Chemical Process Flowsheets. Engineering Application of Artificial Intelligence,2004,17:501-518.
[57]Maurya M R, Rengaswamy R, Venkatasubramaniana V. Incipient fault diagnosis of ennessee Eastman flowsheet using signed directed graph and trend analysis. Proceedings of the 12th European Symposium on Computer Aided Process Engineering (ESCAPE-12),2002.
[58]Maurya M R, Rengaswamy R, Venkatasubramaniana V. A signed directed graph-based systematic framework for chemical engineering science,61,2006:1790-1810
[59]王春利,张卫华,吴重光.故障诊断技术理论与发展趋势及其在石化领域的应用.石油化工自动化,2008,6：7-13.
[60]吴重光,夏涛,张贝克等.基于符号定向图(SDG)深层知识模型的定性仿真.系统仿真学报,2003,15(10)：1351-1355.
[61]杨帆,肖德云.概率SDG模型及故障分析推理.控制与决策,2006,21(5)：487-491.
[62]杨帆,肖德云.结构残差在基于SDG故障分离中的应用.控制工程,2007, 14(3)：320-324.
[63]吴重光.系统建模与仿真.北京：清华大学出版社,2008.
[64]杨帆,萧德云.SDG建模及其应用的进展.控制理论与应用.2005,22(5)：767-774.
[65]UMEDA T, KURUYAMA T. A graphical approach to cause and effect analysis of chemical processing systems[J]. Chemical Engineering Science,1980,35(12): 2379-2388.
[66]LAPP S A, POWERS G J. Computer-aided synthesis of fault-trees [J]. IEEE Trans on Reliability,1977,26(2):2-12.
[67]石宇,邱彤,陈丙珍.用于化工过程的SDG故障分析方法.化工进展.2006,25(12)：1484-1488.
[68李安峰,夏涛,张贝克等.化工过程建模SDG方法[J].系统仿真学报,2003,15(10)：1364-1368.
[69]Cheung J T, Stephanopoulos G.Representation of Process Trends-Part I:A Formal Representation Framework [J]. Computers and Chemical Engineering,1990,14(4-5): 495-510
[70]SHIOZKI J, SHIBATA B, MATSUYAMA H. Fault diagnosis of chemical processes utilizing signed directed graphs-improvement by using temporal information [J]. IEEE Transacations on Industrial Electronics,1989,36(4):469-474
[71]DASH S, RENGASWAMY R, VENKATASUBRAMANIAN V. A novel interval-halving framework for automated identification of process trends [J]. American Institute of Chemical Engineers,2004,50(1):149-162.
[72]J.T.Y Cheung, G Stephanopoulos. Representation of process trends-part I. A formal representation framework[J]. Computers and Chemical Engineering (S0098-1354), 1990,14(4):495-510.
[73]M. R. Maurya, R. Rengaswamy, V. Venkatasubramanian. A signed directed graph and qualitative trend.analysis-based framework for incipient fault diagnosis. Chemical Engineering Research and Design,2007,85(10):1407-1422
[74]Dong Gao, Chongguang Wu, Beike Zhang, Xin Ma. Signed directed graph and qualitative trend analysis based fault diagnosis in chemical industry. Chinese Journal of Chemical Engineering,2010,18(2):265-276
[75]李希,谢刚.PTA技术国产化中的主要化学工程问题及其研究思路.聚酯工业,2001,14(1)：1-7
[76]Lindahl H. A., Abrams K. J., Stark L. E., et al. Method and apparatus for controlling the manufacture of terephthalic acid to control the level and variability of the contaminant content and the optical density. US Patent.4835307
[77]周红军,郭曦虹.PTA市场分析与扬子石化的营销策略.当代石油石化,2005,13(6)：24-27
[78]王丽军,李希,谢刚,等.对二甲苯液相催化氧化动力学(Ⅰ)反应机理和动力学模型.化工学报,2003,54(7)：946-952
[79]罗云,樊运晓,马晓春.风险分析与安全评价[M].北京：化学工业出版社,2004
[80]吴海桥.现代大型客机故障预报专家系统的研究与开发.南京航空航天大学博士学位论文,2002
[81]郑小霞.大型PTA生产过程智能故障预报与优化操作研究.华东理工大学博士学位论文
[82]Nounou M. N., Bakshi B. R. On-line multiscale filtering of random and gross errors without process models. AIChE Journal,1999,45(5):1041-1054
[83]李秀喜,钱宇,宋光均,等.化工过程监测与故障预报系统的设计与开发.计算机工程与应用,2003(31)：211-215
[84]吴海桥.现代大型客机故障预报专家系统的研究与开发.南京航空航天大学博士学位论文,2002
[85]陈敏泽,周东华.动态系统的故障预报技术.控制理论与应用,2003,20(6)：819-824