AI-Based, Continuous, and Automated Condition Monitoring and Optimization of Building Hydraulic Systems for Heat Supply

The heat transition requires significantly more efficient utilization of the heat transferred to a building via district heating. Furthermore, it is essential to achieve a lower overall temperature level in order to effectively utilize renewable sources for space heating and domestic hot water, as well as to maximize the capacity of district heating pipelines. The legally mandated hydraulic balancing is an important step toward this goal. This process is carried out by qualified personnel using a manual and highly detailed procedure. The goal of the project is to develop a method that uses measurement data to continuously identify the buildings where action is actually needed. This is intended to ensure that the HVAC specialists – a key resource for implementing the heating transition – can be deployed in a more targeted manner.

Initial Situation

Today, the hydraulic condition of buildings is generally not monitored on an ongoing basis. There is usually no reliable way to determine which buildings actually require optimization measures–decisions are based on individual site inspections or blanket guidelines rather than on measurement data. The legally required hydraulic balancing (GEG §60) addresses this only once and exclusively through manual work performed by HVAC specialists, who are not expected to be available in sufficient numbers in the foreseeable future. In practice, quality control after implementation rarely takes place, and drinking water heating is usually not taken into account, even though it has a significant impact on the hydraulic quality of the building systems, particularly in multi-family buildings.

Objective

The initial objective of the project is to develop an intelligent digital method that automatically evaluates the hydraulic condition of buildings based on measurement data, thereby identifying the actual need for optimization. Building on this, the second step will involve developing a data-driven method for the continuous, automatic optimization of the entire building’s hydraulic system–including drinking water heating–and for automatic parameter adjustment. These methods are intended to provide approaches for revising the existing regulatory framework for hydraulic balancing and to expand it to include continuous, measurement-data-based quality assurance of the hydraulic system in building heating systems.

Approach

Digital, measurement-data-driven methods can automatically identify which buildings actually require optimization and provide targeted support for improving the system. This allows for more efficient deployment of skilled personnel, more efficient utilization of district heating system capacity, and, as a result, a reduction in CO2 emissions and heating costs.

Fraunhofer ISE is developing the conceptual and methodological foundations for automated hydraulic condition assessment: This includes criteria and classifications for categorizing buildings based on their hydraulic condition, as well as semantic data models that make building and system technology usable for digital processes. Based on historical time series, benchmarks are being developed to evaluate the use of AI algorithms among the project partners. In addition, a digital reporting system is being created to continuously document the hydraulic condition, along with a simplified calculation scheme that compares the labor costs of troubleshooting with the achieved benefits.

The “KIDra” project is funded by the Federal Ministry of Economic Affairs and Energy (BMWE).