AI agent for energy engineers
Waste Heat Recovery Opportunity Agent
A ranked list of feasible heat recovery projects with recoverable heat and payback
What it does
Most plants have waste heat in one place and a need for heat in another, but nobody has lined them up. This agent lists heat sources and heat demands from process data, with temperature, flow and timing. It matches them by temperature level and by time of day, estimates recoverable heat and payback, and ranks matches. For each promising match it tests practical limits such as piping distance, fouling risk and shutdown windows. If a match fails, it moves to the next. The engineer approves which projects to study further. Edge case: an exhaust stream is hot enough but dirty, so the agent tests cleaning cost and drops the match if fouling makes payback too long.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- Survey or process change
- Read process data and operating schedules
- List heat sources and demands with temperature and timing
- Match sources to demands by temperature and time overlap
- Estimate recoverable heat and payback
- Do distance, fouling and timing allow the match?If not: drop the match or propose a buffer, then try the next. Back to step 4.
- Test sensitivity to energy price and utilization
- Is payback within the screening limit under the sensitivity range?If not: mark as marginal and note what would change it. Back to step 5.
- Rank projects and draft the summary
- Engineer approves projects for studyThe agent waits here for your OK.
- Heat recovery opportunity list
How it decides
A match needs the source to be hotter than the demand by a minimum approach temperature and overlap in time. Rank by simple payback.
- Require a minimum approach temperature of 10 degrees
- Reject matches with less than 50% time overlap
- Screen out payback over 5 years
- Flag fouling streams for added cleaning cost
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Minimum approach temperature (default 10 degrees)
- Payback limit (default 5 years)
- Energy price assumptions
- Sources to include
What keeps you in control
It always asks you first
- Projects selected for further study
Hard limits
- Never commits capital
- Never ignores safety limits on process streams
It stops when
- Done: ranked list approved
- Stop: process data is incomplete
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide
