Energy

Heat Load Forecast Agent

On a cold morning, whether the peak load boiler fires up comes down to a spreadsheet and experience.

Brings networks and generation plants together in one schedule and saves boiler hours.

Brings networks and generation plants together in one schedule and saves boiler hours.

Open agent

Enlarge demo

Public administration

Four processes that take up the most time today

Developed in Germany · Hosted in the EU

Developed in Germany · Hosted in the EU or on-premises

Delivered together with

Sovereign hosting in Germany, single-tenant from the first user.

Sovereign hosting in Germany, single-tenant from the first user.

Legal, tax and compliance expertise for legally sound deployment.

Legal, tax and compliance expertise for legally sound deployment.

Advisory and implementation for the public sector, starting with the AI readiness assessment.

Advisory and implementation for the public sector, starting with the AI readiness assessment.

Two boiler hours, caused by a forecast made by hand.

Heat load depends on outdoor temperature, wind, day of the week and the inertia of the network. Operations management estimates it in advance for each network and then decides which plant runs when. If the estimate is too high, the peak load boiler runs for no reason; if it is too low, the boiler has to start up at short notice.

3

heating networks

Each network has its own inertia and its own customer mix.

5

generation plants

Two combined heat and power plants, a heating plant and two peak load boilers share the schedule.

96

quarter-hours

For each one, someone decides which plant covers the load.

180 hours

of peak load per heating season

The boiler runs when the estimate comes too late or is too rough.

The headline figure targets these 180 boiler hours: 30% fewer peak load hours per heating season.

The 30% reduction in peak load hours is this agent's headline figure. Networks, plants, quarter-hours and boiler hours reflect a typical heating season at a municipal utility and do not come from any measurement.

Three steps through the process.

The agent forecasts the heat load for each network and proposes a schedule for the plants; nothing runs until operations management approves it.

1.

1.

Forecast the heat load for each network

The agent calculates a load for every quarter-hour from the weather forecast, the day of the week and the load profiles of previous weeks. Each network gets its own curve.

The agent calculates a load for every quarter-hour from the weather forecast, the day of the week and the load profiles of previous weeks. Each network gets its own curve.

Forecast for 3 networks over 96 quarter-hours, peak of 41.2 MW at 07:15

2.

2.

Schedule plants against the curve

The agent covers the curve with the available plants, taking account of minimum run times and reported unavailability. The peak load boiler comes last.

The agent covers the curve with the available plants, taking account of minimum run times and reported unavailability. The peak load boiler comes last.

CHP 2 on base load from 04:30, boiler 2 on standby from 06:45

3.

3.

Submit the schedule for approval

The proposal goes to operations management with a reason for each plant start. Changes from the previous day are at the top.

The proposal goes to operations management with a reason for each plant start. Changes from the previous day are at the top.

fahrplan_23-01-2026.csv, 5 plants, 2 changes from the previous day

What the agent does not do.

It does not start up any plant.

It does not start up any plant.

The schedule remains a proposal in the control centre. Technical operations management approves it and carries out the switching.

It does not replace the fault report.

It does not replace the fault report.

If CHP 2 fails, it only notices once the report is in the control system. Until then, it works with the reported availability.

It does not decide when a faulty plant returns to service.

It does not decide when a faulty plant returns to service.

It takes the date maintenance enters for boiler 1 after the burner fault unchanged, and plans without the plant until then. *Maintenance reports in the control system whether the spare part delivery will meet that date.*

Connected systems

Network control centre's control system

load profile database

weather data service

Operations

Deployment model

EU cloud or private cloud

EU cloud or private cloud

Data processing

GDPR-compliant, no model training on your data

GDPR-compliant, no model training on your data

Responsible teams

Network control centre, generation, dispatch

Network control centre, generation, dispatch

The questions that come up before the first meeting.

How quickly is the Heat Load Forecast Agent ready to use?

Where is the data processed?

Where does the figure “30% fewer peak load hours” come from?

Related agents.

All agents

Platform

Industries

Services

Resources

Company

Pricing

From AI strategy to a running platform: we show you the way.

Talk to us

Log in

DE

EN

English

Platform

Industries

Services

Resources

Company

Pricing

From AI strategy to a running platform: we show you the way.

Talk to us

Log in

DE

EN

English
neuland.ai

The AI management and orchestration platform from Cologne.

Developed in Cologne. Operated in the EU or in your data center.

Certifications and compliance: ISO 27001, ISO 9001, GDPR, EU AI Act, DORA, StBerG, and BRAO
WirtschaftsWoche Best of Technology 2026 seal, 2nd place in the Process Management category for neuland.ai

© 2026 neuland AI AG