### Abstract

For a daily-basis scheduling of an energy system, energy management system often enough to use not a global optimal scheduling but a near optimal scheduling. The article proposes an online scheduling framework without online optimization. The framework is built from two encoder-decoder architectures to extract features of time series; a multi-layer long short-term memory regression model for multi-step time-series forecasting, and multi-class and single-label classification model for on/off scheduling of a device. The models are estimated at offline, and return scheduling from historical time series as input, at online. We evaluate the accuracy of scheduling from the viewpoint of Kullback-Leibler divergence which measures the dissimilarity between two probability distributions. Through the numerical experiments, we demonstrate the usefulness of the proposed framework.

Original language | English |
---|---|

Title of host publication | ECOS 2019 - Proceedings of the 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems |

Editors | Wojciech Stanek, Pawel Gladysz, Sebastian Werle, Wojciech Adamczyk |

Publisher | Institute of Thermal Technology |

Pages | 1327-1335 |

Number of pages | 9 |

ISBN (Electronic) | 9788361506515 |

Publication status | Published - 2019 |

Event | 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, ECOS 2019 - Wroclaw, Poland Duration: 2019 Jun 23 → 2019 Jun 28 |

### Publication series

Name | ECOS 2019 - Proceedings of the 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems |
---|

### Conference

Conference | 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, ECOS 2019 |
---|---|

Country | Poland |

City | Wroclaw |

Period | 19/6/23 → 19/6/28 |

### Keywords

- Encoder-decoder
- Energy demand forecast
- Home energy management system
- Recurrent neural network
- Scheduling

### ASJC Scopus subject areas

- Energy(all)
- Engineering(all)
- Environmental Science(all)

## Fingerprint Dive into the research topics of 'An online HEMS scheduling method based on deep recurrent neural network'. Together they form a unique fingerprint.

## Cite this

*ECOS 2019 - Proceedings of the 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems*(pp. 1327-1335). (ECOS 2019 - Proceedings of the 32nd International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems). Institute of Thermal Technology.