基于Petri网的微电网故障诊断研究
|
摘要
随着国民经济的发展,电力用户对电能供应的可靠性、稳定性提出了更高的要求。电力系统安全、稳定、可靠的运行,是维持社会日常生产生活以及推动社会发展进步的重要基础。为了实现能源的可持续发展,充分利用新能源,国内外学者提出微电网的概念。微电网是利用经济环保的分布式发电技术向本地重要负荷供电,同时减小分布式电源对大电网的不利影响。
为了快速监测及消除微电网中的故障,确保微电网安全稳定运行,提高供电的可靠性以及快速实现微电网的故障定位,以便于检修和事故后的快速恢复,就需要一个优质的微电网故障诊断系统。目前,国内外专家学者主要基于专家系统、人工神经网络、模糊理论、遗传算法和Petri网等对传统大规模电网进行电力系统故障诊断,并不适用于运行方式灵活,拓扑结构多样化的微电网。随着Petri网理论的发展,基于Petri网的微电网故障诊断具有广阔的发展前景。
基于此,本文对Petri网表示知识,尤其是具有因果关系的Petri网表示知识进行了深入地研究,并对已有的基于Petri网的电力系统诊断模型进行分析。首先,本文对现有基于简单Petri网的故障诊断方法进行研究和分析。由于简单Petri网的模拟能力不强,所建立的Petri网模型对电网的结构依赖性很强,并不适用于运行方式灵活、拓扑结构多样化的微电网。其次,本文根据现在高级Petri网理论的发展,对现有基于着色Petri网理论的故障诊断模型进行分析,并在典型微电网的拓扑结构上进行推理分析,根据算例总结基于着色Petri网理论故障诊断的不足。
本文在上述工作的基础上,分析微电网的特点,介绍典型的微电网形式;并根据微电网保护以及故障诊断的要求,结合面向对象的思想和数据库理论,提出基于高级Petri网的微电网故障诊断模型。在基于高级Petri网的故障诊断模型中,通过加入时间标记来实现诊断逻辑的优先级。考虑到保护拒动以及信息传递丢失的问题,在诊断模型中引入模糊理论,增强诊断模型的鲁棒性。建立以保护动作装置的动作信息为依据的故障诊断模型,实现对微电网内部故障快速监测、故障定位。
With the development of national economy, higher requirements of the reliability and stability of power supply for power users has been put forward. The security and stability, reliability of power system is an important foundation of maintaining social production and daily life, which is also a major basis of promoting social development and progress. However, with the increase of the scale of power system, the problem of the large scale power system is becoming more and more serious. On the other hand, considered the shortage of fossil and the environment problem, coal and oil, natural gas for energy generation have seriously restricted the sustainable development of society. In order to achieve sustainable energy development and make full use of new energy, the concept of Distributed Generation and Micro-grid is proposed. Micro-grid provides power to local loads economically and environmentally, using distribution generation technology, which also reduces the adverse effects of the bulk power system.
Reliability and continuity of power supply is required more and more highly, but the fault in power system is inevitable. It is necessary to monitor and eliminate fault quickly not to only ensure security and stability but also to enhance power reliability and continuity in Micro-grid. A high-quality fault diagnosis system is needed to realize the fault location and restore power supply quickly in Micro-grid. At present, domestic and foreign scholars mainly research on“expert systems”,“artificial neural network”,“fuzzy theory”,“genetic algorithm”and“Petri net”for fault diagnosis in bulk power system. However, the methods used in bulk system do not apply to Micro-grid, the topology of which is flexible and diverse.
The paper has studied on the knowledge of Petri nets in depth and analyzed the fault diagnosis model based Petri nets in bulk system. Firstly, the paper analyzed the existed fault diagnosis model based on simple Petri nets. As the simulation of the simple Petri nets is not strong, the model based on Petri nets has a strong dependence on the structure of the grid. Therefore, it is fit for Micro-grid, which could work in flexible operation mode. Secondly, according to the development of high level Petri nets theory, the paper has analyzed the fault diagnosis model based on colored Petri nets. The reasoning process has been analyzed in typical Micro-grid topology. The shorts of the fault diagnosis based on colored Petri nets have been summarized according to the sample.
