Bibliographic record
Abstract
With the commercial and institutional building sectors using approximately 29% and 34% of all electrical energy consumption in Canada and the United States, respectively, saving electrical energy in commercial and institutional buildings represents an important chal- lenge for both the environment and the energy consumer. Concurrently, a rapid decline in the cost of microprocessing and communication has enabled the profileration of smart me- ters, which allow a customer to monitor energy usage every hour, 15 minutes or even every minute. Algorithmic analysis of this stream of meter readings would allow 1) a building operator to predict the potential cost savings from implemented energy savings measures without engaging the services of an expensive energy expert; and 2) an energy expert to quickly obtain a high-level understanding of a building’s operating parameters without a time-consuming and expensive site visit. This thesis develops an algorithm that takes as input a stream of building meter data and outputs a building’s operating parameters. This output can be used directly by an energy expert to assess a building’s performance; it can also be used as input to other algorithms or systems, such as systems that 1) predict the cost savings from a change in these operating parameters; 2) benchmark a portfolio of building; 3) create baseline models for measurement and verification programs; 4) detect anomalous building behaviour; 5) provide novel data visualization methods; or 6) assess the applicability of demand response programs on a given building. To illustrate this, we show how operating parameters can be used to estimate potential energy savings from energy savings measures and predict building energy consumption. We validate our ap- proach on a range of commercial and institutional buildings in Canada and the United States; our dataset consists of 10 buildings across a variety of geographies and industries and comprises over 21 years of meter data. We use K-fold cross-validation and benchmark our work against a leading black-box prediction algorithm; our model offers comparable prediction accuracy while being far less complex.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".