Providing Fault Detection from Sensor Data in Complex Machines That Build the Smart City
<p>Annual cost reduction due to the incorporation of predictive maintenance techniques [<a href="#B1-sensors-22-00586" class="html-bibr">1</a>].</p> "> Figure 2
<p>Representation of the critical elements of induction motors and their main causes of failure [<a href="#B3-sensors-22-00586" class="html-bibr">3</a>].</p> "> Figure 3
<p>Block diagram of an example of a complex machine, green lines indicate energy flow, and blue lines indicate the data transmission to the cloud.</p> "> Figure 4
<p>Scheme of the proposed methodology for data analysis.</p> "> Figure 5
<p>Graphical representation of the Fourier transform, equation and number of operations [<a href="#B25-sensors-22-00586" class="html-bibr">25</a>].</p> "> Figure 6
<p>Representation of the Hanning window effect.</p> "> Figure 7
<p>Wavelet transform equations and representation of different wavelets.</p> "> Figure 8
<p>Coefficients of approximation and detail obtained by means of the WT of the signal displayed with a Symlets wavelet.</p> "> Figure 9
<p>Diagram of an example of a complex machine with layer differentiation (Yellow, sensor level; orange, board level; red, machine level).</p> "> Figure 10
<p>Image of the prototype used to test the methodology.</p> "> Figure 11
<p>Scheme of application of the methodology to the specific case.</p> "> Figure 12
<p>Disposition of a €50 banknote in the different orientations analyzed.</p> "> Figure 13
<p>Comparison of the Vaux tension measurements of a 5 € banknote on reverse back orientation in the case of normal operation (<b>left</b>) and a failure case (<b>right</b>).</p> "> Figure 14
<p>Comparison of the measurements of the transport motor current of a 50 € banknote on reverse front orientation in the case of normal operation (<b>left</b>) and that of a failure case (<b>right</b>).</p> "> Figure 15
<p>Comparison of the measurements of the double sensor 1 of a 20 € banknote on back orientation in the case of normal operation (<b>left</b>) and one case of failure (<b>right</b>).</p> "> Figure 16
<p>Diagram of the designed monitoring infrastructure.</p> "> Figure 17
<p>Graphs of the different phases of the identification process, from the original FFT, to the FFT without the base noise, with the detected peaks and the colored areas of interest.</p> "> Figure 18
<p>Comparison of the probability distributions obtained for the mean values of the transport motor current in the case of faults associated with the double sensors. Indicator: mean.</p> "> Figure 19
<p>Comparison of the probability distributions obtained for the mean values of the doubles sensor 1 measurements in the case of failures associated with the doubles sensors. Indicators: (<b>top left</b>) mean, (<b>top right</b>) standard deviation, (<b>bottom left</b>) skewness and (<b>bottom right</b>) kurtosis.</p> "> Figure 20
<p>Comparison of the probability distributions obtained for the mean values of the doubles sensor 2 measurements in the case of failures associated with the doubles sensors. Indicators: (<b>top left</b>) mean, (<b>top right</b>) standard deviation, (<b>bottom</b>) skewness.</p> "> Figure 21
<p>Comparison of the probability distributions obtained for the mean values of the voltage Vint in the case of failures associated with the double sensors. Indicator: mean.</p> "> Figure 22
<p>Summary of the implemented MLP.</p> "> Figure 23
<p>Confusion matrices of two neural networks designed to recognize the 14 failure cases.</p> ">
Abstract
:1. Introduction
- Detection of machines that suffer critical breakdowns for the production process.
- Location of the machine element that produces the faults.
- Identification of the causes that provoke the breakdowns (physical reasons why it breaks).
- Definition of the variables to be monitored.
- Selection of the sensors.
- Data acquisition.
- Data curation and extraction of indicators (features).
- Data processing so that the system learns to detect failures.
- -
- Has a 24/7 operation operated by users without detailed knowledge of the operation of all the constituent parts of the machine.
