Sensor fault self-detection based on the mean shift method

Abstract

To accurately identify sensor faults caused by complex environmental conditions and ensure that structural health monitoring systems correctly perceive the structural state, a self-detection method for sensor nodes based on mean shift and sliding window techniques was proposed. The self-detection method comprises two stages, i.e., fault prescreening and fault self-detection. During the fault prescreening stage, the method rapidly identifies potentially abnormal data using the quartile method combined with the sliding window technique, significantly improving the efficiency of the method. During the fault self-detection stage, the method employs the mean shift algorithm to perform adaptive clustering of the abnormal data, effectively detecting various faults. Data from the Canton Tower were used to test the effectiveness of the method by setting four types of sensor faults, i.e., offset, drift, gain, and stuck. Then, the proposed method was compared with extremely randomized trees, random forests, support vector data description, and one-class support vector machines. Results show that the proposed method can detect the four aforementioned faults with high accuracy and computational efficiency.

Keywords

sensor fault detection / mean shift / sliding window / machine learning

References

[1] Ghoshal A, Sundaresan M J, Schulz M J, et al.Structural health monitoring techniques for wind turbine blades[J].Journal of Wind Engineering and Industrial Aerodynamics, 2000, 85(3): 309-324. DOI: 10.1016/S0167-6105(99)00132-4.
[2] Ko J M, Ni Y Q.Technology developments in structural health monitoring of large-scale bridges[J].Engineering Structures, 2005, 27(12): 1715-1725. DOI: 10.1016/j.engstruct.2005.02.021.
[3] Spencer B F Jr, Ruiz-Sandoval M E, Kurata N. Smart sensing technology: Opportunities and challenges[J]. Structural Control and Health Monitoring, 2004, 11(4): 349-368. DOI: 10.1002/stc.48.
[4] Wan H P, Ni Y Q.Bayesian multi-task learning methodology for reconstruction of structural health monitoring data[J].Structural Health Monitoring, 2019, 18(4): 1282-1309. DOI: 10.1177/1475921718794953.
[5] Hu J, Guo J, Ou J P, et al.Measurement of wind field characteristics at a long-span suspension bridge[J]. Journal of Southeast University(English Edition), 2011, 27(3): 328-334. DOI: 10.3969/j.issn.1003-7985.2011.03.020.
[6] Liu X L, Huang Q, Ren Y, Fan Y H, et al.Extraction of cable forces due to dead load in cable-stayed bridges under random vehicle loads[J]. Journal of Southeast University(English Edition), 2015, 31(3): 407-411. DOI: 10.3969/j.issn.1003-7985.2015.03.019.
[7] Wang H, Li A Q, Guo T, et al.Comparable study on typhoon and strong northern wind characteristics of the Runyang Suspension Bridge based on field tests[J]. Journal of Southeast University(English Edition), 2009, 25(1): 99-103. DOI: 10.3969/j.issn.1003-7985.2009.01.021.
[8] Mao J X, Wang H, Spencer B F Jr. Toward data anomaly detection for automated structural health monitoring: Exploiting generative adversarial nets and autoencoders[J].Structural Health Monitoring, 2021, 20(4): 1609-1626. DOI: 10.1177/1475921720924601.
[9] Mahapatro A, Khilar P M.Fault diagnosis in wireless sensor networks: A survey[J]. IEEE Communications Surveys & Tutorials, 2013, 15(4): 2000-2026. DOI: 10.1109/SURV.2013.030713.00062.
[10] Zidi S, Moulahi T, Alaya B.Fault detection in wireless sensor networks through SVM classifier[J]. IEEE Sensors Journal, 2018, 18(1): 340-347. DOI: 10.1109/JSEN.2017.2771226.
[11] Magalhães F, Cunha Á, Caetano E.Vibration based structural health monitoring of an arch bridge: From automated OMA to damage detection[J]. Mechanical Systems and Signal Processing, 2012, 28: 212-228. DOI: 10.1016/j.ymssp.2011.06.011.
