Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps
Forecast developers predominantly assess residuals and error statistics when tuning the targeted model’s quality. With that, eventual cost or rewards of the underlying business application are typically not considered in the model development phase. The analysis of the power system wholesale market...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-05-01
|
Series: | Forecasting |
Subjects: | |
Online Access: | https://www.mdpi.com/2571-9394/4/2/28 |
_version_ | 1797487465466953728 |
---|---|
author | Gonca Gürses-Tran Antonello Monti |
author_facet | Gonca Gürses-Tran Antonello Monti |
author_sort | Gonca Gürses-Tran |
collection | DOAJ |
description | Forecast developers predominantly assess residuals and error statistics when tuning the targeted model’s quality. With that, eventual cost or rewards of the underlying business application are typically not considered in the model development phase. The analysis of the power system wholesale market allows us to translate a time series forecast method’s quality to its respective business value. For instance, near real-time capacity procurement takes place in the wholesale market, which is subject to complex interrelations of system operators’ grid activities and balancing parties’ scheduling behavior. Such forecasting tasks can hardly be solved with model-driven approaches because of the large solution space and non-convexity of the optimization problem. Thus, we generate load forecasts by means of a data-driven based forecasting tool <i>ProLoaF</i>, which we benchmark with state-of-the-art baseline models and the auto-machine learning models <i>auto.arima</i> and <i>Facebook Prophet</i>. |
first_indexed | 2024-03-09T23:48:03Z |
format | Article |
id | doaj.art-b4e97b12118a40a0a3a0bca1e72d1fa6 |
institution | Directory Open Access Journal |
issn | 2571-9394 |
language | English |
last_indexed | 2024-03-09T23:48:03Z |
publishDate | 2022-05-01 |
publisher | MDPI AG |
record_format | Article |
series | Forecasting |
spelling | doaj.art-b4e97b12118a40a0a3a0bca1e72d1fa62023-11-23T16:39:38ZengMDPI AGForecasting2571-93942022-05-014250152410.3390/forecast4020028Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOpsGonca Gürses-Tran0Antonello Monti1Institute for Automation of Complex Power Systems, E.ON Energy Research Center, RWTH Aachen University, 52064 Aachen, GermanyInstitute for Automation of Complex Power Systems, E.ON Energy Research Center, RWTH Aachen University, 52064 Aachen, GermanyForecast developers predominantly assess residuals and error statistics when tuning the targeted model’s quality. With that, eventual cost or rewards of the underlying business application are typically not considered in the model development phase. The analysis of the power system wholesale market allows us to translate a time series forecast method’s quality to its respective business value. For instance, near real-time capacity procurement takes place in the wholesale market, which is subject to complex interrelations of system operators’ grid activities and balancing parties’ scheduling behavior. Such forecasting tasks can hardly be solved with model-driven approaches because of the large solution space and non-convexity of the optimization problem. Thus, we generate load forecasts by means of a data-driven based forecasting tool <i>ProLoaF</i>, which we benchmark with state-of-the-art baseline models and the auto-machine learning models <i>auto.arima</i> and <i>Facebook Prophet</i>.https://www.mdpi.com/2571-9394/4/2/28MLOpsprobabilistic load forecastinguncertaintygrid operationcongestion management |
spellingShingle | Gonca Gürses-Tran Antonello Monti Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps Forecasting MLOps probabilistic load forecasting uncertainty grid operation congestion management |
title | Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps |
title_full | Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps |
title_fullStr | Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps |
title_full_unstemmed | Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps |
title_short | Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps |
title_sort | advances in time series forecasting development for power systems operation with mlops |
topic | MLOps probabilistic load forecasting uncertainty grid operation congestion management |
url | https://www.mdpi.com/2571-9394/4/2/28 |
work_keys_str_mv | AT goncagursestran advancesintimeseriesforecastingdevelopmentforpowersystemsoperationwithmlops AT antonellomonti advancesintimeseriesforecastingdevelopmentforpowersystemsoperationwithmlops |