Novel Leak Detector Based on DWT an Experimental Study
We always face water leakage problems in underground distribution water networks (DWNs). Existing leak detectors suffer from false alarms due to poor leak signal quality affected by external noise, often collected by acoustic or vibratory sensors. This paper introduces a novel Discrete Wavelet Tran...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
ascee
2024
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/12502/1/J17990_cf1915af6d02eca4e02b729c3fb64604.pdf |
| _version_ | 1825638661307236352 |
|---|---|
| author | Meftah, Sabir Bentoumi, Miloud Burhanuddin, Dirman Hanafi Bakhti, Haddi Chabira, Chaima |
| author_facet | Meftah, Sabir Bentoumi, Miloud Burhanuddin, Dirman Hanafi Bakhti, Haddi Chabira, Chaima |
| author_sort | Meftah, Sabir |
| collection | UTHM |
| description | We always face water leakage problems in underground distribution water
networks (DWNs). Existing leak detectors suffer from false alarms due to poor leak signal quality affected by external noise, often collected by acoustic or vibratory sensors. This paper introduces a novel Discrete Wavelet Transform Detector (DWTD) that leverages precise pressure signals non-influenced by environmental noise. Using a prototype of a 100m PEHD pipeline and a diameter of 40mm, Data from two pressure transmitters were collected using a dSPACE MicroLabBox unit. The main idea is to apply the Discrete Wavelet Transform (DWT) with a DONOHO threshold law to cancel noises due to water turbulence fluctuations, ensuring high-quality signals for accurate leak detection and localization. As benchmarks to assess the quality of denoising signals three parameters were calculated, Signal to Noise Ratio (SNR > 26.6763 dB), Normalized Cross-Correlation (NCC≈1), and Mean Square Error (0.20573 < MSE <
48.4761). The denoised temporal signals are obtained from the Inverse Discrete Wavelet Transform (IDWT). A Cross-correlation is employed to these signals to determine the leak’s location. The experimental validation involves positioning the first and second transmitters at specific distances on both sides of the leak position. This allows for comparison between the actual leak position in advance known and calculated positions at various points and leak sizes. With only a few exceptions where the maximum error rate reached 5 meters from the actual leak position, the detector's effectiveness was proven across tests involving four different leak sizes. |
| first_indexed | 2025-03-04T05:05:02Z |
| format | Article |
| id | uthm.eprints-12502 |
| institution | Universiti Tun Hussein Onn Malaysia |
| language | English |
| last_indexed | 2025-03-04T05:05:02Z |
| publishDate | 2024 |
| publisher | ascee |
| record_format | dspace |
| spelling | uthm.eprints-125022025-02-24T07:27:22Z http://eprints.uthm.edu.my/12502/ Novel Leak Detector Based on DWT an Experimental Study Meftah, Sabir Bentoumi, Miloud Burhanuddin, Dirman Hanafi Bakhti, Haddi Chabira, Chaima TJ Mechanical engineering and machinery We always face water leakage problems in underground distribution water networks (DWNs). Existing leak detectors suffer from false alarms due to poor leak signal quality affected by external noise, often collected by acoustic or vibratory sensors. This paper introduces a novel Discrete Wavelet Transform Detector (DWTD) that leverages precise pressure signals non-influenced by environmental noise. Using a prototype of a 100m PEHD pipeline and a diameter of 40mm, Data from two pressure transmitters were collected using a dSPACE MicroLabBox unit. The main idea is to apply the Discrete Wavelet Transform (DWT) with a DONOHO threshold law to cancel noises due to water turbulence fluctuations, ensuring high-quality signals for accurate leak detection and localization. As benchmarks to assess the quality of denoising signals three parameters were calculated, Signal to Noise Ratio (SNR > 26.6763 dB), Normalized Cross-Correlation (NCC≈1), and Mean Square Error (0.20573 < MSE < 48.4761). The denoised temporal signals are obtained from the Inverse Discrete Wavelet Transform (IDWT). A Cross-correlation is employed to these signals to determine the leak’s location. The experimental validation involves positioning the first and second transmitters at specific distances on both sides of the leak position. This allows for comparison between the actual leak position in advance known and calculated positions at various points and leak sizes. With only a few exceptions where the maximum error rate reached 5 meters from the actual leak position, the detector's effectiveness was proven across tests involving four different leak sizes. ascee 2024 Article PeerReviewed text en http://eprints.uthm.edu.my/12502/1/J17990_cf1915af6d02eca4e02b729c3fb64604.pdf Meftah, Sabir and Bentoumi, Miloud and Burhanuddin, Dirman Hanafi and Bakhti, Haddi and Chabira, Chaima (2024) Novel Leak Detector Based on DWT an Experimental Study. International Journal of Robotics and Control Systems, 4 (3). pp. 1109-1134. ISSN 2775-2658 http://dx.doi.org/10.31763/ijrcs.v4i3.1458 |
| spellingShingle | TJ Mechanical engineering and machinery Meftah, Sabir Bentoumi, Miloud Burhanuddin, Dirman Hanafi Bakhti, Haddi Chabira, Chaima Novel Leak Detector Based on DWT an Experimental Study |
| title | Novel Leak Detector Based on DWT an Experimental Study |
| title_full | Novel Leak Detector Based on DWT an Experimental Study |
| title_fullStr | Novel Leak Detector Based on DWT an Experimental Study |
| title_full_unstemmed | Novel Leak Detector Based on DWT an Experimental Study |
| title_short | Novel Leak Detector Based on DWT an Experimental Study |
| title_sort | novel leak detector based on dwt an experimental study |
| topic | TJ Mechanical engineering and machinery |
| url | http://eprints.uthm.edu.my/12502/1/J17990_cf1915af6d02eca4e02b729c3fb64604.pdf |
| work_keys_str_mv | AT meftahsabir novelleakdetectorbasedondwtanexperimentalstudy AT bentoumimiloud novelleakdetectorbasedondwtanexperimentalstudy AT burhanuddindirmanhanafi novelleakdetectorbasedondwtanexperimentalstudy AT bakhtihaddi novelleakdetectorbasedondwtanexperimentalstudy AT chabirachaima novelleakdetectorbasedondwtanexperimentalstudy |