Monitor a brewing process,  demonstrated MCP for Runtime integration.
How-to → Examples → Industry → SolaPanels MCP

Download the solution: SolarPanels MCP Demo.dbsln

This example demonstrates how to integrate MCP for Runtime with MQTT data and Cloud AI. You will set up a Built-in MQTT Broker, map MQTT data to the solution namespace, create assets, and configure custom MCP tools via Script Classes for solar panel monitoring. *

*Note:** This demo uses **MCP for Runtime**, which connects to a running solution (TServer.exe) to query live data. For AI-assisted solution configuration, see MCP for Designer.

 

Overview

This solution showcases:

  • Runs a Built-in MQTT Broker with SparkPlug simulation

  • Maps MQTT data structure to the solution namespace

  • Organizes data through the Europe asset structure

  • Monitors solar panel data via MCP scripts

  • Provides intelligent insights through Cloud AI integration

  • Enables real-time queries and analysis of MQTT data


Technical Information

MQTT Broker and Simulator

  1. Navigate to Data Explorer → MQTT Tools

  2. Start the Built-in Broker

  3. Start the MQTT Publisher Simulator

  4. Verify the MQTT Broker tree is visible in the explorer

  5. Confirm data is being simulated and published

MQTT Tree to Solution

  1. In Data Explorer, locate the MQTT tree structure

  2. Click in the structure you want to import, in this case Cities → Barcelona → Panel_1

  3. Click the Map to the Solution button

  4. Select Collector and New TagProvider Service and confirm the mapping operation

  5. Navigate to Unified Namespace → Tag Provider

  6. Verify that the Tag Provider has been automatically created

Asset Structure

  1. Navigate to Unified Namespace → Assets

  2. Create a New Folder in root

  3. Name the folder "Europe"

  4. Link the Europe folder to the TagProvider by clicking in Insert data from TagProvider

  5. Verify that you can see the data in the asset

ScriptClass with Custom MCP Tool

  1. Navigate to Scripts → Classes

  2. Click New to create a new script

  3. Select Import code from the Library and choose SolarPanelMCP

  4. This Script Class defines custom MCP tools that MCP for Runtime will expose to AI models

  5. Configure the script parameters for solar panel monitoring

  6. Make the changes and save the script

Run the solution

  1. Go in Runtime → Startup and click in Run Startup

Claude AI

Claude AI connects to your running solution via MCP for Runtime, enabling AI-powered queries of live operational data.

  1. Have Claude AI Desktop downloaded

  2. Go in Settings → Developer → Edit Config and select the ”claude_desktop_config.json”

  3. This .json should have the following content:

    {
  "mcpServers": {
    "<SolutionName>": {
      "command": "<ProductPath>\\fx-10\\net8.0\\TMCPServerStdio\\TMCPServerStdio.exe",
      "args": [ "/host:127.0.0.1", "/port:<port>" ],
      "transport": "stdio"
    }
  }
}

  4. Restart the Claude AI

  5. In the Developer setting it shoud show “running” and when you open a new chat in “Search and Tools” you will see the name of your solution there.

Query in Claude AI

You can query any method in a Claude chat. e.g: “Returns the power output of the SolarPanels in Spain” and it returns the value requested.

Besides the method that are presented in the script, you can ask general information about the solution, like:

  • Get tag historian

  • Get alarm online

  • Get value


In this section...