为进一步简化模型结构，提高模式识别性能，提出了一种基于量子计算的限制波尔兹曼机网络模型（Restricted Boltzmann Machine Based on Quantum Computation, QRBM）。在QRBM网络中，依据RBM的网络结构，以量子计算为基础。首先，对数据进行量子化编码。然后，执行量子操作，生成网络的权值矩阵以简化步骤、提高计算效率。之后，确定网络层数以提高准确率，缩短执行时间。最后，实现QRBM模型参数的更新，从而达到故障分类的目的。将本文方法用于齿轮箱模式识别中，提取齿轮箱的正常、齿面磨损、齿根裂纹和断齿等振动信号的数据作为原始特征，采用QRBM神经网络模型进行模式识别。实验结果表明，QRBM分类算法在分类准确率和执行时间上获得的效果比普通神经网络、支持向量机和RBM网络更好，验证了本文方法的有效性和可行性。
Abstract
In order to simplify the structure of model and enhance the performance of pattern recognition, the net model of restricted Boltzmann machine based on quantum computation (QRBM) is proposed. In QRBM network, based on the net structure of RBM and quantum computation, firstly, the data is coded with quantum states. Then, by quantum operation, weight matrix is created for simplifying computation step and enhancing computation efficiency. After that, the number of net layer is confirmed to improve accuracy and shorten execution time. Finally, the parameters in the model are updated. The method is applied in gear fault diagnosis. The original feature is comprised of data which are extracted form vibration signal of gear box with normal states, wearing, crack and broken. QRBM is used for diagnosis with feature set. The results indicated that, compared with neural network, SVM and RBM network, QRBM has better performance in classification accuracy and execution time, which has proved the efficient and feasibility.
关键词
量子计算 /
限制波尔兹曼机 /
神经网络 /
齿轮 /
模式识别
{{custom_keyword}} /
Key words
Quantum computation /
Restricted Boltzmann Machine (RBM) /
neural network /
gear /
pattern recognition
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 温广瑞，李杨，廖与禾，等. 基于精确信息重构的故障转子系统振动加速度信号积分方法[J]. 机械工程学报, 2013, 49 (8): 1-9. Wen Guangrui, Li Yang, Liao Yuhe, et al. Faulty rotor system vibration acceleration signal integration method based on precise information reconstruction [J]. Chinese Journal of Mechanical Engineering, 2013, 49 (8): 1-9.
[2] 李国宾，关德林，李廷举. 基于小波包变换和奇异值分解的柴油机振动信号特征提取研究[J]. 振动与冲击, 2011,30 (8): 149-152. Li Guobin, Guan Delin, Li Tingju. Feature extraction of diesel vibration signal based on wavelet packet transform and singularity value decomposition[J]. Journal of Vibration and Shock, 2011,30 (8): 149-152.
[3] 李永龙，邵忍平，薛腾. 基于小波神经网络的齿轮系统故障诊断[J]. 航空动力学报, 2010, 25(1): 234-240. Li Yonglong, Shao Renping, Xue Teng. Gear fault diagnosis based on wavelet and neural network [J]. Journal of Aerospace Power. 2010, 25(1): 234-240.
[4] 李兵，张培林，任国全，等. 运用EMD和GA-SVM的齿轮故障特征提取与选择[J]. 振动、测试与诊断, 2009,29 (4): 445-449. Li Bing, Zhang Peilin, Ren Guoquan, et al. Gear Fault Diagnosis Using Empirical Mode Decomposition, Genetic Algorithm and Support Vector Machine[J]. Journal of Vibration, Measurement &Diagnosis. 2009,29 (4): 445-449.
[5] Hinton G E and Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504–507.
[6] S.C. Turaga, J.F. Murray, V. Jain, F. Roth, and et al. Convolutional networks can learn to generate affinity graphs for image segmentation [J]. Neural Computation, 2010, 22(2):511–538.
[7] Mohamed A, Dahl G E, Hinton G E. Acoustic Modeling Using Deep Belief Networks [C]. IEEE Trans on Audio, Speech and Language Processing, 2012, 20 (1): 14-22
[8] 陈宇, 郑德权, 赵铁军. 基于Deep Belief Nets 的中文名实体关系抽取[J], 软件学报, 2012, 23(10): 2572-2585. Chen Yu, Zheng Dequan, Zhao Tiejun, Chinese relation extraction based on Deep Belief Nets [J]. Journal of Software, 2012, 23(10):2572-2585.
[9] Mohamed A, Yu D and Deng L. Investigation of full-sequence training of deep belief networks for speech recognition[C]. Proceeding of Interspeech, 2010： 2846-2849.
[10] 李士勇，李盼池. 量子计算与量子优化算法[M]. 哈尔滨工业大学,2009.
[11] 郭荣华，李斌，庄镇泉. 基于混合量子遗传算法的嵌入式系统软硬件协同综合示范[J]. 量子电子学报, 2008,25 (4): 443-451. Guo Ronghua, Li Bin, Zhuang Zhenquan. Hybird quantum probabilistic coding genetic algorithm for hardware-software co-synthesis of embedded system [J]. Chinese Journal of Quantum Electronics. 2008, 25 (4): 443-451.
[12] 李士勇，李盼池. 求解连续空间优化问题的量子粒子群算法[J]. 量子电子学报, 2007,24 (5): 569-574. Li Shiyong, Li Panchi. Quantum particle swarms algorithm for continuous space optimization [J]. Chinese Journal of Quantum Electronics. 2007,24 (5): 569-574.
[13] 周日贵. 量子神经网络模型研究[D]. 南京航空航天大学博士学位论文,2008.
[14] Nielsen M A, Chuang I L. Quantum computation and quantum information [M]. Cambridge University Press, 2010.
[15] 王凯，张永祥，李军. 基于支持向量机的齿轮故障诊断方法研究[J]. 振动与冲击. 2006, 25(6):97-99. Wang Kai, Zhang Yongxiang, Li Jun. Study on Diagnosis of gear fault based on support vector machine [J]. Journal of Vibration and Shock. 2006, 25(6):97-99.
[16] 肖成勇，石博强，王文莉，等. 基于小波包和进化支持向量机的齿轮早期诊断研究[J]. 振动与冲击. 2007, 26(7): 10-12.Xiao Chengyong, Shi Boqiang, Wang Wenli, et al. Gear incipient diagnosing based on wavelet packet and genetic- support vector machine [J]. Journal of Vibration and Shock. 2007, 26(7): 10-12.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1642 KB)
