Development of a contextual bandits-based thermal mass preconditioning algorithm for dynamic electricity pricing
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Preheating / precooling the thermal mass of a building with off-peak electricity can significantly reduce demand for heating / cooling during peak periods. However, an unknown part of this load shifting process is dynamically determining the optimal preconditioning sequence. This paper puts forward a contextual bandits-based algorithm to dynamically optimize preconditioning behaviour. The analysis was conducted by employing a high-fidelity emulator in EnergyPlus, representing a generic small office building. The algorithm iteratively develops univariate change point models for different discrete preconditioning levels, enabling the estimation of a near-optimal preconditioning level for a given building and outdoor temperature forecast for the day. HVAC-related electricity cost savings achieved through this adaptive algorithm varied between 10 and 40% for different peak pricing and envelope scenarios. For all peak pricing and envelope scenarios, the adaptive algorithm was superior to the baseline preconditioning sequences and within 2% of those estimated using a global optimization approach.HighlightsDevelopment of a contextual bandits-based thermal mass preconditioning algorithm for dynamic electricity pricing Optimal load shifting via preconditioning is studied through simulation.EnergyPlus model of a small office building was used as a controls emulator.Control algorithms interacted with the emulator via EnergyPlus' Python API.An adaptive algorithm was developed to autonomously estimate the optimal preconditioning behaviour.The algorithm's performance was comparable to that of global optimization.
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.
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 it