> For the complete documentation index, see [llms.txt](https://docs.myrspoven.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.myrspoven.com/how-the-ai-saves-energy/overview.md).

# How myCoreAI works

myCoreAI runs a continuous optimization loop on every building: it trains a model from the building's own data, predicts how setpoint changes affect indoor climate and energy use, and writes setpoints to the Building Management System (BMS) that reduce consumption while staying within agreed comfort bounds. The pages in this chapter document the individual features that shape that loop. Which ones are active depends on contract scope and on what the BMS exposes for reading and writing.

**Asset and facility management:** myCoreAI lowers heating and electricity consumption without capital expenditure, while keeping comfort within contractual limits — improving operating costs and supporting energy management obligations under ISO 50001 and EPBD.

**Technical operations:** myCoreAI provides a transparent optimization layer over the existing BMS. Setpoint writes, comfort bounds, and the energy/comfort trade-off are all configurable per building and per zone.

**Integrators:** myCoreAI consumes BMS datapoints over the standard integrations (BACnet, Modbus, vendor APIs) and writes back through the same channels. No parallel control system is introduced.

## Core behavior

The optimization loop — model training, comfort-vs-energy trade-off, setpoint writing — is always on for any building under myCoreAI control. The features below tune or extend it.

1. **Model.** myCoreAI trains a building-specific model from historical sensor and setpoint data. A new building usually needs about 30 days of data first (see the [FAQ](https://docs.myrspoven.com/get-help/faq.md)).
2. **Predict.** It forecasts indoor climate and energy response to candidate setpoint moves over the optimization horizon.
3. **Decide.** It selects the operating point that minimizes energy or cost under the active comfort bounds.
4. **Write.** It pushes the resulting setpoints to the BMS at the configured cadence.

## How myCoreAI handles equipment

* **Stopped pumps.** myCoreAI writes no setpoints to a heating circuit whose pumps are all stopped. One running pump is enough to keep control.
* **Summer behavior.** Heating circuits switch to summer behavior once half of them have their pumps stopped.
* **Heat recovery first.** On air handling units, myCoreAI uses heat recovery before direct heating.
* **Return-air control.** On air handling units controlled on return temperature, return-air flow is a controllable signal.

## Features

* [**Adaptive min/max**](https://docs.myrspoven.com/how-the-ai-saves-energy/adaptive-min-max.md) — Limits how far heating-circuit setpoints may move from the baseline supply curve.
* [**AI Freedom**](https://docs.myrspoven.com/how-the-ai-saves-energy/ai-freedom.md) — Lets myCoreAI use learned comfort bounds and unoccupied-hours setbacks for larger savings.
* [**Comfort schedule**](https://docs.myrspoven.com/how-the-ai-saves-energy/comfort-schedule.md) — Relaxes comfort bounds and ventilation during scheduled or learned unoccupied hours.
* [**AI Penalty**](https://docs.myrspoven.com/how-the-ai-saves-energy/ai-penalty.md) — The trade-off function that balances energy use against comfort deviations.
* [**Draught control**](https://docs.myrspoven.com/how-the-ai-saves-energy/draught-control.md) — Keeps radiator water warm enough on cold office days to prevent cold draughts at windows.
* [**Groupings**](https://docs.myrspoven.com/how-the-ai-saves-energy/groupings.md) — Logical grouping of sensors for control and reporting.
* [**Indoor climate baseline**](https://docs.myrspoven.com/how-the-ai-saves-energy/indoor-climate-baseline.md) — Comfort bounds learned from the building's own history, used when AI Freedom is on.
* [**myLoadshift**](https://docs.myrspoven.com/how-the-ai-saves-energy/loadshift.md) — Moves heating energy use toward cheaper hours while keeping comfort.
* [**Mixed circuit control**](https://docs.myrspoven.com/how-the-ai-saves-energy/mixed-circuit-control.md) — Control logic for circuits that feed both a downstream circuit and an air handling unit.
* [**Dew point function**](https://docs.myrspoven.com/how-the-ai-saves-energy/dew-point-function.md) — Keeps chilled-water setpoints above the dew point to prevent condensation.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.myrspoven.com/how-the-ai-saves-energy/overview.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
