Learning to build Azure Digital Twins.

Most of the docs around ADT focus on building applications, which makes sense. As a data analyst, I think more about using ADT for modeling. For that reason, my approach to using the API is very different.

This repo will, in addition to giving you an updated understanding of Azure Digital Twin's API, give you templates that you can use to build your own digital twin models.

Getting started

from the project root directory

Building the azure resources

docs will have all of the instructuions to build the Azure resources.

Building the local environment

Assuming you already have Anaconda installed and activated, build the environment:

conda env create -f env.yml

Then you need to test your connection to digital twin:

python scripts/test_connection.py

If it returns anything similar to [<azure.digitaltwins.core._generated.models._models_py3.DigitalTwinsModelData object at 0x00000251B4FB3400>, <azure.digitaltwins.core._generated.models._models_py3.DigitalTwinsModelData object at 0x00000251B4FB3730>] You are connected and ready to begin.

Case Studies

Examples of using a digital twin for various purposes. Each one has its own readme and models.

name	description
Basic Tickets and Lines	Basic walkthrough of creating model objects, twins and queries.
Pipes and Tanks	Tank and pipes diagram for simulaiton and monitoring.

Scripts

name	description
test connection	confirms that you can connect to the ADT API.

Here are the docs (the good ones):

• this is the python repo. That's what I'm going through here (usually the main) branch.

Models

• Digital Twins Definition Language

Querying:

• A good overview of the syntax
• Some great examples

Hardware:

• Device catalogue. This is useful because it is a list of devices that are already vetted to work well with ADT, and come included with a json model snippet that you can use for that application. Less discovery work for your team.