为了提高末制导雷达故障诊断的效率和精度，提出一种基于属性粒化聚类与回声状态网络(ESN)的故障诊断方法.定义一种基于属性值影响度的属性区分能力测量指标，并以此作为相似性度量依据，利用近邻传播(AP)聚类算法得到区分能力相当的若干属性粒，通过选取聚类中心属性来完成故障征兆属性的约简；在储备池构建中，采用Bienenstock-Cooper-Munro (BCM)规则进行连接权矩阵的预训练，在目标函数中添加L1/2范数惩罚项，以提高ESN储备池对样本的动态适应性；同时，采用半阈值迭代法对模型进行求解，通过末制导雷达信号处理模块的故障诊断实例，验证了所提方法的有效性和优越性.结果表明：与BP神经网络和传统ESN模型相比，所提方法的稳定性及诊断准确率更高；将其与基于属性重要度和AP聚类的属性粒化约简算法相结合，能够进一步提高故障诊断的效率和精度，其仿真实验的训练时间仅为 8.98s，诊断正确率可达 95.2%.
In order to improve the efficiency and precision of terminal guidance radar fault diagnosis, a fault diagnosis method based on attribute granulation clustering and echo state network (ESN) was proposed. Firstly, an attribute distinguishing ability index was defined by attribute value influence degree. As the basis of similarity measure, a number of attribute granules of similar distinguish are obtained through affinity propagation clustering algorithm, and then fault attribute reduction was completed by selecting clustering center attributes. In order to improve the dynamic adaptability of ESN reservoir to samples, Bienenstock-Cooper-Munro (BCM) rule was introduced into the reservoir construction to train the connection weight matrix. Meanwhile, the L1/2-norm penalty term was added to the objective function in order to improve the sparsification efficiency, solving a numerical oscillation problem by using a smoothing L1/2 regularizer, the model was solved by using the half threshold iteration method at last. The effectiveness and superiority of the proposed method are verified by a fault diagnosis example of terminal guidance radar signal processing module. The training time of the simulation experiment is only 8.98s, and the diagnostic accuracy can reach 95.2%.
［1］潘红兵, 蔡云龙. 基于故障树及LabVIEW的雷达设备故障诊断［J］. 电子测量技术, 2013, 36(9): 115-118.
PAN Hongbing, CAI Yunlong. Fault diagnosis of radar equipment based on fault tree and LabVIEW［J］. Electronic Measurement Technology, 2013, 36(9): 115-118.
［2］SHAN X, YANG H, ZHANG P. Fault diagnosis expert system of artillery radar based on neural network［C］∥2010 International Conference on Computer Design and Applications (ICCDA). Qinhuangdao, China: IEEE, 2010, 2: V2-426-V2-429.
［3］涂望明, 宋执环, 陈运涛, 等. 基于小波变换和LS-SVM的雷达故障诊断［J］. 控制工程, 2013, 20(2): 309-312.
TU Wangming, SONG Zhihuan, CHEN Yuntao, et al. Radar fault diagnosis based on wavelet transform and LS-SVM［J］. Control Engineering, 2013, 20 (2): 309-312.
［4］KANG J, WU K, CHI K, et al. Multi-class intelligent fault diagnosis approach based on modified relevance vector machine［C］∥2016 International Conference on Intelligent Networking and Collaborative Systems (INCOS). Ostrava, Czech Republic: IEEE, 2016: 27-30.
［5］李伟, 梁玉英, 朱赛. 基于神经网络和证据理论的信息融合在故障诊断中的应用［J］. 计算机测量与控制, 2012, 20(11): 2888-2890.
LI Wei, LIANG Yuying, ZHU Sai. Application of information fusion based on neural networks and evidence theory in fault diagnosis［J］. Computer Measurement and Control, 2012, 20 (11): 2888-2890.
［6］JAEGER H. The “echo state” approach to analysing and training recurrent neural networks［EB/OL］. ［2017-05-10］. https:∥citeseerx.ist.psu.edu/showci-ting?cid=418118.
［7］LUN S X, YAO X S, QI H Y, et al. A novel model of leaky integrator echo state network for time-series prediction［J］. Neurocomputing, 2015, 159: 58-66.
［8］郭嘉, 雷苗, 彭喜元.基于相应簇回声状态网络静态分类方法［J］. 电子学报, 2011, 39(3A): 14-18.
GUO Jia, LEI Miao, PENG Xiyuan. Static classification method based on corresponding cluster echo state network［J］. Acta Electronica Sinica, 2011, (3A): 14-18.
［9］SCARDAPANE S, UNCINI A. Semi-supervised echo state networks for audio classification［J］. Cognitive Computation, 2017, 9(1): 125-135.
［10］MARTIN C E, REGGIA J A. Fusing swarm intelligence and self-assembly for optimizing echo state networks［J］. Computational Intelligence and Neuroscience, 2015(9): 1-15. doi: 10.1155/2015/642429.
［11］KUMP P, BAI E W, CHAN K, et al. Variable selection via RIVAL (removing irrelevant variables amidst Lasso iterations) and its application to nuclear material detection［J］. Automatica, 2012, 48(9): 2107-2115.
［12］XU H, ZHANG R, LIN C, et al. Novel approach of fault diagnosis in wireless sensor networks node based on rough set and neural network model［J］. International Journal of Future Generation Communication and Networking, 2016, 9(4): 1-16.
［13］YAO Y Y. Granular computing: Past, present and future［C］∥International Conference on Rough Sets and Knowledge Technology. Chengdu, China: IEEE, 2008.
［14］FREY B J, DUECK D. Clustering by passing messages between data points［J］. Science, 2007, 315: 972-976.
［15］CASTELLANI G C, INTRATOR N, SHOUVAL H, et al. Solutions of the BCM learning rule in a network of lateral interacting nonlinear neurons［J］. Network: Computation in Neural Systems, 1999, 10(2): 111-121.
［16］XU Z, ZHANG H, WANG Y, et al. L1/2 regularization［J］. Science China Information Sciences, 2010, 53(6): 1159-1169.
［17］DAUBECHIES I, DEVORE R, FORNASIER M, et al. Iteratively reweighted least squares minimization for sparse recovery［J］. Communications on Pure and Applied Mathematics, 2010, 63(1): 1-38.