Furthermore, the paper has analyzed the characteristics of Micro-grid and described the typical form of Micro-grid. Considering the requirements of protection and fault diagnosis in Micro-grid, combined with object-oriented thinking and database theory, the paper puts forward fault diagnosis model of Micro-grid based on high-level Petri net. The model also adds time stamps to the elements to achieve the priority of diagnostic logic. Taking into account that the protection fail to operate and the loss may happen in information transmissions, fuzzy theory is introduced in fault diagnosis model to enhance the robustness of the fault diagnosis model. Building a model based on protection action information to achieve the rapid internal fault monitoring and location in Micro-grid.
为了快速监测及消除微电网中的故障,确保微电网安全稳定运行,提高供电的可靠性以及快速实现微电网的故障定位,以便于检修和事故后的快速恢复,就需要一个优质的微电网故障诊断系统。目前,国内外专家学者主要基于专家系统、人工神经网络、模糊理论、遗传算法和Petri网等对传统大规模电网进行电力系统故障诊断,并不适用于运行方式灵活,拓扑结构多样化的微电网。随着Petri网理论的发展,基于Petri网的微电网故障诊断具有广阔的发展前景。
基于此,本文对Petri网表示知识,尤其是具有因果关系的Petri网表示知识进行了深入地研究,并对已有的基于Petri网的电力系统诊断模型进行分析。首先,本文对现有基于简单Petri网的故障诊断方法进行研究和分析。由于简单Petri网的模拟能力不强,所建立的Petri网模型对电网的结构依赖性很强,并不适用于运行方式灵活、拓扑结构多样化的微电网。其次,本文根据现在高级Petri网理论的发展,对现有基于着色Petri网理论的故障诊断模型进行分析,并在典型微电网的拓扑结构上进行推理分析,根据算例总结基于着色Petri网理论故障诊断的不足。
本文在上述工作的基础上,分析微电网的特点,介绍典型的微电网形式;并根据微电网保护以及故障诊断的要求,结合面向对象的思想和数据库理论,提出基于高级Petri网的微电网故障诊断模型。在基于高级Petri网的故障诊断模型中,通过加入时间标记来实现诊断逻辑的优先级。考虑到保护拒动以及信息传递丢失的问题,在诊断模型中引入模糊理论,增强诊断模型的鲁棒性。建立以保护动作装置的动作信息为依据的故障诊断模型,实现对微电网内部故障快速监测、故障定位。
With the development of national economy, higher requirements of the reliability and stability of power supply for power users has been put forward. The security and stability, reliability of power system is an important foundation of maintaining social production and daily life, which is also a major basis of promoting social development and progress. However, with the increase of the scale of power system, the problem of the large scale power system is becoming more and more serious. On the other hand, considered the shortage of fossil and the environment problem, coal and oil, natural gas for energy generation have seriously restricted the sustainable development of society. In order to achieve sustainable energy development and make full use of new energy, the concept of Distributed Generation and Micro-grid is proposed. Micro-grid provides power to local loads economically and environmentally, using distribution generation technology, which also reduces the adverse effects of the bulk power system.
Reliability and continuity of power supply is required more and more highly, but the fault in power system is inevitable. It is necessary to monitor and eliminate fault quickly not to only ensure security and stability but also to enhance power reliability and continuity in Micro-grid. A high-quality fault diagnosis system is needed to realize the fault location and restore power supply quickly in Micro-grid. At present, domestic and foreign scholars mainly research on“expert systems”,“artificial neural network”,“fuzzy theory”,“genetic algorithm”and“Petri net”for fault diagnosis in bulk power system. However, the methods used in bulk system do not apply to Micro-grid, the topology of which is flexible and diverse.
The paper has studied on the knowledge of Petri nets in depth and analyzed the fault diagnosis model based Petri nets in bulk system. Firstly, the paper analyzed the existed fault diagnosis model based on simple Petri nets. As the simulation of the simple Petri nets is not strong, the model based on Petri nets has a strong dependence on the structure of the grid. Therefore, it is fit for Micro-grid, which could work in flexible operation mode. Secondly, according to the development of high level Petri nets theory, the paper has analyzed the fault diagnosis model based on colored Petri nets. The reasoning process has been analyzed in typical Micro-grid topology. The shorts of the fault diagnosis based on colored Petri nets have been summarized according to the sample.
