Project: TransParEnergy

The systematic recording of energy consumption in buildings has traditionally been a time-consuming and labor-intensive process, limiting its application primarily to isolated large-scale studies. As a result, comprehensive and up-to-date energy data at the level of individual buildings or neighborhoods remains scarce. The introduction of a “digital energy twin,” a digital representation that integrates spatial and energy-related data, continues to face major challenges. These include the limited availability and granularity of input data, insufficient mechanisms for secure and standardized data exchange, and issues related to data privacy. As a result, the current digital twin system has shortcomings in terms of the detail, completeness, and consistency of energy data. This research is motivated by the need to close this gap by enabling the seamless integration of various data sources—including architectural plans, images, energy performance certificates, and user-provided data—into a unified framework. By developing a digital energy interface that is compatible with both BIM and GIS platforms, cities will be provided with a robust digital energy twin that can be used to visualize and analyze energy consumption. This tool will not only improve the accessibility of energy data across different areas, but also support evidence-based decision-making, thereby accelerating the energy transition in the built environment.

The goal of this research project is to develop an intelligent, interoperable framework for urban energy analysis by leveraging existing machine learning techniques and digital modeling tools. The project begins with a comprehensive literature review of current machine learning approaches relevant to predicting, classifying, and modeling the energy efficiency of buildings. Data will then be collected from various sources, including architectural designs, energy performance certificates, sensor data, and user-generated inputs. These datasets will serve as the basis for training and validating an AI-based model capable of estimating or classifying the energy consumption characteristics of buildings. The trained AI model is integrated into the digital energy interface, a software platform that supports data analysis, the digital energy twin, and user interaction.

DKSR

Reich & Hölscher

Kreis Lippe (asosciated)

Sennestadt GmbH (associated)