- -
- Integrates tens of sensors and actuators managed by several microcontrollers and microprocessors communicated by control buses.
- -
- Requires energy from the mains to work, sometimes has a battery, but as a short-time backup.
- -
- Has IP (Internet Protocols) connectivity.
2. Materials and Methods
2.1. Data Processing Methodology
2.1.1. Sensor Level—Variable Targeting
2.1.2. Board Level—Embedded Data Curation and Feature Extraction
2.1.3. Machine Level—Feature Integration and Pattern Finding
3. Results and Discussion
3.1. Testbench Definition
3.2. Data Analysis
3.2.1. Layer 1: Sensor Data
3.2.2. Layer 2: Board Data
3.2.3. Layer 3: Machine Data
4. Conclusions
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Why Predictive Maintenance Is Driving Industry 4.0. Available online: https://files.solidworks.com/partners/pdfs/why-predictive-maintenance-is-driving-industry-4.0405.pdf (accessed on 3 December 2021).
- Henriquez, P.; Alonso, J.B.; Ferrer, M.A.; Travieso, C.M. Review of Automatic Fault Diagnosis Systems Using Audio and Vibration Signals. IEEE Trans. Syst. Man Cybern. Syst. 2014, 44, 642–652. [Google Scholar] [CrossRef]
- Choudhary, A.; Goyal, D.; Shimi, S.L.; Akula, A. Condition Monitoring and Fault Diagnosis of Induction Motors: A Review. Arch. Comput. Methods Eng. 2019, 26, 1221–1238. [Google Scholar] [CrossRef]
- Castelli, M.; Andrade, M. Metodología de Monitoreo, Detección y Diagnóstico de Fallos En Motores Asíncronos de Inducción. Mem. Investig. Ing. 2007, 5, 65–76. [Google Scholar]
- Mardacany, E. Smart cities characteristics: Importance of buit environments components. In Proceedings of the IET Conference on Future Intelligent Cities, London, UK, 4–5 December 2014; pp. 1–6. [Google Scholar]
- Purnachand, K.; Shabbeer, M.; Rao, P.N.V.S.M.; Babu, C.M. Predictive maintenance of machines and industrial equipment. In Proceedings of the 2021 10th IEEE International Conference on Communication Systems and Network Technologies (CSNT), Bhopal, India, 18–19 June 2021; pp. 318–324. [Google Scholar]
- Ayvaz, S.; Alpay, K. Predictive Maintenance System for Production Lines in Manufacturing: A Machine Learning Approach Using IoT Data in Real-Time. Expert Syst. Appl. 2021, 173, 114598. [Google Scholar] [CrossRef]
- Mi, S.; Feng, Y.; Zheng, H.; Li, Z.; Gao, Y.; Tan, J. Integrated Intelligent Green Scheduling of Predictive Maintenance for Complex Equipment Based on Information Services. IEEE Access 2020, 8, 45797–45812. [Google Scholar] [CrossRef]
- Nangia, S.; Makkar, S.; Hassan, R. IoT Based Predictive Maintenance in Manufacturing Sector. SSRN Electron. J. 2020, 1–7. [Google Scholar] [CrossRef]
- Dalzochio, J.; Kunst, R.; Pignaton, E.; Binotto, A.; Sanyal, S.; Favilla, J.; Barbosa, J. Machine Learning and Reasoning for Predictive Maintenance in Industry 4.0: Current Status and Challenges. Comput. Ind. 2020, 123, 103298. [Google Scholar] [CrossRef]
- Liang, B.; Iwnicki, S.D.; Zhao, Y. Application of Power Spectrum, Cepstrum, Higher Order Spectrum and Neural Network Analyses for Induction Motor Fault Diagnosis. Mech. Syst. Signal Process. 2013, 39, 342–360. [Google Scholar] [CrossRef]
- Bendjama, H.; Bouhouche, S.; Boucherit, M.S. Application of Wavelet Transform for Fault Diagnosis in Rotating Machinery. Int. J. Mach. Learn. Comput. 2012, 2, 82–87. [Google Scholar] [CrossRef] [Green Version]
