The Applicability of a Machine Learning Methodology to Generate TMY Weather Files
Bibliographic record
Abstract
To effectively decarbonize buildings accurate energy models must be created to predict building energy performance. Typical meteorological year (TMY) weather files represent long-term weather conditions and are used in energy modelling to help evaluate energy performance. This thesis explores generating TMY files using machine learning to improve accuracy, which can significantly influence energy simulation results. The current TMY generation approach relies on expert judgment, often overlooking seasonal, climate and application-based variations. \nManuscript #1 introduces a machine learning methodology using feature importance to determine the relevant generation parameters used in the Sandia method to enhance the current TMY generation approach. The proposed methodology is applied to a medium office building in Montreal. The results reveal an improved representativeness of the long-term average building energy demand for the TMY generated using the proposed methodology. \nManuscript #2 aims to (1) assess the applicability of the methodology across Canadian climates; (2) investigate the feasibility of using standardized climate zone-based weighting factors to reduce the computational time associated with extracting location-based weighting factors to facilitate wider adoption of the proposed methodology. The methodology is applied to 18 cities across six Canadian climate zones and generates two weather files for each location. TMYSTATION uses location-based weighting factors while TMYCZ uses climate zone-based weighting factors. The CV(RMSE) and NMBE indicate the proposed weather files outperform the conventional weather files in predicting the long-term energy performance of buildings. Although the TMYSTATION files performed marginally better, the convenience of standardized climate zone-based weighting factors can enhance the methodology’s adaptability.
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 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".