[12] Mao J X, Wang H, Feng D M, et al.Investigation of dynamic properties of long-span cable-stayed bridges based on one-year monitoring data under normal operating condition[J].Structural Control and Health Monitoring, 2018, 25(5): e2146. DOI: 10.1002/stc.2146.
[13] Muhammed T, Shaikh R A.An analysis of fault detection strategies in wireless sensor networks[J].Journal of Network and Computer Applications, 2017, 78: 267-287. DOI: 10.1016/j.jnca.2016.10.019.
[14] Lau B C P, Ma E W M, Chow T W S. Probabilistic fault detector for Wireless Sensor Network[J].Expert Systems with Applications, 2014, 41(8): 3703-3711. DOI: 10.1016/j.eswa.2013.11.034.
[15] Abid A, Kachouri A, Guiloufi A B F, et al. Centralized KNN anomaly detector for WSN[C]//2015 IEEE 12th International Multi-Conference on Systems, Signals & Devices(SSD15). Mahdia, Tunisia, 2015: 1-4. DOI: 10.1109/SSD.2015.7348091.
[16] Warriach E U, Tei K.Fault detection in wireless sensor networks: A machine learning approach[C]//2013 IEEE 16th International Conference on Computational Science and Engineering. Sydney, Australia, 2014: 758-765. DOI: 10.1109/CSE.2013.116.
[17] Panda M, Khilar P M.Energy efficient distributed fault identification algorithm in wireless sensor networks[J]. Journal of Computer Networks and Communications, 2014, 2014: 1-16. DOI: 10.1155/2014/323754.
[18] Chen J R, Kher S, Somani A.Distributed fault detection of wireless sensor networks[C]//Proceedings of the 2006 Workshop on Dependability Issues in Wireless Ad Hoc Networks and Sensor Networks. Los Angeles, CA, USA, 2006: 65-72. DOI: 10.1145/1160972.1160985.
[19] Obst O.Distributed fault detection in sensor networks using a recurrent neural network[J].Neural Processing Letters, 2014, 40(3): 261-273. DOI: 10.1007/s11063-013-9327-4.
[20] Younis O, Fahmy S, Santi P.An architecture for robust sensor network communications[J].International Journal of Distributed Sensor Networks, 2005, 1(3/4): 305-327. DOI: 10.1080/15501320500330786.
[21] Barborak M, Dahbura A, Malek M.The consensus problem in fault-tolerant computing[J]. ACM Computing Surveys, 1993, 25(2): 171-220. DOI: 10.1145/152610.152612.
[22] Ni K, Ramanathan N, Chehade M N H, et al. Sensor network data fault types[J].ACM Transactions on Sensor Networks, 2009, 5(3): 1-29. DOI: 10.1145/1525856.1525863.
[23] Fukunaga K, Hostetler L.The estimation of the gradient of a density function, with applications in pattern recognition[J].IEEE Transactions on Information Theory, 1975, 21(1): 32-40. DOI: 10.1109/TIT.1975.1055330.
[24] Xiao C X, Liu M.Efficient mean-shift clustering using Gaussian KD-tree[J].Computer Graphics Forum, 2010, 29(7): 2065-2073. DOI: 10.1111/j.1467-8659.2010.01793.x.
[25] Ni Y Q, Xia Y, Liao W Y, et al.Technology innovation in developing the structural health monitoring system for Guangzhou New TV Tower[J].Structural Control and Health Monitoring, 2009, 16(1): 73-98. DOI: 10.1002/stc.303.
[26] Ni Y Q, Xia Y, Lin W, et al.SHM benchmark for high-rise structures: A reduced-order finite element model and field measurement data[J].Smart Structures and Systems, 2012, 10(4/5): 411-426. DOI: 10.12989/sss.2012.10.4_5.411.
[27] Saeed U, Jan S U, Lee Y D, et al.Fault diagnosis based on extremely randomized trees in wireless sensor networks[J].Reliability Engineering & System Safety, 2021, 205: 107284. DOI: 10.1016/j.ress.2020.107284.
[28] Noshad Z, Javaid N, Saba T, et al.Fault detection in wireless sensor networks through the random forest classifier[J]. Sensors, 2019, 19(7): 1568. DOI: 10.3390/s19071568.

Funding

Wan Huaping, male, doctor, professor, [e-mail].