Using Prophet Algorithm for Pattern Recognition and Short Term Forecasting of Load Demand Based on Seasonality and Exogenous Features
Bibliographic record
Abstract
As smart meters have proliferated in recent years, electrical power companies are dealing with a large volume of data, known as Big Data. Consistent with this issue, data science techniques are necessary to extract the patterns of consumption and forecast load at a specific area. Short-term load forecasting (STLF) models are crucial for electric utilities and play a vital role in generation planning, system operation, system security (false data detection), energy trading (buying/purchasing), reconfiguration, and contingency analysis. Nevertheless, electric consumption forecasting is challenging due to uncertainty incorporated in consumers' behavior. In this paper, we propose a comprehensive two-stage STLF method based on the Prophet Algorithm (PA). In the first stage and after data mining, load consumption patterns are extracted based on trend and seasonality features (e.g., spring, summer, autumn, winter, daily, weekly, etc.) along with holiday effects. Also, to make the model more robust, weather and lag data are added as an exogenous regressors. In the second step the mean absolute percentage error (MAPE) is defined as an objective function to find hyper parameters of the PA. The optimal hyper parameters maximize the accuracy of the fitted model and result in an appropriate prediction based on the historical data. To show the performance of the proposed method Commonwealth Edison Company (CornEd., Northern Illinois, PJM RTO) data set including load and weather data is used. Three case studies are investigated to predict day-ahead load profiles employing different PA hyper parameters. The Simulation results developed in Python programing language reveal that different load patterns can be extracted by this method. Furthermore, different hyper setting results in distinct forecasting while one case can trace light load hours, the other one can follow peak load hours, more accurately. For all of the case studies, the next 24-hour load can be forecasted within an acceptable error range (less than 3%) within the confidence interval.
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".