基于神经网络的软件可靠性预测研究与应用
|
摘要
在计算机技术飞速发展的今天，人们对计算机系统的依赖性越来越高。而在造成计算机系统错误的因素中，软件占了绝大部分。随着软件体系规模的日益增大及其复杂性的日益增强，软件的可靠性问题更为突出。因此，如何保证软件的质量，有效地对软件产品特性进行度量和预测，对开发期间的现状进行控制，设计并开发可靠的软件己成为当务之急。而软件可靠性预测能够使开发和测试的相关人员对软件的可靠性有一个大致的了解，所以对于软件如何进一步开发、测试和质量的控制都具有十分重要的意义。
神经网络自开创以来一直深受许多学者的重视，并广泛运用于各种领域，取得了辉煌的成就。预测是神经网络的又一个重要应用领域，这是因为神经网络具有优良的非线性特性，特别适用于高度非线性系统的处理。所以基于神经网络的智能预测是解决非线性预测问题的有效方法，为预测理论开辟了新的广阔发展空间。
本文概述了国内外关于软件可靠性预测方法的研究现状，重点研究和分析了传统的软件可靠性模型和BP神经网络的结构和特点，指出了它们的不足之处。并结合前人运用神经网络进行软件可靠性预测的相关理论和成果，针对它们的不足之处进行了改进。特别是BP神经网络具有的收敛速度慢和易陷入局部极小值等问题，主要从算法和网络结构两方面着手进行了改进。在启发式改进算法方面通过增加动量项和可变的学习速率；数值优化方面主要使用Levenberg-Marquardt算法训练网络。而网络结构主要针对初始权值的选取和隐含层节点数的确定进行了优化设计；同时结合区间探测法、逐步搜索法、数据归一化和交叉验证等数学工具，提出了基于神经网络的软件可靠性预测模型。
最后，通过运用MATLAB仿真工具对各种模型进行了数值仿真分析，证实了新模型同传统模型相比预测精度更高，泛化能力更强和良好的一致性。为了进一步验证这一模型的实用性，特将其应用到某手机软件系统以估测此软件的可靠性。
With the rapid development of computer technology today, we depend more andmore on computer systems. Software is the primary factor among all that result incomputer system errors. The increasingly extended scale and complexity of softwaresystem make its reliability issue much more important. So how to secure the qualityof software as well as design and develop reliable software is our urgent task. As theprediction of software reliability enables the developers and testers to get generalideas about the software reliability before testing and using it, it is important forfurther development, testing and quality control of software.
Neural Network (NN) has been deeply appreciated by many scholars. Now it isused with great success in many fields. Prediction is one of the important applicationfields of NN. Most of the general predicting methods are based on linear analysis,when it comes to non-linear they met many difficulties. Therefore NN is competentfor non-linear proceeding for its excellent non-linear character. Predicting methodsbased on NN extend the space of predicting research.
This paper summarizes the status of software reliability prediction methods inhome and abroad. Through research and analysis of traditional software reliabilitymodels and neural network structure and characteristics we pointed out theirdeficiencies. The low speed of learning is a main problem for the BP neural networksin fault diagnosis systems. On the basis of others theory and results of softwarereliability prediction on neural network, the momentum term and variable studyingrate are increased to the standard BP algorithm, and LM algorithm is used to train theneural network because the algorithm can solve problem of velocity effectively, anddesigning Optimally of the network structure and numbers of the hidden layer nodes.We come up with the better software reliability prediction model based on ANN withthe help of many mathematic tools such as the Zone-Probing method, the Searchmethod gradually, data scaling, cross validation and so on.
At last, Simulation shows that compared with the classic models, the new modelhas better prediction precision, better generalization ability and lower dependence on the number of samples.for the further verification of this model we apply it to somecellular phone software to predict the software reliability of it.
神经网络自开创以来一直深受许多学者的重视，并广泛运用于各种领域，取得了辉煌的成就。预测是神经网络的又一个重要应用领域，这是因为神经网络具有优良的非线性特性，特别适用于高度非线性系统的处理。所以基于神经网络的智能预测是解决非线性预测问题的有效方法，为预测理论开辟了新的广阔发展空间。
本文概述了国内外关于软件可靠性预测方法的研究现状，重点研究和分析了传统的软件可靠性模型和BP神经网络的结构和特点，指出了它们的不足之处。并结合前人运用神经网络进行软件可靠性预测的相关理论和成果，针对它们的不足之处进行了改进。特别是BP神经网络具有的收敛速度慢和易陷入局部极小值等问题，主要从算法和网络结构两方面着手进行了改进。在启发式改进算法方面通过增加动量项和可变的学习速率；数值优化方面主要使用Levenberg-Marquardt算法训练网络。而网络结构主要针对初始权值的选取和隐含层节点数的确定进行了优化设计；同时结合区间探测法、逐步搜索法、数据归一化和交叉验证等数学工具，提出了基于神经网络的软件可靠性预测模型。
最后，通过运用MATLAB仿真工具对各种模型进行了数值仿真分析，证实了新模型同传统模型相比预测精度更高，泛化能力更强和良好的一致性。为了进一步验证这一模型的实用性，特将其应用到某手机软件系统以估测此软件的可靠性。
With the rapid development of computer technology today, we depend more andmore on computer systems. Software is the primary factor among all that result incomputer system errors. The increasingly extended scale and complexity of softwaresystem make its reliability issue much more important. So how to secure the qualityof software as well as design and develop reliable software is our urgent task. As theprediction of software reliability enables the developers and testers to get generalideas about the software reliability before testing and using it, it is important forfurther development, testing and quality control of software.
Neural Network (NN) has been deeply appreciated by many scholars. Now it isused with great success in many fields. Prediction is one of the important applicationfields of NN. Most of the general predicting methods are based on linear analysis,when it comes to non-linear they met many difficulties. Therefore NN is competentfor non-linear proceeding for its excellent non-linear character. Predicting methodsbased on NN extend the space of predicting research.
This paper summarizes the status of software reliability prediction methods inhome and abroad. Through research and analysis of traditional software reliabilitymodels and neural network structure and characteristics we pointed out theirdeficiencies. The low speed of learning is a main problem for the BP neural networksin fault diagnosis systems. On the basis of others theory and results of softwarereliability prediction on neural network, the momentum term and variable studyingrate are increased to the standard BP algorithm, and LM algorithm is used to train theneural network because the algorithm can solve problem of velocity effectively, anddesigning Optimally of the network structure and numbers of the hidden layer nodes.We come up with the better software reliability prediction model based on ANN withthe help of many mathematic tools such as the Zone-Probing method, the Searchmethod gradually, data scaling, cross validation and so on.
At last, Simulation shows that compared with the classic models, the new modelhas better prediction precision, better generalization ability and lower dependence on the number of samples.for the further verification of this model we apply it to somecellular phone software to predict the software reliability of it.
引文
[1] 徐仁佐，谢曼，郑人杰．软什可靠性模型及应用[M]．北京：清华大学出版社，1994，1，第1版．
[2] 邱邦清．国外软件可靠性模型研究与发展综述[J]．质量与可靠性，1994，1：14～18
[3] Michael R. Lyu. Handbook of Software Reliability Engineering[M].New York: McGraw-Hill, IEEE Computer Society Press, 1996.4
[4] John D. Musa, Anthony Iannino, Kazuhira Okumoto. Software Reliability: Measurement, Prediction, Application[M]. New York: McGraw-Hill, 1989,12
[5] 黄锡滋．软件可靠性、安全性与质量保证[M]．北京：电子工业出版社，2002，10，第1版
[6] 陈光宇，黄锡滋．软件可靠性学科发展现状及展望[J]．电子科技大学学报(社会科学版)，2002(3)：99～102
[7] 刘云，赵玮．软件可靠性研究与进展[J]．微机发展，2003(2)：12～15
[8] M.Xie. A shock model for software failures[J]. Microel.Rel. 1987, 27:717～724
[9] 郑骏．软件可靠性建模的发展和存在的问题[J]．计算机科学，1995(6)：74～78
[10] Whittake J A, Poore J H, Markov. Analysis of Software Specification [J]. ACM Trans. Software Engineering and Methodology, 1993, 2(2): 93～103
[11] 韩峰岩，覃征，王昕．软件可靠性预测的小波分析方法[J]．空军工程大学学报(自然科学版)，2004，12(6)：72～75
[12] 眭俊华，范灵春，张海盛．软件可靠性预测技术研究[J]．计算机工程，2006，1(1)：67～70
[13] Inouem, Nagayosh I A. A Chaos Neural computer[J]. Physical Letter A. 1991, 158(8): 373～376
[14] 解光军，杨俊安，庄镇泉．基于量子双缝干涉实验的神经网络模型[J]．模式识别与人工智能，2003，16(1)：28～32
[15] 朱大奇，陈小平．计算机过程控制技术[M]．南京：南京大学出版社，2001，12
[16] 朱大奇．航空电子设备故障诊断新技术研究：[博士学位论文][D]．南京：南京航空航天大学，2002
[17] 黄锡滋．软件的可靠性与安全性『M1．北京：科学出版社，1993，10，第1版
[18] Z. Jelinski, P.B. Moranda. Software reliability research. Statistical Computer Performance Evaluation, New York: Academic Press, 1972:438-484
[19] 周瑞，徐仁佐，杨晓．基于测试覆盖的NHPP模型[J]．小型微型计算机系统．1999，8(8)：587～591
[20] 王文康．关于软件评估的可靠性模型假设[J]．西北民族学院学报(自然科学版)．1999，3(1)：45～53
[21] 陈守余，周梅春．人工神经网络模拟实现与应用[M]．武汉：中国地质大学出版社，2000，9：77～82
[22] 朱大奇，史慧．人工神经网络原理及应用[M]．北京：科学出版社，2006，2：33～39
[23] 高洪深，陶有德．BP神经网络模型的改进[J]．系统工程理论与实践．1996(1)：67～71．
[24] 高雪鹏，从爽．BP网络改进算法的性能对比研究[J]．控制与决策．2001，16(2)：167～171
[25] Yongjoon Cho, Rhee S. Quality estimation of resistance spot welding by using pattern recognition with neural networks[J]. IEEE Transactions on Instrumentation and Measurement, 2004, 53(4): 330～334
[26] A.D.Doulamis, N.D. Doulamis, S.D. Kollias. On-line retrainable neural networks: improving the performance of neural networks in image analysis problems[J]. IEEE Transactions Neural Networks, 2000,11:137～155
[27] 朱曼丽，孙志锋，孙志林．一种基于神经网络的数字识别方法[J]．计算机工程与应用．2002(12)：74～77
[28] Zh.Zhou, Y.Jiang. Medical diagnosis with C4.5 rule preceded by artificial neural network ensemble[J]. IEEE Transactions on Information Technology in Biomedicine, 2003, 7(1): 37-42
[29] S.M.Yamany, A.A.Farag. Adaptive object identification and recognition using neural networks and surface signatures[J]. Proceedings. IEEE Conferenceon Advanced Video and Signal Based Surveillance.2003,4:137～142
[30] S. H. Ling, F. H. F. Leung, H. K. Lam, etal. A novel genetic-algorithm-based neural network for short-term load forecasting[J]. IEEE Transactions on Industrial Electronics. 2003, 50(4): 793～799
[31] 刘哗，邹建龙，张赞等．神经网络在传感信号处理中的应用[J]．计算机自动测量与控制．2000，8(5)：63～66