- Kiangala, K.S.; Wang, Z. An Effective Predictive Maintenance Framework for Conveyor Motors Using Dual Time-Series Imaging and Convolutional Neural Network in an Industry 4.0 Environment. IEEE Access 2020, 8, 121033–121049. [Google Scholar] [CrossRef]
- Strauss, P.; Schmitz, M.; Wostmann, R.; Deuse, J. Enabling of predictive maintenance in the brownfield through low-cost Sensors, an IIoT-architecture and machine learning. In Proceedings of the 2018 IEEE International Conference on Big Data (Big Data), Seattle, WA, USA, 10–13 December 2018; pp. 1474–1483. [Google Scholar]
- Saponara, S.; Fanucci, L.; Bernardo, F.; Falciani, A. A Network of Vibration Measuring Nodes with Integrated Signal Processing for Predictive Maintenance of High Power Transformers. In Proceedings of the 2015 IEEE 9th International Symposium on Intelligent Signal Processing (WISP), Sienna, Italy, 15–17 May 2015; pp. 1–4. [Google Scholar]
- Pereira, R.R.; Diniz da Silva, V.A.; Brito, J.N.; Daniel Nolasco, J. On-line monitoring induction motors by fuzzy logic: A study for predictive maintenance operators. In Proceedings of the 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), Changsha, China, 13–15 August 2016; pp. 1341–1346. [Google Scholar]
- Novoa, C.G.; Berríos, G.A.G.; Söderberg, R.A. Predictive maintenance for motors based on vibration analysis with compact rio. In Proceedings of the 2017 IEEE Central America and Panama Student Conference (CONESCAPAN), Panama City, Panama, 20–22 September 2017; pp. 1–6. [Google Scholar]
- Markiewicz, M.; Wielgosz, M.; Bochenski, M.; Tabaczynski, W.; Konieczny, T.; Kowalczyk, L. Predictive Maintenance of Induction Motors Using Ultra-Low Power Wireless Sensors and Compressed Recurrent Neural Networks. IEEE Access 2019, 7, 178891–178902. [Google Scholar] [CrossRef]
- Patil, S.S.; Gaikwad, J.A. Vibration analysis of electrical rotating machines using FFT: A method of predictive maintenance. In Proceedings of the 2013 Fourth International Conference on Computing, Communications and Networking Technologies (ICCCNT), Tiruchengode, India, 4–6 July 2013; pp. 1–6. [Google Scholar]
- Kavana, V.; Neethi, M. Fault analysis and predictive maintenance of induction motor using machine learning. In Proceedings of the 2018 International Conference on Electrical, Electronics, Communication, Computer, and Optimization Techniques (ICEECCOT), Msyuru, India, 14–15 December 2018; pp. 963–966. [Google Scholar]
- Rosli, N.S.B.; Ibrahim, R.B.; Ismail, I. Optimized neural network of predictive maintenance for Air Booster Compressor (ABC) motor failure. In Proceedings of the 2019 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), Auckland, New Zealand, 20–23 May 2019; pp. 1–6. [Google Scholar]
- Godoy, W.F.; Silva, I.N.; Goedtel, A.; Palácios, R.H.C.; Lopes, T.D. Application of Intelligent Tools to Detect and Classify Broken Rotor Bars in Three-phase Induction Motors Fed by an Inverter. IET Electr. Power Appl. 2016, 10, 430–439. [Google Scholar] [CrossRef]
- Amarnath, M.; Sugumaran, V.; Kumar, H. Exploiting Sound Signals for Fault Diagnosis of Bearings Using Decision Tree. Measurement 2013, 46, 1250–1256. [Google Scholar] [CrossRef]
- Patidar, S.; Soni, P.K. An Overview on Vibration Analysis Techniques for the Diagnosis of Rolling Element Bearing Faults. Int. J. Eng. Trends Technol. IJETT 2013, 4, 1804–1809. [Google Scholar]
- My NTi Audio. Available online: https://www.nti-audio.com/es/servicio/conocimientos/transformacion-rapida-de-fourier-fft (accessed on 1 December 2021).