Furthermore, the paper has analyzed the characteristics of Micro-grid and described the typical form of Micro-grid. Considering the requirements of protection and fault diagnosis in Micro-grid, combined with object-oriented thinking and database theory, the paper puts forward fault diagnosis model of Micro-grid based on high-level Petri net. The model also adds time stamps to the elements to achieve the priority of diagnostic logic. Taking into account that the protection fail to operate and the loss may happen in information transmissions, fuzzy theory is introduced in fault diagnosis model to enhance the robustness of the fault diagnosis model. Building a model based on protection action information to achieve the rapid internal fault monitoring and location in Micro-grid.
引文
[1]鲁宗相,王彩霞,闵勇,等.微电网研究综述.电力系统自动化, 2007, 31(19): 100~105.
[2]时珊珊,鲁宗相,周双喜,等.中国微电网的特点和发展方向.中国电力, 2009, 42(7): 21~25.
[3]何仰赞,温增银.电力系统分析[M].武汉:华中科技大学, 2002.
[4] European Photovoltaic Industry Association. Setting the Pace of The Solar Age. SET FOR 2020, June 25, 2009.
[5] Giraud F, Salameh Z M. Steady-state performance of a grid connected rooftop hybrid wind-photovoltaic power system with battery storage. Power Engineering Review, IEEE. 2001, 21(2): 54.
[6] Murray Thomson, David G. Network power-flow analysis for a high penetration of distributed generation. IEEE Transactions on Power system, 2007, 22(3): 1157-1162.
[7]中华人民共和国国务院,国家中长期科学和技术发展规划纲要(2006-2020年),[online].Available:http://www.gov.cn/jrzg/2006/02/09/content_183787.htm
[8] Robert Lasseter, Abbas Akhil, Chris Marnay, The CERTS MicroGrid Concept, Consultant Report, 2002.4.
[9] Carl Adam Petri. Communication with automata, Darmstadt University of Technology, 1962.
[10] Kurt Jensen. Coloured Petri nets[M]. Heidelberg: Springer Berlin, 1987.
[11] Hura GS, Atwood JW. The use of Petri nets to analyse coherent fault trees[J]. IEEE Trans on Reliability, 1998, 10(5): 469-474.
[12] Stephen JH. Fuzzy rule base systems verification using high-level Petri nets[J]. IEEE Trans on Knowledge and Data Engineering, 2003, 15(2): 457-473.
[13] Lo KL, Ng HS, Trecat J. Power systems fault diagnosis using Petri nets[J]. IEE Proceedings-Generations, Transmissions and Distributions, 1997, 144(3): 231-236.
[14] Camurri, A. Coglio, A.“Extended Simple Colored Petri Neits: A Tool For Plant Simulation”[J]. 1997 IEEE International Conference, 1997, 12(15): 2909-2914 vol. 3.
[15]史丽萍,时培磊,常伟,等.煤矿继电保护专家系统的开发.煤矿安全, 2011, 42(8): 95-98.
[16]殷建,舒洲.液压挖掘机故障诊断专家系统研究.机械工程师, 2011, 8: 57-58.
[17]张学军,刘小冰,阎彩萍,等.基于正反向推理的电力系统故障诊断.电力系统自动化, 1998, 22(5): 30~32.
[18]杜一,张沛超,郁惟镛,等.基于事例和规则混合推理的变电站故障诊断系统.电网技术, 2004, 28(1): 34~37.
[19]毕天姝,倪以信,吴复立,等.基于新型神经网络的电网故障诊断方法.中国电机工程学报, 2002, 22(2): 73~78.
[20]周明,任建文,李庚银,等.基于模糊推理的分布式电力系统故障诊断专家系统.电力系统自动化, 2001, 25(24): 33~36.
[21]束洪春,孙向飞,于继来,等.粗糙集理论在电力系统中的应用.电力系统自动化, 2006, 39(5):90~95.
[22]陈玉林,陈允平,孙金莉,等.电网故障诊断方法综述.中国电力, 2006.5, 39(5): 27~31.
[23]袁崇义. Petri网原理与应用.北京:电子工业出版社,2005.
[24]玄志成,陈章位,黄克强,等.基于Petri网的复合故障诊断方法的研究.机械科学技术, 1999, 18(2): 291-293.
[25]盛鹍,孔力,齐智平,等.新型电网-微电网(Microgrid)研究综述.继电器, 2007, 35(12):75~80.
