Designing Your Agent

Before we start building an agent, we need to design it - to decide what it will do, and what it will do it on.

Let's start with a fairly basic example.

In our database, we an asset called the Main Pump, which has the property Motor Load.

This is a percentage. We'd like to create an AI that tells us whenever the motor current is over 100%.

So, referencing the anatomy of an agent, we can start filling in our details for inputs and our algorithm.

The inputs are the two speeds we want to compare. We will call the first input 'Actual', and the second input 'Target'.

"inputs": [ 
     { "name": "Load" } 
   ],

Next, we choose which algorithm we want to use.

The weak Property Pattern algorithm is ideal to use here - it returns a '0' or a '1' based on if your properties fit a specific pattern.

   "style": "PropertyPattern"

Next, we want to go through the settings for this algorithm.