[32] M.J. Aitkenhead, A.J.S.McDonald. A neural network face recognition system[J]. Engineering Applications of Artificial Intelligence. 2003, 50(4): 793～799
[33] 张雷．基于神经网络技术的结构可靠性分析与优化设计：[博士学位论文][D]．吉林大学，2004
[34] 蔡自兴，徐光佑．人工智能及其神经网络[M]．北京：清华大学出版社．2004，第3版
[35] N.Logothetis. The Role of Data Transformationin Taguchi Analysis[J]. Quality and Reliability Engineering International. 1988 (40): 49～61
[36] Hagan M T, Demuth H B, Beale M H. Neural Network Design[M]. Boston: PWS Publishing. 1996
[37] 贺昌政，李晓峰，俞海．BP人工神经网络模型的新改进及其应用[J]．数学的实践与认识2002，7(4)：554～561
[38] Irie, S. Miyake. Capacity of three-layered perceptrons[J], Proceedings of International Joint Conference on Neural Networks. 1988, 1: 641～648
[39] S. M. Carrol, W. Dickinson. Construction of neural nets using Random transform[J]. Proceedings of International Joint Conference on Neural Networks. 1989, 1:607～611
[40] G. Cybenko. Approximation by superpositions of asigmoidal function[J]. Mathematics of Control, Signals, and Systems. 1989, 2(4): 303～314
[41] K. Funabashi. On the approximate realization of continuous mappings by neural networks[J].Neural Networks. 1989, 2:183～193
[42] K. Fomik. Approximation capabilities of multilayer feed forward network[J]. Neural Networks. 1991, 4(2): 251～257
[43] R. Hecht-Nielsen. Theory of the backpropagation[J]. Proceedings of International Joint Conference on Neural Networks. 1989, 1: 539～605
[44] 易帆．神经网络预测研究：[硕士论文][D]．西南交通大学．2005
[45] 管野文友．软件可靠性[M]．．北京：机械工业出版社，1988：247
[46] 董长虹．Matlab神经网络与应用[M]．北京：国防工业出版社，2005
[2] 邱邦清．国外软件可靠性模型研究与发展综述[J]．质量与可靠性，1994，1：14～18
[3] Michael R. Lyu. Handbook of Software Reliability Engineering[M].New York: McGraw-Hill, IEEE Computer Society Press, 1996.4
[4] John D. Musa, Anthony Iannino, Kazuhira Okumoto. Software Reliability: Measurement, Prediction, Application[M]. New York: McGraw-Hill, 1989,12
[5] 黄锡滋．软件可靠性、安全性与质量保证[M]．北京：电子工业出版社，2002，10，第1版
[6] 陈光宇，黄锡滋．软件可靠性学科发展现状及展望[J]．电子科技大学学报(社会科学版)，2002(3)：99～102
[7] 刘云，赵玮．软件可靠性研究与进展[J]．微机发展，2003(2)：12～15
[8] M.Xie. A shock model for software failures[J]. Microel.Rel. 1987, 27:717～724
[9] 郑骏．软件可靠性建模的发展和存在的问题[J]．计算机科学，1995(6)：74～78
[10] Whittake J A, Poore J H, Markov. Analysis of Software Specification [J]. ACM Trans. Software Engineering and Methodology, 1993, 2(2): 93～103
[11] 韩峰岩，覃征，王昕．软件可靠性预测的小波分析方法[J]．空军工程大学学报(自然科学版)，2004，12(6)：72～75
[12] 眭俊华，范灵春，张海盛．软件可靠性预测技术研究[J]．计算机工程，2006，1(1)：67～70
[13] Inouem, Nagayosh I A. A Chaos Neural computer[J]. Physical Letter A. 1991, 158(8): 373～376
[14] 解光军，杨俊安，庄镇泉．基于量子双缝干涉实验的神经网络模型[J]．模式识别与人工智能，2003，16(1)：28～32
[15] 朱大奇，陈小平．计算机过程控制技术[M]．南京：南京大学出版社，2001，12
[16] 朱大奇．航空电子设备故障诊断新技术研究：[博士学位论文][D]．南京：南京航空航天大学，2002
[17] 黄锡滋．软件的可靠性与安全性『M1．北京：科学出版社，1993，10，第1版
[18] Z. Jelinski, P.B. Moranda. Software reliability research. Statistical Computer Performance Evaluation, New York: Academic Press, 1972:438-484
[19] 周瑞，徐仁佐，杨晓．基于测试覆盖的NHPP模型[J]．小型微型计算机系统．1999，8(8)：587～591
[20] 王文康．关于软件评估的可靠性模型假设[J]．西北民族学院学报(自然科学版)．1999，3(1)：45～53