[26]郭创新,朱传柏,曹一家,等.电力系统故障诊断的研究现状与发展趋势.电力系统自动化, 2006, 30(8): 98~103.
[27]李训领.中国能源发展评价研究: [硕士学位论文],北京:华北电力大学, 2008.6.
[28]林闯.随机Petri网和系统性能评价.北京:清华大学出版社, 2005.
[29]方培培,李永丽,杨晓军,等. Petri网与专家系统结合的输电网络故障诊断方法.电力系统及其自动化学报, 2005, 17(2): 26~30.
[30]孙静,秦世引,宋永华,等.一种基于Petri网和概率信息的电力系统故障诊断方法.电力系统自动化, 2003, 27(13): 10~13.
[31]段振国,高曙.一种电网故障智能诊断求解模型的研究.中国电机工程学报, 1997, 17(6): 399~402.
[32]毕天姝,杨春发,黄少锋,等.基于改进Petri网模型的电网故障诊断方法.电网技术, 2005, 29(21): 52~56.
[33]杨兵,丁辉,罗为民,等.基于知识库的变压器故障诊断专家系统.中国电机工程学报, 2002, 22(10): 121~124.
[34]何耀华,韩守木,程尚模,等.基于多神经网络协同推理的故障诊断系统的研制.中国电机工程学报, 1999, 19(12): 57~60.
[35]毕天姝,严正,文福栓,等.基于径向基函数神经网络的在线分布式故障诊断系统.电网技术, 2001, 25(11): 27~32.
[36]曾庆锋,何正友,杨健维,等.基于有色Petri网的电力系统故障诊断模型研究.电力系统保护与控制, 2010, 38(14): 5~11.
[37]任惠,米增强,赵洪山,等.基于编码Petri网的电力系统故障诊断模型研究.中国电机工程学报, 2005, 20(25): 44~49.
[38]刘思华.电网故障诊断方法的研究: [博士学位论文],山东济南:山东大学, 2010.11.
[39]杨健维,何正友.基于时序模糊Petri网的电力系统故障诊断.电力系统自动化, 2011, 35(15): 46~51.
[40]翁汉琍,毛鹏,林湘宁,等.一种改进的电网故障诊断优化模型.电力系统自动化, 2007, 31(7): 66~70.
[41]吴欣,郭创新,曹一家,等.基于贝叶斯网络及信息时序属性的电力系统故障诊断方法.中国电机工程学报, 2005, 25(5): 14~18.
[42]梅念,石东源,李银红,等.计及信息畸变影响的电网故障诊断分级优化方法.电工技术学报, 2009, 24(9): 178~185.
[43]汤磊,孙宏斌,张伯明,等.基于信息理论的电力系统在线故障诊断.电机工程学报, 2003, 23(7): 5~11.
[44]李旭光.微电网的建模仿真及运行特性分析:硕士学位论文,天津:天津大学, 2009.6.
[45]潘炎,汤庸.模糊时态工作流网的建模和时间可能性分析.计算机集成制造系统, 2006, 12(11): 1779~1784.
[46]宋巍,窦万春,刘茜萍,时间约束的Petri网及其可调度性分析与验证.软件学报, 2007, 18(1): 11~21.
[47]江志斌. Petri网及其在制造系统建模与控制中的应用.北京:机械工业出版社, 2004.
[48]陶晓峰,蒋昌骏,段江涛,等.一种面向对象Petri网的构造及应用研究.机械科学与技术, 2003, 22(5): 746.
[49]曹政才,乔非,吴启迪,等. Petri网技术在半导体生产线建模中的应用.同济大学学报:自然科学版, 2008, 36(12): 1707.
[50]文登敏.面向对象理论与实践.成都:西南交通大学出版社, 2005.
[51]杨行健.面向对象技术与面向对象数据库.西安:西北工业大学出版社, 1996.
[52]胡成志,卢继平,胡利华,等.分布式电源对配电网继电保护影响的分析.重庆大学学报, 2006, 29(8): 36~39.
[53] Natthaphob N., Fault current contribution from synchronous machine and inverter based distributed generators. Power Delivery, IEEE Transactions on, 2007, 22(1): 634~641.
[54] Jayawarna, N., Jenkins, N., Barnes, M., Safety analysis of micro-gird. Future Power Systems, 2005 International Conference on, Amsterdam, 2005, 7~14.
