电源车传感器故障检测和数据重构方法

蒋栋年; 把余江; 李炜

doi:10.13700/j.bh.1001-5965.2021.0501

电源车传感器故障检测和数据重构方法

doi: 10.13700/j.bh.1001-5965.2021.0501

蒋栋年^{1, 2},
把余江¹,
李炜^{1, 3, ,}

1.
兰州理工大学电气工程与信息工程学院，兰州 730050
2.
国家电网甘肃省电力科学研究院，兰州 730050
3.
兰州理工大学甘肃省工业过程先进控制重点实验室，兰州 730050

基金项目: 国家自然科学基金(62263020)；甘肃省杰出青年科学基金(20JR10RA202)；兰州理工大学红柳优秀青年人才资助计划；兰州市科技计划(2022-2-69)

详细信息

通讯作者:
E-mail：liwei@lut.edu.cn

中图分类号: TP277
计量
- 文章访问数: 218
- HTML全文浏览量: 55
- PDF下载量: 22
- 被引次数: 0
出版历程
- 收稿日期: 2021-08-30
- 录用日期: 2021-11-20
- 网络出版日期: 2021-12-15
- 整期出版日期: 2023-07-31

Sensor fault detection and data reconstruction method of power supply vehicle

JIANG Dongnian^{1, 2},
BA Yujiang¹,
LI Wei^{1, 3
, ,}

1.
College of Electrical and Information Engineering，Lanzhou University of Technology，Lanzhou 730050，China
2.
State Grid Gansu Electric Power Research Institute，Lanzhou 730050，China
3.
Key Laboratory of Gansu Advanced Control for Industrial Processes，Lanzhou University of Technology，Lanzhou 730050，China

Funds: National Natural Science Foundation of China (62263020); Outstanding Youth Fund of Gansu Province (20JR10RA202); Hongliu Outstanding Young Talents Support Project of Lanzhou University of Technology; Lanzhou Science and Technology Plan (2022-2-69)

More Information

Corresponding author: E-mail：liwei@lut.edu.cn

摘要

摘要:
针对电源车由于运行环境复杂而容易发生传感器故障的问题，提出了一种基于时空相关性的传感器故障检测和数据重构方法。针对单个传感器运行数据的时序关系特征，借助具有选择与遗忘机制的极限学习机(SF-ELM)建立了电源车传感器时间序列预测子模型，并据此实现对电源车传感器的故障检测；针对已检测的故障传感器，利用不同传感器之间的空间相关性，通过冗余度分析，使用改进后的互信息熵筛选出与故障传感器数据相关性较高的辅助传感器数据，实现对故障传感器失效数据的在线重构；通过仿真验证了所提方法在电源车传感器故障检测和数据重构中的可行性与有效性。
- 传感器 /
- 时空相关性 /
- 极限学习机 /
- 故障检测 /
- 数据重构
Abstract:
Aiming at the problem that the power supply vehicle is prone to sensor fault due to the complex operating environment, a sensor fault detection and data reconstruction method based on spatiotemporal correlation is proposed in this paper. Firstly, according to the time-series relationship characteristics of single sensor operation data, the fault detection sub-model of the power vehicle sensor is established with the help of the selective forgetting extreme learning machine (SF-ELM) mechanism, and the fault detection of power vehicle sensor is realized. Secondly, simultaneous interpreting the fault sensors, using the spatial correlation among different sensors, and through redundancy analysis, the improved mutual information entropy is used to screen out the auxiliary sensor data which is highly correlated with the fault sensor data, and the online reconstruction of the failure sensor data is realized. Finally, the feasibility and effectiveness of the proposed method in sensor fault detection and data reconstruction of power vehicles are verified by simulation.
- sensor /
- spatiotemporal correlation /
- extreme learning machine /
- fault detection /
- data reconstruction

HTML全文

图 1 电源车传感器故障检测和数据重构流程

Figure 1. Sensor fault detection and data recovery process of power supply vehicle

下载: 全尺寸图片幻灯片

图 2 信息熵V氏图

Figure 2. Information entropy V diagram

下载: 全尺寸图片幻灯片

图 3 电源车模块关系

Figure 3. Relationship of power vehicle module

下载: 全尺寸图片幻灯片

图 4 传感器正常工作时OS-ELM的预测结果及误差

Figure 4. Prediction results and errors of OS-ELM during normal sensor operation

下载: 全尺寸图片幻灯片

图 5 传感器正常工作时SF-ELM的预测结果及误差

Figure 5. Prediction results and errors of SF-ELM during normal sensor operation

下载: 全尺寸图片幻灯片

图 6 传感器发生恒偏差故障时的预测结果及残差

Figure 6. Prediction results and residuals in case of constant bias fault of sensor