[21] 陈守余，周梅春．人工神经网络模拟实现与应用[M]．武汉：中国地质大学出版社，2000，9：77～82
[22] 朱大奇，史慧．人工神经网络原理及应用[M]．北京：科学出版社，2006，2：33～39
[23] 高洪深，陶有德．BP神经网络模型的改进[J]．系统工程理论与实践．1996(1)：67～71．
[24] 高雪鹏，从爽．BP网络改进算法的性能对比研究[J]．控制与决策．2001，16(2)：167～171
[25] Yongjoon Cho, Rhee S. Quality estimation of resistance spot welding by using pattern recognition with neural networks[J]. IEEE Transactions on Instrumentation and Measurement, 2004, 53(4): 330～334
[26] A.D.Doulamis, N.D. Doulamis, S.D. Kollias. On-line retrainable neural networks: improving the performance of neural networks in image analysis problems[J]. IEEE Transactions Neural Networks, 2000,11:137～155
[27] 朱曼丽，孙志锋，孙志林．一种基于神经网络的数字识别方法[J]．计算机工程与应用．2002(12)：74～77
[28] Zh.Zhou, Y.Jiang. Medical diagnosis with C4.5 rule preceded by artificial neural network ensemble[J]. IEEE Transactions on Information Technology in Biomedicine, 2003, 7(1): 37-42
[29] S.M.Yamany, A.A.Farag. Adaptive object identification and recognition using neural networks and surface signatures[J]. Proceedings. IEEE Conferenceon Advanced Video and Signal Based Surveillance.2003,4:137～142
[30] S. H. Ling, F. H. F. Leung, H. K. Lam, etal. A novel genetic-algorithm-based neural network for short-term load forecasting[J]. IEEE Transactions on Industrial Electronics. 2003, 50(4): 793～799
[31] 刘哗，邹建龙，张赞等．神经网络在传感信号处理中的应用[J]．计算机自动测量与控制．2000，8(5)：63～66
[32] M.J. Aitkenhead, A.J.S.McDonald. A neural network face recognition system[J]. Engineering Applications of Artificial Intelligence. 2003, 50(4): 793～799
[33] 张雷．基于神经网络技术的结构可靠性分析与优化设计：[博士学位论文][D]．吉林大学，2004
[34] 蔡自兴，徐光佑．人工智能及其神经网络[M]．北京：清华大学出版社．2004，第3版
[35] N.Logothetis. The Role of Data Transformationin Taguchi Analysis[J]. Quality and Reliability Engineering International. 1988 (40): 49～61
[36] Hagan M T, Demuth H B, Beale M H. Neural Network Design[M]. Boston: PWS Publishing. 1996
[37] 贺昌政，李晓峰，俞海．BP人工神经网络模型的新改进及其应用[J]．数学的实践与认识2002，7(4)：554～561
[38] Irie, S. Miyake. Capacity of three-layered perceptrons[J], Proceedings of International Joint Conference on Neural Networks. 1988, 1: 641～648
[39] S. M. Carrol, W. Dickinson. Construction of neural nets using Random transform[J]. Proceedings of International Joint Conference on Neural Networks. 1989, 1:607～611
[40] G. Cybenko. Approximation by superpositions of asigmoidal function[J]. Mathematics of Control, Signals, and Systems. 1989, 2(4): 303～314
[41] K. Funabashi. On the approximate realization of continuous mappings by neural networks[J].Neural Networks. 1989, 2:183～193
[42] K. Fomik. Approximation capabilities of multilayer feed forward network[J]. Neural Networks. 1991, 4(2): 251～257
[43] R. Hecht-Nielsen. Theory of the backpropagation[J]. Proceedings of International Joint Conference on Neural Networks. 1989, 1: 539～605
[44] 易帆．神经网络预测研究：[硕士论文][D]．西南交通大学．2005
[45] 管野文友．软件可靠性[M]．．北京：机械工业出版社，1988：247
[46] 董长虹．Matlab神经网络与应用[M]．北京：国防工业出版社，2005