- Vilela, R.M.; Metrglho, J.C.; Cardoso, J.C. Machine and industrial monitorization system by analysis of acoustic signatures. In Proceedings of the 12th IEEE Mediterranean Electrotechnical Conference (IEEE Cat. No. 04CH37521), Dubrovnik, Croatia, 12–15 May 2004. [Google Scholar]
- Zhang, W.; Jia, M.-P.; Zhu, L.; Yan, X.-A. Comprehensive Overview on Computational Intelligence Techniques for Machinery Condition Monitoring and Fault Diagnosis. Chin. J. Mech. Eng. 2017, 30, 782–795. [Google Scholar] [CrossRef] [Green Version]
- Kingma, D.P.; Ba, J. Adam: A Method for Stochastic Optimization. arXiv 2014, arXiv:1412.6980. [Google Scholar]
Variable | Board | Abreviation | Unit | Bits |
---|---|---|---|---|
Transport engine current | Engines | I_trans | mA | 12 |
Feeding engine current | Engines | I_feed | 12 | |
Transport engine encoder ticks | Engines | N_pul_trans | Number of counter ticks between two encoder pulses | 12 |
Feeding engine encoder ticks | Engines | N_pul_feed | 12 | |
Infrarred sensor 1a | Engines | IR1a | 0 no obstacle, 1 obstacle | 1 |
Infrarred sensor 1b | Engines | IR1b | 1 | |
Infrarred sensor 2 | Engines | IR2 | 1 | |
Infrarred sensor 3a | Engines | IR3a | 1 | |
Infrarred sensor 3b | Engines | IR3b | 1 | |
FFT of microphone measures | Engines | FFT | - | 1024 |
Doubles sensor 1 | Banknotes | Doubles1 | Measure proprtional to banknote’s thickness | 32 |
Doubles sensor 2 | Banknotes | Doubles2 | 32 | |
Temperature | Energy | Temp | Celsius degrees | 16 |
Internal voltage | Energy | Vint | V | 16 |
Auxiliary voltage | Energy | Vaux | 16 |
Identifier | Name of Failure | |
---|---|---|
0 | Normal operation case | |
1 | Effect of eccentricity in axle 2 | |
A | Concentrity deviation of 0.2 mm | |
B | Concentrity deviation of 0.5 mm | |
2 | Effect of eccentricity in axle 4 | |
A | Concentrity deviation of 0.2 mm. | |
B | Concentrity deviation of 0.5 mm. | |
3 | Effect of dented bearings: | |
A | Dented bearing in axle 2. | |
B | Dented bearing in axle 3. | |
4 | Effect of defective springs: | |
A | Spring without screw at BNF. | |
B | Spring without screw at the entrance of the safe. | |
5 | Effect of defective doubles sensors: | |
A | Perforated doubles wheel. | |
B | Eccentricity of 0.04 mm of the outer wheel. | |
C | Eccentricity of 0.08 mm of the outer wheel. | |
6 | Deteriorated pulleys and worn belts: | |
A | Deteriorated 32 z pulley. | |
B | Worn S2M 180 belt and deteriorated exit pulley. |
Divergence | KL > 4 | KL > 5 | KL > 10 | KL > 15 |
---|---|---|---|---|
Higher | LS | S | SS | SSS |
Lower | LI | I | II | III |
Failures | 1 | 2 | 3 | 4 | 5 | 6 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Indicators | A | B | A | B | A | B | A | B | A | B | C | A | B | |
