Development and Demonstration of Surrogate Models to Predict Energy-Related Building Features from Heating and Cooling Load Signature
Bibliographic record
Abstract
This thesis introduces an innovative, efficient, and cost-effective approach for screening office buildings to identify the buildings with the most retrofit potential.Traditional auditing methodologies, including advanced inverse-based methods, often face constraints in terms of engineering cost, time, high fidelity BAS data availability, and scalability.To address these challenges, this research leverages inverse-based machine learning techniques to develop surrogate models for 12 rectangular mid to high rise office buildings.These models predict energy-related building features from widely accessible heating and cooling load data, offering a more accessible and scalable solution.Two methodologies are explored in this research: 1) unsupervised learning of load signatures with clustering algorithms to group probable energy-related building features, and 2) supervised learning of load signatures using an ensemble of artificial neural networks (ANNs) to predict probable energy-related building feature sets.Both methodologies utilize a large training dataset generated through simulation of energyrelated features to obtain the heating and cooling load data.From this simulated data, threeparameter change point models (3P CPMs) are extracted as load signatures, which are then used to predict the energy-related building features.This approach allows for the efficient and accurate characterization of building energy performance, providing a scalable alternative to traditional auditing methodologies.The results indicate that the surrogate models can predict most energy-related building features with reasonable accuracy.Furthermore, the use of clustering to handle multicollinearity in the predictions highlights the potential of these models as an alternative to conventional energy auditing methodologies.This research also presents a user-friendly software tool, demonstrating the practical application of the surrogate model using real-world building consumption data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".