Smart house measurements

Load Forecasting Dataset

Readme File

VARLAB – The Centre for Research & Technology, Hellas [CERTH] - Informatics and Telematics Institute [ITI] - https://varlab.iti.gr/

Authors: Chrysovalantis-George Kontoulis, Georgios Stavropoulos, Dimosthenis Ioannidis

Publication Date: February -, 2023

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreements No. 957406 (TERMINET).

1.Introduction

This dataset features information from a smarthome located at Greece, which features the Mediterranean climate. The building is utilized as a modern workplace that is being used for various every day activities. It is equipped with numerous smart devices and appliances, from smart lights to smart a elevator, while also featuring PVTs.

2.Dataset Overview

2.1Dataset Collection

The system is built on multiple communication protocols including EnOcean, Zigbee, Modbus, BACnet, and, LTE/IEEE 802.15.4 at 2.4GHz. For the sensor data collection, a raspberry Pi microcontroller was used, and data were subsequently transmitted to the storage database.

The extraction period of the data is between 2021-01-01 through 2022-12-20. Along this period there is a total of 66619 unique recordings and the time granularity of the data is set to 15 minutes for all devices.

2.2Data Peculiarities

The place is occupied from Monday to Friday from 9:00 AM GMT+2 (Greenwich Mean Time) all the way through 5:00 PM GMT+2. Note that the building is not active during Greek public holidays, but some computers or servers might be on and consuming electrical energy. Also, there are some irregularities in the data reporting consistency at summer, Christmas & Easter as the building is not occupied for a long time of period. Timestamps of the dataset are in the GMT+2 timezone.

2.3Dataset Structure

This dataset includes a total of six features and it can be used for Electrical, Thermal and Cooling Load forecasting. Electricity Consumption is the consumption of the whole house, Air-condition Status is either 1 or 0 for on and off, respectively, Luminance is how bright a space is, Light Dimming is the dimming of the lights in each room. Finally we have the Indoor Temperature for each room and the Outdoor Temperature.

Data are extracted from four rooms in total. Note that in rooms 1 and 3, there is only one indoor temperature device, thus values are identical for temperature_room_1 and temperature_room_3. Note that sensors have some null values, which is generally either due to inactivity, e.g., the Light Dimming sensor and the Air-condition Status are event-based or due to potential system downtime.

The provided dataset is stored in csv format. A brief overview of the dataset is presented at the Table 3.1.

Table 2.1 Dataset overview

2.4Descriptive Statistics

Table 2.2 provides a brief overview of the key statistical characteristics of the data to. The table presents a summary of important metrics and measures, including measure of central tendency such as the mean, as well as measures of variability such as the standard deviation.

Table 2.2 Descriptive Statistics

3.Acknowledgment

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreements No. 957406 (TERMINET).