Current | I_trans | SSS | = | II | II | I | = | III | III | III | II | III | III | III |
I_feed | = | = | = | = | = | = | = | = | = | = | = | = | = | |
Time between IR | T_IR11 | = | = | S | S | = | = | = | = | = | = | = | = | = |
T_IR31 | = | = | = | = | = | = | = | = | = | = | = | = | = | |
T_IR33 | = | = | = | = | = | = | = | = | = | = | = | = | = | |
N_pul_trans | = | = | = | = | = | = | = | = | = | = | = | = | = | |
N_pul_feed | = | = | = | = | = | = | = | = | = | = | = | = | = | |
Doubles sensors | Doubles 1 | III | III | S | = | S | SSS | SSS | S | AVG SSS DES S | AVG SSS DES SSS SK III KUR SSS | AVG SSS DES SSS SK III KUR SSS | SS | AVG SSS DES S |
Doubles 2 | AVG I DES II | AVG III DES II | S | LS | SS | SSS | SSS | S | AVG SSS | AVG III DES SSS | AVG III DES SSS SK I | SS | AVG SSS DES SSS SK SS | |
Voltages | Vint | = | = | = | = | = | = | = | = | = | = | LS | LS | = |
Vaux | = | = | = | = | = | S | = | = | = | = | = | = | = | |
FFTs | Energy 1 | = | = | = | = | = | = | = | I | = | = | = | LI | = |
Energy 2 | = | = | = | = | = | = | = | I | II | = | = | I | I |
I_Trans | Kullback–Leibler | |||||
---|---|---|---|---|---|---|
0-5A | 5A-0 | 0-5B | 5B-0 | 0-5C | 5C-0 | |
Mean | 25.322 | 26.229 | 6.836 | 14.215 | 12.087 | 17.932 |
Doubles 1 | Kullback–Leibler | |||||
---|---|---|---|---|---|---|
0-5A | 5A-0 | 0-5B | 5B-0 | 0-5C | 5C-0 | |
Mean | 26.985 | 27.402 | 19.838 | 21.173 | 25.869 | 25.379 |
Std.Dev. | 6.509 | 3.008 | 26.270 | 25.759 | 28.101 | 26.075 |
Skewness | 0.355 | 0.202 | 22.922 | 22.689 | 25.899 | 26.216 |
Kurtosis | 0.297 | 0.469 | 15.573 | 10.018 | 26.375 | 16.953 |
Doubles 2 | Kullback–Leibler | |||||
---|---|---|---|---|---|---|
0-5A | 5A-0 | 0-5B | 5B-0 | 0-5C | 5C-0 | |
Mean | 26.516 | 27.226 | 25.925 | 26.189 | 25.012 | 24.793 |
Std. Dev | 0.101 | 0.231 | 26.762 | 25.599 | 26.639 | 26.132 |
Skewness | 0.588 | 0.663 | 2.222 | 0.761 | 5.800 | 0.983 |
Vint | Kullback–Leibler | |||||
---|---|---|---|---|---|---|
0-5A | 5A-0 | 0-5B | 5B-0 | 0-5C | 5C-0 | |
Mean | 2.196 | 0.669 | 0.516 | 0.42 | 4.760 | 1.414 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Gascón, A.; Casas, R.; Buldain, D.; Marco, Á. Providing Fault Detection from Sensor Data in Complex Machines That Build the Smart City. Sensors 2022, 22, 586. https://doi.org/10.3390/s22020586
Gascón A, Casas R, Buldain D, Marco Á. Providing Fault Detection from Sensor Data in Complex Machines That Build the Smart City. Sensors. 2022; 22(2):586. https://doi.org/10.3390/s22020586
Chicago/Turabian StyleGascón, Alberto, Roberto Casas, David Buldain, and Álvaro Marco. 2022. "Providing Fault Detection from Sensor Data in Complex Machines That Build the Smart City" Sensors 22, no. 2: 586. https://doi.org/10.3390/s22020586