Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm
Abstract This research utilizes time series models to forecast electricity generation from renewable energy sources and electricity consumption. The configuration of optimal parameters for these models typically requires optimization algorithms, but conventional algorithms may struggle with fixed se...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-87013-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585776917905408 |
|---|---|
| author | Yang Cao Jun Yu Rui Zhong Masaharu Munetomo |
| author_facet | Yang Cao Jun Yu Rui Zhong Masaharu Munetomo |
| author_sort | Yang Cao |
| collection | DOAJ |
| description | Abstract This research utilizes time series models to forecast electricity generation from renewable energy sources and electricity consumption. The configuration of optimal parameters for these models typically requires optimization algorithms, but conventional algorithms may struggle with fixed search patterns and limited robustness. To address this, we propose an auto-evolution hyper-heuristic algorithm named AE-GAPB. AE-GAPB integrates a genetic algorithm (GA) at the high-level component and employs particle swarm optimization (PSO) and the bat algorithm (BA) at the low-level component. The GA continuously finds the best hyperparameters for PSO and BA based on prediction accuracy, which significantly accelerates the optimization and improves the accuracy. Additionally, the crossover and mutation rates of GA evolve over iteration time and fitness value space, further enhancing its adaptability. We validated AE-GAPB on six time series forecasting models and compared it with five well-known optimization algorithms as well as GAPB without auto-evolution at the high-level component. As a result, AE-GAPB achieved excellent results on renewable energy generation and electricity consumption datasets from the Hokkaido, Kyushu, and Tohoku regions of Japan. |
| format | Article |
| id | doaj-art-602eacc84f344db391b9d5bf35c4404c |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-602eacc84f344db391b9d5bf35c4404c2025-01-26T12:29:56ZengNature PortfolioScientific Reports2045-23222025-01-0115112010.1038/s41598-025-87013-8Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithmYang Cao0Jun Yu1Rui Zhong2Masaharu Munetomo3Graduate School of Information Science and Technology, Hokkaido UniversityInstitute of Science and Technology, Niigata UniversityInformation Initiative Center, Hokkaido UniversityInformation Initiative Center, Hokkaido UniversityAbstract This research utilizes time series models to forecast electricity generation from renewable energy sources and electricity consumption. The configuration of optimal parameters for these models typically requires optimization algorithms, but conventional algorithms may struggle with fixed search patterns and limited robustness. To address this, we propose an auto-evolution hyper-heuristic algorithm named AE-GAPB. AE-GAPB integrates a genetic algorithm (GA) at the high-level component and employs particle swarm optimization (PSO) and the bat algorithm (BA) at the low-level component. The GA continuously finds the best hyperparameters for PSO and BA based on prediction accuracy, which significantly accelerates the optimization and improves the accuracy. Additionally, the crossover and mutation rates of GA evolve over iteration time and fitness value space, further enhancing its adaptability. We validated AE-GAPB on six time series forecasting models and compared it with five well-known optimization algorithms as well as GAPB without auto-evolution at the high-level component. As a result, AE-GAPB achieved excellent results on renewable energy generation and electricity consumption datasets from the Hokkaido, Kyushu, and Tohoku regions of Japan.https://doi.org/10.1038/s41598-025-87013-8Electricity ForecastingRenewable EnergyElectricity ConsumptionAuto-Evolution Hyper-heuristicsTime Series Models |
| spellingShingle | Yang Cao Jun Yu Rui Zhong Masaharu Munetomo Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm Scientific Reports Electricity Forecasting Renewable Energy Electricity Consumption Auto-Evolution Hyper-heuristics Time Series Models |
| title | Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm |
| title_full | Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm |
| title_fullStr | Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm |
| title_full_unstemmed | Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm |
| title_short | Forecasting Renewable energy and electricity consumption using evolutionary hyperheuristic algorithm |
| title_sort | forecasting renewable energy and electricity consumption using evolutionary hyperheuristic algorithm |
| topic | Electricity Forecasting Renewable Energy Electricity Consumption Auto-Evolution Hyper-heuristics Time Series Models |
| url | https://doi.org/10.1038/s41598-025-87013-8 |
| work_keys_str_mv | AT yangcao forecastingrenewableenergyandelectricityconsumptionusingevolutionaryhyperheuristicalgorithm AT junyu forecastingrenewableenergyandelectricityconsumptionusingevolutionaryhyperheuristicalgorithm AT ruizhong forecastingrenewableenergyandelectricityconsumptionusingevolutionaryhyperheuristicalgorithm AT masaharumunetomo forecastingrenewableenergyandelectricityconsumptionusingevolutionaryhyperheuristicalgorithm |