Automated Detection of Anomalous Holes in Fuel Injector Nozzles Using High-Definition Microphones (DH-FINM) and Machine Learning Algorithms

International Journal of Electronics and Communication Engineering

© 2024 by SSRG - IJECE Journal

Volume 11 Issue 9

Year of Publication : 2024

Authors : T. A. Mohanaprakash, D. Siva, J. Jegan, S. Janagiraman, M. Therasa

10.14445/23488549/IJECE-V11I9P114

How to Cite?

T. A. Mohanaprakash, D. Siva, J. Jegan, S. Janagiraman, M. Therasa, "Automated Detection of Anomalous Holes in Fuel Injector Nozzles Using High-Definition Microphones (DH-FINM) and Machine Learning Algorithms," SSRG International Journal of Electronics and Communication Engineering, vol. 11,  no. 9, pp. 149-162, 2024. Crossref, https://doi.org/10.14445/23488549/IJECE-V11I9P114

Abstract:

This study addresses a critical issue faced by fuel injector nozzle manufacturers: the identification of defects in holes caused by drill bit overheating. Manual inspection methods are often inadequate, leading to production delays and the creation of substandard products. To overcome this challenge, we propose a novel approach that combines high-resolution microphones with advanced machine learning algorithms. The system is designed to effectively isolate drilling sounds from background noise, enabling the use of a deep learning model to accurately differentiate between the sound signatures of a properly functioning drill bit and one that is damaged. This approach offers real-time monitoring and prompt alerts, ensuring that defective products are detected early in the production process, significantly reducing the likelihood of errors in the final output. By automating the detection process, the system not only enhances productivity but also improves overall product quality. This solution outperforms traditional methods by eliminating human error, minimizing downtime, and maintaining consistent production standards. The integration of high-resolution microphones allows for precise acoustic analysis, which is critical in identifying subtle differences in sound that may indicate defects. The machine learning model is trained on a comprehensive dataset of sound patterns, ensuring robust performance even in varied manufacturing environments. The proposed technology (DH-FINM –Detection of Holes in Fuel Injector Nozzles using Microphones) offers a cost-effective, scalable, and reliable solution that aligns with the industry’s need for efficient and accurate quality control mechanisms. This innovative approach provides a substantial improvement over conventional inspection techniques, contributing to higher efficiency, reduced waste, and improved product reliability. The adoption of this system could set a new standard for quality control in fuel injector nozzle manufacturing, paving the way for broader applications of similar technologies across other sectors where precision and quality are paramount.

Keywords:

Machine learning algorithms, Nozzle, Injector, Microphones, Noise.

References:

[1] Jong-Oh Sun, and Kwang-Joon Kim, “Isolation of Vibrations due to Speakers in Audio-Visual Electronic Devices without Deteriorating Vibration of Speaker Cone,” Journal of Mechanical Science and Technology, vol. 26, pp. 723-730, 2012.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Jingjing Zhang et al., “Isolated Word Recognition with Audio Derivation and CNN,” 2017 IEEE 29th International Conference on Tools with Artificial Intelligence, Boston, MA, USA, pp. 336-341, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Huy Phan et al., “Unifying Isolated and Overlapping Audio Event Detection with Multi-Label Multi-Task Convolutional Recurrent Neural Networks,” ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing, Brighton, UK, pp. 51-55, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Keunwoo Choi et al., “A Comparison of Audio Signal Preprocessing Methods for Deep Neural Networks on Music Tagging,” 2018 26th European Signal Processing Conference (EUSIPCO), Rome, Italy, pp. 1870-1874, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Jouni Pohjalainen et al., “Spectral and Cepstral Audio Noise Reduction Techniques in Speech Emotion Recognition,” Proceedings of the 24th ACM International Conference on Multimedia, Amsterdam, Netherlands, pp. 670-674, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Kin Wai Cheuk et al., “nnAudio: An on-the-Fly GPU Audio to Spectrogram Conversion Toolbox Using 1D Convolutional Neural Networks,” IEEE Access, vol. 8, pp. 161981-162003, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Monika Dörfler, Roswitha Bammer, and Thomas Grill, “Inside the Spectrogram: Convolutional Neural Networks in Audio Processing,” 2017 International Conference on Sampling Theory and Applications (SampTA), Tallinn, Estonia, pp. 152-155, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Muhammad Shadab Alam Hashmi et al., “Railway Track Inspection Using Deep Learning Based on Audio to Spectrogram Conversion: An On-the-Fly Approach,” Sensors, vol. 22, no. 5, pp. 1-21, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Zhiheng Ouyang et al., “A Fully Convolutional Neural Network for Complex Spectrogram Processing in Speech Enhancement,” ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing, Brighton, UK, pp. 5756-5760, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Guanxiong Shen et al., “Radio Frequency Fingerprint Identification for LoRa Using Spectrogram and CNN,” IEEE INFOCOM 2021 - IEEE Conference on Computer Communications, Vancouver, BC, Canada, pp. 1-10, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Chieh-Feng Cheng et al., “Activity Analysis of Construction Equipment Using Audio Signals and Support Vector Machines,” Automation in Construction, vol. 81, pp. 240-253, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Alessandro Maccagno et al., “A CNN Approach for Audio Classification in Construction Sites,” Progresses in Artificial Intelligence and Neural Systems, Smart Innovation, Systems and Technologies, vol. 184, pp. 371-381, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Chieh-Feng Cheng et al., “Evaluation of Software and Hardware Settings for Audio-Based Analysis of Construction Operations,” International Journal of Civil Engineering, vol. 17, pp. 1469-1480, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[14] R. Huber, and B. Kollmeier, “PEMO-Q—A New Method for Objective Audio Quality Assessment Using a Model of Auditory Perception,” IEEE Transactions on Audio, Speech, and Language Processing, vol. 14, no. 6, pp. 1902-1911, 2006.
[CrossRef] [Google Scholar] [Publisher Link]
[15] V.A. Abramov et al., “On-Board Transmission Quality Assessment Using Short Audio Signal,” 2021 Systems of Signals Generating and Processing in the Field of on Board Communications, Moscow, Russia, pp. 1-6, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Nallamilli P.G. Bhavani et al., “Real-Time Inspection in Detection Magnetic Flux Leakage by Deep Learning Integrated with Concentrating Non-Destructive Principle and Electromagnetic Induction,” IEEE Instrumentation & Measurement Magazine, vol. 25, no. 7, pp. 48-54, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[17] S. Sreethar et al., “Implementation of Cross Layer Design with Localization Techniques in Wireless Sensor Networks Using Deep Learning,” 2024 International Conference on Expert Clouds and Applications, Bengaluru, India, pp. 607-613, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[18] V. Elizabeth Jesi et al., “Advancements in Sensor Technology and Computer Vision for Enhanced Road Safety and Anomaly Detection using Machine Learning Approach,” 2023 Third International Conference on Ubiquitous Computing and Intelligent Information Systems, Gobichettipalayam, India, pp. 457-462, 2023.
[CrossRef] [Google Scholar] [Publisher Link]