Advanced control and monitoring system for HVAC assets

More Information

eMarketplace

Software, Technologies and Building systems

DESCRIPTION:

A new smart controller helps heat pump systems run more efficiently and reliably. It uses two advanced tools in a small, easy-to-use device. The first tool is a model predictive control (MPC) algorithm that looks at weather forecasts, building use, and energy prices to run heating and cooling at the best times to reduce energy costs. The second tool is a fault detection and diagnosis (FDD) algorithm that watches how well the system is working, spotting problems early and helping fix them before they get worse. Both tools are built into a simple platform—like a Raspberry Pi—making the system flexible and easy to test.

BUSINESS ROUTE

Target Users

Landlords, Real-estate developers, Service providers, Social, public and cooperative housing providers

VALUE

Heat pumps often run at less than 80% of their intended efficiency due to undetected soft faults or inadequate maintenance. While major faults that shut down the system are quickly noticed, soft faults—those that reduce performance without stopping operation—can quietly drive up electricity use and energy bills. The Fault Detection and Diagnosis (FDD) algorithm tackles this issue by identifying and diagnosing these subtle inefficiencies, enabling timely and effective maintenance. At the same time, heat pumps frequently operate during costly peak electricity hours or without regard for actual building occupancy. To address this, the Model Predictive Controller (MPC) uses forecasts and simplified system models to anticipate changes in weather, occupancy, and energy prices. It then schedules heat pump operation during the most cost-effective times, optimizing performance while lowering energy costs and reducing environmental impact.

APPLICATION

The algorithms offer strong potential for continued research and development, with opportunities to expand their features and refine their architecture for broader applicability across diverse building types and systems. Ongoing laboratory experiments are testing their performance in a hardware-in-the-loop setup, validating their effectiveness in new case studies. The long-term vision is to bring these algorithms to market as an integrated, user-friendly solution under a Software-as-a-Service (SaaS) model. Commercialization will be pursued once a high technology readiness level (TRL) is achieved and real-world demonstrations confirm significant energy savings—positioning the product as a smart, scalable tool for next-generation building energy management.