[55] Zhou M C, McDermott K, Patel P A, Petri net synthesis and analysis of a flexible manufacturing system cell, study of object-oriented petri net and its application in flexible manufacturing cells. IEEE Transactions on Systems, Man and Cybernetics, 1993,23(2): 523.
[2]时珊珊,鲁宗相,周双喜,等.中国微电网的特点和发展方向.中国电力, 2009, 42(7): 21~25.
[3]何仰赞,温增银.电力系统分析[M].武汉:华中科技大学, 2002.
[4] European Photovoltaic Industry Association. Setting the Pace of The Solar Age. SET FOR 2020, June 25, 2009.
[5] Giraud F, Salameh Z M. Steady-state performance of a grid connected rooftop hybrid wind-photovoltaic power system with battery storage. Power Engineering Review, IEEE. 2001, 21(2): 54.
[6] Murray Thomson, David G. Network power-flow analysis for a high penetration of distributed generation. IEEE Transactions on Power system, 2007, 22(3): 1157-1162.
[7]中华人民共和国国务院,国家中长期科学和技术发展规划纲要(2006-2020年),[online].Available:http://www.gov.cn/jrzg/2006/02/09/content_183787.htm
[8] Robert Lasseter, Abbas Akhil, Chris Marnay, The CERTS MicroGrid Concept, Consultant Report, 2002.4.
[9] Carl Adam Petri. Communication with automata, Darmstadt University of Technology, 1962.
[10] Kurt Jensen. Coloured Petri nets[M]. Heidelberg: Springer Berlin, 1987.
[11] Hura GS, Atwood JW. The use of Petri nets to analyse coherent fault trees[J]. IEEE Trans on Reliability, 1998, 10(5): 469-474.
[12] Stephen JH. Fuzzy rule base systems verification using high-level Petri nets[J]. IEEE Trans on Knowledge and Data Engineering, 2003, 15(2): 457-473.
[13] Lo KL, Ng HS, Trecat J. Power systems fault diagnosis using Petri nets[J]. IEE Proceedings-Generations, Transmissions and Distributions, 1997, 144(3): 231-236.
[14] Camurri, A. Coglio, A.“Extended Simple Colored Petri Neits: A Tool For Plant Simulation”[J]. 1997 IEEE International Conference, 1997, 12(15): 2909-2914 vol. 3.
[15]史丽萍,时培磊,常伟,等.煤矿继电保护专家系统的开发.煤矿安全, 2011, 42(8): 95-98.
[16]殷建,舒洲.液压挖掘机故障诊断专家系统研究.机械工程师, 2011, 8: 57-58.
[17]张学军,刘小冰,阎彩萍,等.基于正反向推理的电力系统故障诊断.电力系统自动化, 1998, 22(5): 30~32.
[18]杜一,张沛超,郁惟镛,等.基于事例和规则混合推理的变电站故障诊断系统.电网技术, 2004, 28(1): 34~37.
[19]毕天姝,倪以信,吴复立,等.基于新型神经网络的电网故障诊断方法.中国电机工程学报, 2002, 22(2): 73~78.
[20]周明,任建文,李庚银,等.基于模糊推理的分布式电力系统故障诊断专家系统.电力系统自动化, 2001, 25(24): 33~36.
[21]束洪春,孙向飞,于继来,等.粗糙集理论在电力系统中的应用.电力系统自动化, 2006, 39(5):90~95.
[22]陈玉林,陈允平,孙金莉,等.电网故障诊断方法综述.中国电力, 2006.5, 39(5): 27~31.
[23]袁崇义. Petri网原理与应用.北京:电子工业出版社,2005.
[24]玄志成,陈章位,黄克强,等.基于Petri网的复合故障诊断方法的研究.机械科学技术, 1999, 18(2): 291-293.
[25]盛鹍,孔力,齐智平,等.新型电网-微电网(Microgrid)研究综述.继电器, 2007, 35(12):75~80.
[26]郭创新,朱传柏,曹一家,等.电力系统故障诊断的研究现状与发展趋势.电力系统自动化, 2006, 30(8): 98~103.
[27]李训领.中国能源发展评价研究: [硕士学位论文],北京:华北电力大学, 2008.6.
