TDengine Free Tier Workshop Manual

Hands-On Scenario: Solar Power Monitoring

Workshop objective: You are responsible for monitoring a solar installation with multiple inverters. During the workshop, you’ll turn raw solar telemetry into an organized equipment model, operational analytics, a live dashboard, and AI-powered insights.

Before We Start: Your Data Is Already Ready

There is no data ingestion setup required for this workshop.

TDengine Free Tier includes the sample solar dataset we will use throughout the session. When you start the workshop, the data is already available and ready to explore.

That means there is:

• No OPC UA configuration
• No MQTT configuration
• No CSV import
• No connector setup

We’ll briefly explain how real industrial data gets into TDengine, but we’re intentionally skipping ingestion configuration so we can spend the full workshop working with the data and building something useful.

Please install TDengine here if you have not done so.

0–5 min — Welcome & Understand the Solar Scenario

Scenario: We operate a solar site with multiple inverters. Our job is to understand how the site is performing and identify potential problems.

Steps:

1. Log into TDengine Free Tier.
2. Introduce the sample solar environment.
3. Look at the equipment and measurements available.
4. Explain the workflow we’re going to build: Data → Equipment → Analysis → Dashboard → AI
5. Preview what we’ll have completed by the end.

Checkpoint: Everyone is logged in and can access the preloaded sample data.

5–15 min — Explore the Solar Data

Goal: Understand the industrial data that TDengine has already prepared for us.

Steps:

1. Open the sample solar dataset.
2. Identify the different solar inverters.
3. Look at available measurements such as:
   • AC Power
   • DC Power
   • Voltage
   • Current
   • Temperature
   • Energy Generation
   • Operating Status
4. Explore recent measurements.
5. Look at historical data.
6. Compare measurements between inverters.
7. Explain how this same data could normally arrive through OPC UA, MQTT, PI System, or other industrial sources.

Checkpoint: We understand what data we have and where it comes from—without spending workshop time configuring ingestion.

15–25 min — Organize the Solar Equipment

Goal: Move from raw time-series data to an equipment-centric view.

Steps:

1. Open the solar equipment model.
2. Identify the Solar Site.
3. Locate individual inverters such as:
   • INV-0101
   • INV-0102
   • INV-0103
4. Explore the attributes associated with an inverter.
5. Connect measurements such as power, temperature, and voltage to their equipment context.
6. Review equipment metadata such as model, capacity, and location.
7. Navigate from the site level down to an individual inverter.

Checkpoint:

We’ve gone from:

Raw measurements → Solar Site → Inverters → Attributes

Now the data has operational context.

25–35 min — Analyze Solar Performance

Goal: Turn inverter measurements into useful operational information.

Steps:

1. Open the Analysis Workbench.
2. Plot AC power over time.
3. Compare INV-0101, INV-0102, and INV-0103.

35–47 min — Build a Live Solar Operations Dashboard

Goal: Turn our exploration and calculations into something an operations team could actually use.

Steps:

1. Create a new Solar Operations Dashboard.
2. Create a trend showing inverter power over time.
3. Compare INV-0101, INV-0102, and INV-103.
4. Arrange the visualizations into an operator-friendly view.
5. Save the dashboard.

Checkpoint: We’ve transformed time-series data into a live operational view.

47–55 min — Investigate the Solar Site with AI

Goal: Use AI to investigate the same data we’ve been analyzing manually.

Steps:

1. Open AI Insights.
2. Ask:

“Summarize the performance of the solar site.”

3. Follow up with:

“Which inverter is producing the least power?”

4. Investigate further:

“Compare INV-0101 with the other inverters.”

5. Ask:

“Are there any unusual patterns in INV-03?”

6. Explore possible relationships:

“Is temperature related to the reduction in power output?”

7. Finish with:

“Create a short operational summary and tell me what I should investigate.”

Checkpoint: Instead of manually navigating every measurement, we can use AI to help investigate our industrial data.

55–60 min — From Sample Data to Your Industrial Data

Now we connect everything attendees built back to the real world.

The sample data removed the complexity of ingestion for today’s workshop, but the same workflow applies when TDengine is connected to real industrial systems.

OPC UA → TDengine → Equipment → Analysis → Dashboard → AI

PI System → TDengine → Equipment → Analysis → Dashboard → AI

MQTT → TDengine → Equipment → Analysis → Dashboard → AI

The solar example could just as easily represent:

Solar → Inverters Manufacturing → Machines Utilities → Meters Oil & Gas → Pumps Buildings → HVAC Mining → Conveyors

What You Built in 60 Minutes

Started with: Preloaded solar time-series data
↓
Explored: Inverter measurements and history
↓
Organized: Equipment, attributes, and metadata
↓
Analyzed: Performance and calculated metrics
↓
Visualized: Live operational dashboard
↓
Investigated: AI-powered operational insights

60–90 min — Optional Live Q&A

Use the remaining time for attendees’ actual environments: OPC UA, MQTT, PI System, architecture, migration, dashboards, analytics, AI, deployment, or how to reproduce today’s workflow using their own industrial data.

The important workshop promise becomes very clear: you don’t spend the hour getting data into TDengine—you spend the hour actually using industrial data.