下载: 全尺寸图片幻灯片

图 7 重构误差箱线图和频域曲线

Figure 7. Reconstruction error box line plots and frequency domain curves

下载: 全尺寸图片幻灯片

图 8 未进行辅助变量筛选时ELM重构结果及重构误差

Figure 8. ELM reconstruction results and reconstruction errors without auxiliary variable screening

下载: 全尺寸图片幻灯片

图 9 辅助变量筛选前后ELM的重构结果及重构误差

Figure 9. Reconstruction results and reconstruction errors of ELM before and after auxiliary variable screening

下载: 全尺寸图片幻灯片

表 1 时间序列预测模型测试结果

Table 1. Test results of time series prediction model

方法	初始训练样本	相对误差/%	时间/s
OS-ELM	200	10.15	0.95464
SF-ELM	200	1.01	0.85689

下载: 导出CSV

表 2 电源车传感器之间的互信息熵

Table 2. Mutual information entropy between power vehicle sensors

参数	有功功率	无功功率	功率因子	定子电流	定子电压	转子速度	励磁电压
有功功率		1.3201×10⁻⁴	3.6484×10⁻⁵	1.3473×10⁻⁴	1.5541×10⁻⁴	3.1343×10⁻⁶	3.3021×10⁻⁵
无功功率	1.3201×10⁻⁴		3.1738×10⁻⁴	5.1552×10⁻⁵	1.3239×10⁻⁴	2.5767×10⁻⁵	2.6637×10⁻⁵
功率因子	3.6484×10⁻⁵	3.1738×10⁻⁴		1.2910×10⁻⁴	3.9005×10⁻⁴	7.3856×10⁻⁵	8.5718×10⁻⁵
定子电流	1.3473×10⁻⁴	5.1552×10⁻⁵	1.2910×10⁻⁴		1.9872×10⁻⁴	3.0838×10⁻⁵	1.0739×10⁻⁴
定子电压	1.5541×10⁻⁴	1.3239×10⁻⁴	3.9005×10⁻⁴	1.9872×10⁻⁴		1.0635×10⁻⁴	1.1331×10⁻⁴
转子速度	3.1343×10⁻⁶	2.5767×10⁻⁵	7.3856×10⁻⁵	3.0838×10⁻⁵	1.0635×10⁻⁴		1.2210×10⁻⁶
励磁电压	3.3021×10⁻⁵	2.6637×10⁻⁵	8.5718×10⁻⁵	1.0739×10⁻⁴	1.1331×10⁻⁴	1.2210×10⁻⁶

下载: 导出CSV

表 3 不同辅助变量个数的重构结果对比

Table 3. Comparison of reconstruction results with different number of auxiliary variables

k	相对误差/%	时间/s
4	0.14	0.87169
6	0.17	1.02634

下载: 导出CSV

参考文献(20)

