lds-pipelines 0.2.0

A tool to analyze the leakage in an pipeline using modified SPRT technique

🛠️ Leak Detection Analysis Dashboard

👋 Welcome to the Leak Detection Analysis Dashboard

This dashboard helps you analyze likelihood values of leakage from your data — whether it's static (CSV/Excel) or real-time (from OSI-PI server).

---

🚀 Key Features

• 📁 CSV/Excel Analysis: Upload historical data for static analysis using interactive charts.
• 🌐 Real-time Analysis: Connect live to an OSI-PI server and analyze streaming sensor data in real time.

---

🛠️ How to Use

1. Select the desired analysis mode from the sidebar.

2. For static analysis:

   • Upload a CSV or Excel file containing your sensor data.
   • View the interactive visualizations and Zn statistics.

3. For real-time analysis:

   • Choose a start date and time.
   • The dashboard will begin streaming and visualizing the live Zn values.

---

ℹ️ About Likelihood Values (Zn)

The Zn score represents the system’s confidence in detecting disturbances or leaks.

• Zn is computed using a modified SPRT (Sequential Probability Ratio Test) technique — designed for early and reliable detection.
• When Zn crosses the upper threshold, the system is confident that a leak or disturbance has likely occurred.
• If Zn stays below the lower threshold, the system is confident that no disturbance has occurred so far.

---

📝 Sample Data Files

Three sample CSV files have been included in the core module of this project. You can use these files for a quick static analysis demo. These files are based on real-time data collected from the actual field. Please note:

• Pressure values were missing, so we used volume values as pressure values since they don't affect the analysis.

Feel free to experiment with these sample files to get a feel for the application.

---

🖥️ Mock API and Configuration

Since many users may not have access to an API key for connecting to a real-time OSI-PI database, I've created a mock API within the main logic. This mock API will simulate a sample run of the software using mock data.

If you have your own API key:

• You will need to uncomment the relevant sections in the api_processor.py file located in the core module.
• The logic for connecting to the OSI-PI server is already in place. You only need to provide a few details such as:
  • Web ID tags of the sensors
  • Duration for updating data

These parameters can be found and configured within the api_processor.py file.

---

🚀 Running the Program Locally

To run the program locally, use the following command in your terminal:

dashboard

To stop the execution use Ctrl+C in the terminal.