Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for energy engineers

Demand Charge Peak Shaving Agent

Lower monthly peaks through changes that do not disturb production, verified against the next bills.

Demand Charge Peak Shaving Agent: what goes in, what the agent does and what you get

What it does

A plant pays for its single highest 15-minute peak each month, and one start-up of three machines at once can cost thousands. This agent reads the interval data and finds the peaks that set each month's demand charge. It identifies what caused each one by lining up loads, schedules and logs. It then simulates options such as staggering starts, shifting a batch, pre-cooling or adding storage, and checks the savings against operating limits such as production schedules, comfort and equipment constraints. It rechecks the actual result after a change is made. You approve control changes. Edge case: a peak caused by a one-time test is excluded from the pattern but still listed.

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.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueNoNo 1 STARTS WHEN Interval data available 2 USES A TOOL Read interval data and find the peaks that set thedemand charge 3 DOES Identify causes by lining up equipment schedules andlogs 4 DOES Simulate options: staggered starts, load shifts,pre-cooling, storage 5 CHECKS THE RESULT Does each option stay inside the operating limits? If not: adjust the option or drop it and simulate thenext one. Back to step 4. 6 DOES Estimate savings per option 7 YOU APPROVE Manager approves control changes 8 USES A TOOL After changes, read the next period's interval data 9 CHECKS THE RESULT Did the peak drop as simulated? If not: find the new cause and revise the plan. Back tostep 3. 10 RESULT Peak analysis and savings tracker
Read the steps as a list
  1. Interval data available
  2. Read interval data and find the peaks that set the demand charge
  3. Identify causes by lining up equipment schedules and logs
  4. Simulate options: staggered starts, load shifts, pre-cooling, storage
  5. Does each option stay inside the operating limits?If not: adjust the option or drop it and simulate the next one. Back to step 4.
  6. Estimate savings per option
  7. Manager approves control changesThe agent waits here for your OK.
  8. After changes, read the next period's interval data
  9. Did the peak drop as simulated?If not: find the new cause and revise the plan. Back to step 3.
  10. Peak analysis and savings tracker

How it decides

It targets the peaks setting the charge, simulates changes that reduce them, and keeps only those that respect the operating limits and give savings above the threshold.

  • Peak caused by simultaneous starts: stagger by 5 minute steps
  • Option reduces output or comfort beyond the limit: reject
  • Savings under 500 dollars per month: record only
  • Peak from a one-time test: exclude from the pattern, list it

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Demand interval length
  • Operating limits
  • Savings threshold
  • Equipment included
  • Storage options to model

What keeps you in control

It always asks you first

  • Any change to controls or schedules

Hard limits

  • Never change controls directly
  • Never propose changes that violate safety interlocks

It stops when

  • Done: changes verified by data
  • Stop: no option fits operating limits

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.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • 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
Get access to this agent

An example run

What happensIn January, the peak was 1,640 kW at 7:10, caused by three compressors and the oven starting together. Staggering starts by 10 minutes cut it to 1,420 kW in simulation and saved 3,300 dollars monthly. In February the peak was 1,500, higher than simulated. The check failed, and the agent traced it to a chiller reset. The manager approved a second schedule change.

More agents for energy engineers