Digital Cities - Water Utilities

Application

Three level of a digital city water utilities. Starting with an overview where the live data and status of reservoirs, pump station and treatment plants can be viewed in whole on an interactive map. Flood prediction is actively displayed for the whole area. The next level is the facility level, specifically a pump station for this demonstration, which focuses on aggregate asset performance and energy efficiency. Finally a drilldown of the pump assets for a refined condition monitoring insights and preventative actions. The application is configured using:

1.0 Landing Page

2.0 Pump Station Overview

3.0 Pump Drilldown

Recommendations

Flood Prediction

Proactive detection of potential flood events for timely preventive action.

Condition Monitoring Overview

Comprehensive surveillance of water utility infrastructure for operational health.

Condition Monitoring Asset

Detailed monitoring of each pump’s performance and condition within the system

Data Streams

[1] Flood Prediction

An example of how to contextualize simulated data, predict flooding, receive intelligent suggestions, run recommendations and output the data and results to the Application Designer. The data stream is configured using:

[2] Water Facilities

An example of how to stream facility data to the Application Designer. The data stream is configured using:

[3] Pump Station

Data related to the pump station assets.

[4] Water Pumps

Monitor multiple pump data in real-time for pumps in a water pump station.

[5] Flow Control

Monitor flow data in real-time for a water pump station.

[6] Flow Control Telemetry

Simulate flow data in real-time for a water pump station.

[7] Pumps

Simulate multiple pump data in real-time for pumps in a water pump station.

Notebook

The Notebook Flood Prediction.ipynb can re-run to generate the model file for the Python agent. The datasets used with this model development are in the same folder as the notebook file, you can find them by clicking on the CSV - Weather Data SIngaporeitem in the links list. This process involves training a model and saving the weights - be sure to place the resulting file in a location that the Stream Host can access. Also update the path in the Calculated Field agent for [1] Flood Prediction so that the correct file is used to supply the weights.

Steps to Import

1. Create/confirm variables

Ensure the following variables are available to be used in the data stream:

• App Designer URL
• App Designer Integration Key (Encrypted)

2. Import SQL Tables

• Execute SQL files so that the DemoWorkManagementPump and DemoMaintenanceHistory tables are present and contain data.

3. Save the Python Model

• Download and save the svr_pipeline_model.joblib Joblib file to a chosen directory that is accessible by the Stream Host.
• Note this directory as it needs to be updated in the [1] Flood Prediction Data Stream.
• Ensure scikit-learn, pandas and joblib are installed in the same Python environment.
  • example: pip install pandas in terminal.

4. Import the Data Streams

Follow the guide for applicable parts:

• Select the highest agent version number on import, if prompted
• Assign Access to others as required
• XMPro agents () - ensure the URL & Integration Key are selected
• Recommendation agent () - ensure the URL & Integration Key are selected
• Read Recommendation () - Do not configure yet, this will be configured once the recommendation is imported
• Calculated Field () - ensure the correct model file paths are configured (path chosen in step 2)
• Python agent () - ensure the correct Python version is selected, a stream host has access to a Python runtime, the paths are set and the script is applied.
• SQL Context Provider () - ensure the correct SQL credentials are entered.
• Click Apply and save the data stream
• Publish the data stream and open the live view
• Ensure there is data in the live view by monitoring the agents

5. Import the Recommendations

Map the data stream to import

Assign Access to others as required

6. Set Read Recommendation Agents

Configure the following Read Recommendation agents along with the URL and Key fields:

6.1 Pump Unity & Pumps Unity

Recommendation: Condition Monitoring Asset
Rules: All

7. Import the Application

1.0 Landing Page

Map the data source on import:

2.0 Pump Station Overview

Map the data source on import:

3.0 Pump Drilldown

Map the data source on import:

• After importing, on the Landing Page download Water Utilities Map.html Change the URL in the D3 map to link to the Pump Station Overview page and reupload to the D3 element
  • Specifically the lines:
    "url": "https://xmpro-ad.azurewebsites.net/render;appId=2379;pageId=9890;appVersion=1;categoryId=;categoryName="
• Assign Access to others as required
• Ensure the App Data connection properties are configured and valid

• Edit the application to link the recommendations - Select all recommendations mentioned in this guide.

  Page	Location	Entity ID
  1.0 Landing Page	Bottom Right	null
  2.0 Pump Station Overview	Bottom Right	null
  3.0 Pump Drilldown	Middle Right	Parameter.AssetId

• Save the application
• Publish the application
• Ensure there is data in the application and that the Unity model is receiving its data by hovering over and observing the values