Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Comprehensive Energy Management Application Method considering Smart Home Occupant Behavior using IoT and Real Big Data

S. Saba Rafiei, Mahdi S. Naderi, Mehrdad Abedi

TL;DR

This work presents an IoT-enabled, decentralized energy management framework for smart homes that fuses high-resolution forecasting with MILP-based load scheduling. It utilizes four years of 1-minute real data including PV and EV integration, comparing multiple deep learning predictors and identifying Seq2Seq with attention as superior for day-ahead forecasts. The optimization minimizes daily grid cost while honoring comfort and power-balance constraints, incorporating TOU and RTP pricing and allowing grid selling. Results show substantial gains in cost reduction ($\$62.05\%$), load variability reduction ($\leq 19.70\%$), and PAR smoothing ($\leq 20.89\%$), underscoring the importance of accurately modeling non-controllable loads and the benefits of edge-enabled, IoT-driven smart-home energy management.

Abstract

One of the most far-reaching use cases of the internet of things is in smart grid and smart home operation. The smart home concept allows residents to control, monitor, and manage their energy consumption with minimum loss and self-involvement. Since each household's lifestyle and energy consumption is unique, the management system needs background knowledge about residents' energy consumption behavioral patterns for more accurate planning. To obtain this information, data related to residents' consumption records must be processed. This research has attempted to provide an optimal decentralized management system consisting of interoperable sections to forecast, optimize, schedule, and implement load management in a smart home. Comparing different prediction models using 4 years of 1-min interval real data of a smart home with photovoltaic generation (PV) and electric vehicle (EV), forecasting non-controllable loads and taking a deterministic approach in different scenarios, the system uses mixed integer linear programming (MILP) to provide load scheduling with the objective of an optimal total energy cost reduction with minimum changes in the household's desired consumption compared to the initial state. The results have shown that the proposed system has reliable performance due to the high precision of the forecast and has led to increased energy efficiency, reduced energy cost (up to 62. 05\%), reduced peak-to-average ratio (PAR) (up to 44. 19\%) and reduced standard deviation (SD) (up to 19. 70\%) in net consumption.

A Comprehensive Energy Management Application Method considering Smart Home Occupant Behavior using IoT and Real Big Data

TL;DR

This work presents an IoT-enabled, decentralized energy management framework for smart homes that fuses high-resolution forecasting with MILP-based load scheduling. It utilizes four years of 1-minute real data including PV and EV integration, comparing multiple deep learning predictors and identifying Seq2Seq with attention as superior for day-ahead forecasts. The optimization minimizes daily grid cost while honoring comfort and power-balance constraints, incorporating TOU and RTP pricing and allowing grid selling. Results show substantial gains in cost reduction (62.05\%\leq 19.70\%\leq 20.89\%$), underscoring the importance of accurately modeling non-controllable loads and the benefits of edge-enabled, IoT-driven smart-home energy management.

Abstract

One of the most far-reaching use cases of the internet of things is in smart grid and smart home operation. The smart home concept allows residents to control, monitor, and manage their energy consumption with minimum loss and self-involvement. Since each household's lifestyle and energy consumption is unique, the management system needs background knowledge about residents' energy consumption behavioral patterns for more accurate planning. To obtain this information, data related to residents' consumption records must be processed. This research has attempted to provide an optimal decentralized management system consisting of interoperable sections to forecast, optimize, schedule, and implement load management in a smart home. Comparing different prediction models using 4 years of 1-min interval real data of a smart home with photovoltaic generation (PV) and electric vehicle (EV), forecasting non-controllable loads and taking a deterministic approach in different scenarios, the system uses mixed integer linear programming (MILP) to provide load scheduling with the objective of an optimal total energy cost reduction with minimum changes in the household's desired consumption compared to the initial state. The results have shown that the proposed system has reliable performance due to the high precision of the forecast and has led to increased energy efficiency, reduced energy cost (up to 62. 05\%), reduced peak-to-average ratio (PAR) (up to 44. 19\%) and reduced standard deviation (SD) (up to 19. 70\%) in net consumption.

Paper Structure

This paper contains 31 sections, 43 equations, 31 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Structure of the Energy Management System
  3. Forecasting Unit
  4. Data Collection and Pre-processing
  5. Handling Missing Data
  6. Data Normalization
  7. Feeding Data to the Model
  8. Configuration of Predictive Models
  9. Prediction Models
  10. Control Unit
  11. Energy Cost Optimization
  12. Objective Function
  13. Power Balance Constraint
  14. Selling to the Grid
  15. Operation Unit
  16. ...and 16 more sections

Figures (31)

  • Figure 1: The structure of the proposed management system
  • Figure 2: Comparison of superior results from search on hyper-parameters to form optimal deep recurrent model
  • Figure 3: Price data relevant to the studied smart home
  • Figure 4: Non-controllable loads based on Max-Mean determination
  • Figure 5: Non-controllable loads based on Median-Mean determination
  • ...and 26 more figures