Short term electric load forecasting for British Columbia, Canada: an exploration of the use of numerical weather prediction data as a predictor in an artificial neural network
Bibliographic record
Abstract
Short term load forecasting (STLF) is used by electric utility companies in their daily operations to match generation with anticipated load. Load forecasting is challenging because electricity demand is dependent on human behaviour and weather. Temperature is the weather variable most commonly used as input to STLF models. The largest electric utility in British Columbia (BC), Canada, BC Hydro, uses Vancouver temperature data as the only input to forecast load for the whole province. To better account for weather patterns across British Columbia, this research explores the use of gridded numerical weather prediction (NWP) data in multi-layer perceptron (MLP) artificial neural network (ANN) short-term load forecast models. Seven experiments are run, that differ by the source of input weather data or number of hidden layers, as follows: (1) point temperature data for Vancouver, mimicking BC Hydro’s operational model; (2) gridded temperature data for BC; (3) gridded temperature, humidity, precipitation, precipitable water, snow depth, and wind speed data for BC; (4) point temperature data for five major BC load centres: Vancouver, Victoria, Abbotsford, Kelowna, Prince George; (5) as in experiment 1, but with a two hidden layer MLP, rather than one; (6) as in experiment 2, but with a two hidden layer MLP; and (7) an ensemble method using weather model ensemble member temperature point forecasts for Vancouver. In all experiments, non-weather input variables including (a) day of the week, (b) hour of the day, (c) month, and (d) previous load values are also used. Results for both hour-ahead and 8-day forecasts show that the use of NWP data does not improve load forecast accuracy, but ensemble forecasts do. Of all seven experiments, an ensemble model (7) is the best, closely followed by a model using Vancouver point temperature data with two hidden layers (5). In both cases where two hidden layers are used in the ANN rather than one, model performance improves.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".