[28]林闯.随机Petri网和系统性能评价.北京:清华大学出版社, 2005.
[29]方培培,李永丽,杨晓军,等. Petri网与专家系统结合的输电网络故障诊断方法.电力系统及其自动化学报, 2005, 17(2): 26~30.
[30]孙静,秦世引,宋永华,等.一种基于Petri网和概率信息的电力系统故障诊断方法.电力系统自动化, 2003, 27(13): 10~13.
[31]段振国,高曙.一种电网故障智能诊断求解模型的研究.中国电机工程学报, 1997, 17(6): 399~402.
[32]毕天姝,杨春发,黄少锋,等.基于改进Petri网模型的电网故障诊断方法.电网技术, 2005, 29(21): 52~56.
[33]杨兵,丁辉,罗为民,等.基于知识库的变压器故障诊断专家系统.中国电机工程学报, 2002, 22(10): 121~124.
[34]何耀华,韩守木,程尚模,等.基于多神经网络协同推理的故障诊断系统的研制.中国电机工程学报, 1999, 19(12): 57~60.
[35]毕天姝,严正,文福栓,等.基于径向基函数神经网络的在线分布式故障诊断系统.电网技术, 2001, 25(11): 27~32.
[36]曾庆锋,何正友,杨健维,等.基于有色Petri网的电力系统故障诊断模型研究.电力系统保护与控制, 2010, 38(14): 5~11.
[37]任惠,米增强,赵洪山,等.基于编码Petri网的电力系统故障诊断模型研究.中国电机工程学报, 2005, 20(25): 44~49.
[38]刘思华.电网故障诊断方法的研究: [博士学位论文],山东济南:山东大学, 2010.11.
[39]杨健维,何正友.基于时序模糊Petri网的电力系统故障诊断.电力系统自动化, 2011, 35(15): 46~51.
[40]翁汉琍,毛鹏,林湘宁,等.一种改进的电网故障诊断优化模型.电力系统自动化, 2007, 31(7): 66~70.
[41]吴欣,郭创新,曹一家,等.基于贝叶斯网络及信息时序属性的电力系统故障诊断方法.中国电机工程学报, 2005, 25(5): 14~18.
[42]梅念,石东源,李银红,等.计及信息畸变影响的电网故障诊断分级优化方法.电工技术学报, 2009, 24(9): 178~185.
[43]汤磊,孙宏斌,张伯明,等.基于信息理论的电力系统在线故障诊断.电机工程学报, 2003, 23(7): 5~11.
[44]李旭光.微电网的建模仿真及运行特性分析:硕士学位论文,天津:天津大学, 2009.6.
[45]潘炎,汤庸.模糊时态工作流网的建模和时间可能性分析.计算机集成制造系统, 2006, 12(11): 1779~1784.
[46]宋巍,窦万春,刘茜萍,时间约束的Petri网及其可调度性分析与验证.软件学报, 2007, 18(1): 11~21.
[47]江志斌. Petri网及其在制造系统建模与控制中的应用.北京:机械工业出版社, 2004.
[48]陶晓峰,蒋昌骏,段江涛,等.一种面向对象Petri网的构造及应用研究.机械科学与技术, 2003, 22(5): 746.
[49]曹政才,乔非,吴启迪,等. Petri网技术在半导体生产线建模中的应用.同济大学学报:自然科学版, 2008, 36(12): 1707.
[50]文登敏.面向对象理论与实践.成都:西南交通大学出版社, 2005.
[51]杨行健.面向对象技术与面向对象数据库.西安:西北工业大学出版社, 1996.
[52]胡成志,卢继平,胡利华,等.分布式电源对配电网继电保护影响的分析.重庆大学学报, 2006, 29(8): 36~39.
[53] Natthaphob N., Fault current contribution from synchronous machine and inverter based distributed generators. Power Delivery, IEEE Transactions on, 2007, 22(1): 634~641.
[54] Jayawarna, N., Jenkins, N., Barnes, M., Safety analysis of micro-gird. Future Power Systems, 2005 International Conference on, Amsterdam, 2005, 7~14.
[55] Zhou M C, McDermott K, Patel P A, Petri net synthesis and analysis of a flexible manufacturing system cell, study of object-oriented petri net and its application in flexible manufacturing cells. IEEE Transactions on Systems, Man and Cybernetics, 1993,23(2): 523.