Machine learning (ML) applications often require large datasets, a requirement that can pose a major challenge in fields where data is sparse or inconsistent. To address this issue, we propose a novel approach that combines prior knowledge with data-driven methods to significantly reduce data dependency. This study represents knowledge by the method of Component-Based Machine Learning (CBML) as the structure of the system and parameters in the context of energy-efficient building engineering. In this way, CBML incorporates semantic domain knowledge within the structure of a data-driven model. To understand the advantage of CBML, we conducted a case experiment to assess the effectiveness of this knowledge-encoded ML approach in scenarios with sparse data input (1% - 0.0125% sampling rate) and several typical ML methods. Our findings reveal three key advantages of this approach over traditional ML methods: 1) It significantly improves the robustness of ML models when dealing with extremely small and inconsistent datasets; 2) It allows for efficient utilization of data from diverse record collections; 3) It can handle incomplete data while maintaining high interpretability and reducing training time. These features offer a promising solution to the challenges associated with deploying data-intensive methods and contribute to more efficient real-world data usage. Additionally, we outline four essential prerequisites to ensure the successful integration of prior knowledge and ML generalization in target scenarios and open-sourced the code and dataset for community reproduction.

Increasing sustainability requirements make evaluating different design options for identifying energy-efficient design ever more important. These requirements demand simulation models that are not only accurate but also fast. Machine Learning (ML) enables effective mimicry of Building Performance Simulation (BPS) while generating results much faster than BPS. Component-Based Machine Learning (CBML) enhances the capabilities of the monolithic ML model. Extending monolithic ML approach, the paper presents deep-learning architectures, component development methods and evaluates their suitability for space exploration in building design. Results indicate that deep learning increases the performance of models over simple artificial neural network models. Methods such as transfer learning and Multi-Task Learning make the component development process more efficient. Testing the deep-learning model on 201 new design cases indicates that its cooling energy prediction (R2: 0.983) is similar to BPS, while errors for heating energy predictions (R2: 0.848) are higher than BPS. Higher heating energy prediction error can be resolved by collecting heating data using better design space sampling methods that cover the heating demand distribution effectively. Given that the accuracy of the deep-learning model for heating predictions can be increased, the major advantage of deep-learning models over BPS is their high computation speed. BPS required 1145 s to simulate 201 design cases. Using the deep-learning model, similar results can be obtained in 0.9 s. High computation speed makes deep-learning models suitable for design space exploration.

Retrofitting the existing building stock is among the most important objectives and imperative to meet societal goals to reduce primary energy demand and anthropogenic greenhouse gas emissions. District heating systems have proven to supply heat for buildings both energy and cost efficiently. Thermal microgrids (TMGs) can be understood as a subcategory of district heating systems: small scale, bidirectional, and potentially fed by different thermal sources. Given a suitable combination of loads, the number of and distance between buildings, they can offer economic and environmental advantages compared to the supply by individual heating systems per building. We present a novel method using data analysis techniques on georeferenced building stock data to identify suitable configurations of buildings that yield a cost-efficient TMG. For the identification, both semantic and spatial data from a database are combined using fuzzy logics and cost-benefit analysis. We apply the method using a case study featuring a database of 306 buildings potentially to be retrofitted. As a result, we can identify nine groups of 25 buildings that would form a microgrid featuring up to 17.4% cost benefits compared to an individual heat supply. This would save approximately 30% of the building-induced CO2emission of the community.