Leakage-Detection-in-Smart-Water-Distribution-Systems

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #go #machine-learning #rust #docker #macos #awesome #cli #vue #ios #mcp

Leakage-Detection-in-Smart-Water-Distribution-Systems

(★ 13)

This repository contains the coding aspect of the project Leakage Detection in Smart Water Distribution Systems, where the collected data is used for training an ANN model to detect leakage.

Leakage-Detection-in-Smart-Water-Distribution-Systems 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Leakage-Detection-in-Smart-Water-Distribution-Systems

![Type](https://img.shields.io/badge/Type-Research_Project-yellow)
![Concepts](https://img.shields.io/badge/Concepts-Deep_Learning-blue)
![Language](https://img.shields.io/badge/Language-Python-red)
![Libraries](https://img.shields.io/badge/Libraries-Keras,_Tensorflow,_Scikit_Learn-green)

This project was done in CSIR-CEERI, Pilani in the Cyber Physical Systems department under the mentorship of Principal Scientist [Dr. Bhausaheb Ashoke Botre](https://www.ceeri.res.in/profiles/bhausaheb-ashok-botre/). This repository contains the coding aspect of the project, where the collected data is used for training an ANN model to detect leakage.

Data - Data was collected experimentally from an IOT-based Smart Water Grid testbed and thus can't be shared publically.

Team Members - Pavni Jaiswal, Neelabh Sinha

### Elements of the Project -

- A major chunk of project involved the study of Leakage Detection techniques, and then, conducting experiments on a physical testbed (existing in CSIR-CEERI, Pilani) which has a pipeline network with all relevant sensors installed at various places, which mimics the arrangement of a typical water distribution network. PLC and SCADA system is installed for control and data collection.

- Thereafter, data was collected by conducting experiments by simulating various types of leak and studying the behaviour of the system.

- Further, the data was carefully analysed to study the dependence of various parameters at various nodes on a leak, and study of how these parameters are interdependent. Further, deciding features were extracted which were to be used as inputs for the classification.

- An ANN based model was trained to after carefully deciding the features (which were temperature, pressure, flow rate and vibration at different nodes of the network) which classifies a state of operation as leak or not a leak.

### Included Files with the repository -
- Data : A portion of used Data to observe the nature
- Analysis_and_Model.ipynb : Analysis of Data, Training the model, Testing the model

### Feature Dependency -

Only the % dependency of leak on various parameters are shown here. Individual Data Analysis graphs and visualizations can be seen from [Visualizations](https://github.com/neelabhsinha/Leakage-Detection-in-Smart-Water-Distribution-Systems/tree/master/Visualizations) Folder.

### Model -

### Results -

Metric | Value (%)
-------|----------
Accuracy | 92%
Precision | 92.5%
Recall | 90.7%
F1-score | 91.4%

Leakage-Detection-in-Smart-Water-Distribution-Systems

Leakage-Detection-in-Smart-Water-Distribution-Systems 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Leakage-Detection-in-Smart-Water-Distribution-Systems

Leakage-Detection-in-Smart-Water-Distribution-Systems 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README