[1]	李炜, 周丙相, 蒋栋年. 基于深度学习序贯检验的电源车故障诊断方法[J]. 系统仿真学报, 2020, 32(4): 638-648. doi: 10.16182/j.issn1004731x.joss.18-0288 LI W, ZHOU B X, JIANG D N. Fault diagnosis method of vehicle power supply based on deep learning and sequential test[J]. Journal of System Simulation, 2020, 32(4): 638-648(in Chinese). doi: 10.16182/j.issn1004731x.joss.18-0288
[2]	李炜, 周丙相, 蒋栋年, 等. 基于多状态时间序列预测学习的电源车故障预测方法[J]. 吉林大学学报, 2020, 50(4): 1532-1544. LI W, ZHOU B X, JIANG D N, et al. Fault prediction of power supply vehicle based on multi-state time series prediction learning[J]. Journal of Jilin University, 2020, 50(4): 1532-1544(in Chinese).
[3]	MANDAL S, SANTHI B, SRIDHAR S, et al. Minor fault detection of thermocouple sensor in nuclear power plants using time series analysis[J]. Annals of Nuclear Energy, 2019, 134: 383-389.
[4]	MEJIA J, OCHOA-ZEZZATTI A, CRUZ-MEJÍA O, et al. Prediction of time series using wavelet Gaussian process for wireless sensor networks[J]. Wireless Networks, 2020, 26(12): 5751-5758.
[5]	WU J, HU K, CHENG Y W, et al. Data-driven remaining useful life prediction via multiple sensor signals and deep long short-term memory neural network[J]. ISA Transactions, 2020, 97: 241-250. doi: 10.1016/j.isatra.2019.07.004
[6]	NIU G, XIONG L, QIN X, et al. Fault detection isolation and diagnosis of multi-axle speed sensors for high-speed trains[J]. Mechanical Systems and Signal Processing, 2019, 131: 183-198. doi: 10.1016/j.ymssp.2019.05.053
[7]	JEONG S, FERGUSON M, HOU R, et al. Sensor data reconstruction using bidirectional recurrent neural network with application to bridge monitoring[J]. Advanced Engineering Informatics, 2019, 42(10): 1-14.
[8]	XIAO Y, YIN H, ZHANG Y, et al. A dual-stage attention-based Conv-LSTM network for spatio-temporal correlation and multivariate time series prediction[J]. International Journal of Intelligent Systems, 2021, 36(5): 2036-2057. doi: 10.1002/int.22370
[9]	LIN T H, WANG T C, WU S C. Deep learning schemes for event identification and signal reconstruction in nuclear power plants with sensor faults[J]. Annals of Nuclear Energy, 2021, 154: 108-113.
[10]	ELNOUR M, MESKIN N, AL-NAEMI M. Sensor data validation and fault diagnosis using auto-associative neural network for HVAC systems[J]. Journal of Building Engineering, 2020, 27: 100935. doi: 10.1016/j.jobe.2019.100935
[11]	NASROLAHI S S, ABDOLLAHI F. Sensor fault detection and recovery in satellite attitude control[J]. Acta Astronautica, 2018, 145(4): 275-283.
[12]	DU J, CHEN H, ZHANG W. A deep learning method for data recovery in sensor networks using effective spatio-temporal correlation data[J]. Sensor Review, 2018, 39(3): 1-11.
[13]	TAYEH G B, MAKHOUL A, PERERA C, et al. A spatial-temporal correlation approach for data reduction in cluster-based sensor networks[J]. IEEE Access, 2019, 7: 50669-50680. doi: 10.1109/ACCESS.2019.2910886
[14]	ZHAO Z, LIU Z, SUN Y, et al. WOS-ELM-based double redundancy fault diagnosis and reconstruction for aeroengine sensor[J]. Journal of Control Science and Engineering, 2017, 2017: 1-14.
[15]	张弦, 王宏力. 具有选择与遗忘机制的极端学习机在时间序列预测中的应用[J]. 物理学报, 2011, 60(8): 74-80. ZHANG X, WANG H L. Selective forgetting extreme learning machine and its application to time series prediction[J]. Acta Physica Sinica, 2011, 60(8): 74-80(in Chinese) .
[16]	SMEJKAL T, MIKYŠKA J. Efficient solution of linear systems arising in the linearization of the VTN-phase stability problem using the Sherman-Morrison iterations[J]. Fluid Phase Equilibria, 2021, 527: 1-12.
[17]	蒋栋年, 李炜. 基于自适应阈值的粒子滤波非线性系统故障诊断[J]. 北京航空航天大学学报, 2016, 42(10): 2099-2106. doi: 10.13700/j.bh.1001-5965.2015.0611 JIANG D N, LI W. Fault diagnosis of particle filter nonlinear systems based on adaptive threshold[J]. Beijing University of Aeronautics and Astronautics, 2016, 42(10): 2099-2106(in Chinese). doi: 10.13700/j.bh.1001-5965.2015.0611
[18]	SEKERKA R F. Entropy and information theory[M]//SEKERKA R F. Thermal physics: Thermodynamics and statistical mechanics for scientists and engineers. Berlin: Springer, 2015: 247-256.
[19]	OSBORNE J W. Improving your data transformations: Applying the Box-Cox transformation[J]. Practical Assessment Research & Evaluation, 2010, 15(1): 1-10.
[20]	SARMADI H, KARAMODIN A. A novel anomaly detection method based on adaptive Mahalanobis-squared distance and one-class kNN rule for structural health monitoring under environmental effects[J]. Mechanical Systems and Signal Processing, 2020, 140: 1-24.