"Deep Learning Analysis” for potential water leakage (Based on Noise Logger)

 

Objective

To propose an in-house developed project which utilize “Deep Learning” to check “possibility of water leakage” by using audio file record by “Noise Logger” (or device with audio recording capability)

Assumption

1.         Available of hardware (Noise logger or audio recording device) which can record audio file of pipeline on site

2.         Available training set of audio files (around 50 - 100 sample files): Around 50 samples from pipeline which has leakage, and 50 samples from pipeline which is normal

 

Work Flow

1.         Front-end staff (PPP, WS, AI) record audio file of pipeline on site (by using Noise Logger or audio recording device), assume audio file is in .WAV format.

2.         Through a simple Web UI:

l   WSD staff (WS or AI) upload .WAV file to SFTP server of this project, with Facility ID of pipeline as a key

l   Click a button of the Web UI to start analysis

3.         Web UI returns result of analysis – about possibility of water leakage

 

Hardware Platform

l   Beside sourcing of “Noise Logger” of “Audio Recording Device”, there is no hardware investment in Phase I (for Phase II development, assume mobile phone is used so that user can get result of analysis on site). Development will base on existing infrastructure of WSDINMS which utilize VMs (VM Servers) of “Gov Cloud” (GCIS)

Software Development (in-house, around 6 .. 9 months)

l   Design script to upload audio file to SFTP folder of “Gov Cloud”

l   Develop software (C#) to convert .WAV file to spectrogram file (ie: image file)

l   Develop software (Python) to train “Deep Learning Engine”

l   Develop software (Python) to predict possibility of water leakage.

l   Develop simple Web UI (.NET ASPX, hosted in IIS) to upload WAV file, return result of prediction, maintain historical records, etc

l   Result will be saved to a “MS-SQL Express Server”

Remark:  All components above is either free-of-charge or “open-license”