Electricity Consumption Forecasting: An Approach Using Cooperative Ensemble Learning with SHapley Additive exPlanations
arXiv:2605.25305v1 Announce Type: new Abstract: Electricity expense management presents significant challenges, as this resource is susceptible to various influencing factors. In universities, the demand for this resource is rapidly growing with institutional expansion and has a significant environmental impact. In this study, the machine learning models long short-term memory (LSTM), random forest (RF), support vector regression (SVR), and extreme gradient boosting (XGBoost) were trained with historical consumption data from the Federal Institute of Paran\'a (IFPR) over the last seven years a
The increasing focus on sustainable operations and the expansion of institutions like universities are driving the need for more efficient electricity management, making AI-driven forecasting crucial.
Accurate electricity consumption forecasting is vital for optimizing resource allocation, reducing operational costs, and mitigating environmental impact in energy-intensive sectors.
This research demonstrates how cooperative ensemble learning can enhance the precision and interpretability of electricity demand predictions, offering a clearer pathway for practical implementation.
- · Energy utilities
- · Large institutions (universities, data centers)
- · AI/ML model developers
- · Energy management software providers
- · Traditional energy forecasting methods
- · Institutions with inefficient energy management
Improved electricity management leads to reduced energy waste and operational expenses for large consumers.
Wider adoption of such AI models could stabilize local grids by better matching supply and demand, contributing to grid reliability.
The enhanced efficiency might enable more aggressive deployment of compute-intensive technologies by reducing the energy cost burden.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG