A Kyiv Private Household Smart Plug Dataset for AI-Driven Digital Twin Modeling of Sustainable Residential Energy Systems (doi:10.48788/DVUA/SXITMW)

View:

Part 1: Document Description
Part 2: Study Description
Part 3: Data Files Description
Part 4: Variable Description
Part 5: Other Study-Related Materials
Entire Codebook

(external link)

Document Description

Citation

Title:

A Kyiv Private Household Smart Plug Dataset for AI-Driven Digital Twin Modeling of Sustainable Residential Energy Systems

Identification Number:

doi:10.48788/DVUA/SXITMW

Distributor:

DataverseUA

Date of Distribution:

2026-09-23

Version:

1

Bibliographic Citation:

Boiko, Olha; Anton, Komin; Parfenenko, Yuliia, 2026, "A Kyiv Private Household Smart Plug Dataset for AI-Driven Digital Twin Modeling of Sustainable Residential Energy Systems", https://doi.org/10.48788/DVUA/SXITMW, DataverseUA, V1, UNF:6:7De53OI0uklpWD5HTNMzWw== [fileUNF]

Study Description

Citation

Title:

A Kyiv Private Household Smart Plug Dataset for AI-Driven Digital Twin Modeling of Sustainable Residential Energy Systems

Identification Number:

doi:10.48788/DVUA/SXITMW

Authoring Entity:

Boiko, Olha (Sumy State University)

Anton, Komin (Sumy State University)

Parfenenko, Yuliia (Sumy State University)

Software used in Production:

Python

Grant Number:

No. 0126U000866

Distributor:

DataverseUA

Access Authority:

Boiko, Olha

Depositor:

Olha Boiko

Date of Deposit:

2026-09-22

Holdings Information:

https://doi.org/10.48788/DVUA/SXITMW

Study Scope

Keywords:

Computer and Information Science, Energy consumption--Forecasting, Artificial intelligence, Real-time data processing, Machine learning, Time-series analysis, Digital twins (Computer simulation)

Abstract:

This dataset contains energy consumption data collected for research on energy consumption forecasting and intelligent energy management in hybrid energy systems. The dataset can be used to develop, train, validate, and evaluate machine learning models for short-term electricity consumption forecasting and energy management. It is intended for research in energy informatics, edge AI, TinyML, and smart energy systems.

Kind of Data:

Time Series Data

Notes:

The dataset consists of three CSV files plus a README, comprising a plug-level hourly energy consumption dataset from a private household in Kyiv, Ukraine. The primary file, export_energy_plain.csv, contains 26,263 hourly readings (datetime in UTC, consumption in kWh, and entity_id) collected between 27 August 2025 and 2 September 2026 from three Shelly Smart Plugs: one in the living room (TV, Apple TV, and LED night light; 8,857 records), one in the bathroom (towel dryer and washing machine; 8,629 records), and one in the home office (monitor, laptop, speakers, and the Raspberry Pi 5 that also served as the Home Assistant/InfluxDB data-collection hub; 8,777 records), with an overall missing-data rate of 1.69%. This is supplemented by export_outdoor_temp_plain.csv, containing 8,888 hourly outdoor temperature readings in degrees Celsius that can be joined to the energy data by UTC timestamp, and by vacation_periods.csv, which records two known household absence periods (a winter vacation from 22–28 February 2026 and a summer vacation from 15–24 July 2026) for occupancy-aware analysis. The README also notes qualitative context on scheduled and emergency power outages in Kyiv between 15 October 2025 and 1 March 2026, though these are not provided as exact day-level labels.

Methodology and Processing

Sources Statement

Data Access

Other Study Description Materials

Related Publications

Citation

Title:

Komin, A., & Boiko, O. (2025). A privacy-preserving edge data aggregation for TinyML energy forecasting in households. Technology Audit and Production Reserves, 6(2(86), 31–38. https://doi.org/10.15587/2706-5448.2025.339277

Identification Number:

10.15587/2706-5448.2025.339277

Bibliographic Citation:

Komin, A., & Boiko, O. (2025). A privacy-preserving edge data aggregation for TinyML energy forecasting in households. Technology Audit and Production Reserves, 6(2(86), 31–38. https://doi.org/10.15587/2706-5448.2025.339277

File Description--f1932

File: export_energy_plain.tab

  • Number of cases: 26263

  • No. of variables per record: 3

  • Type of File: text/tab-separated-values

Notes:

UNF:6:EuOrLva31wab1y7cAd2yJw==

File Description--f1930

File: export_outdoor_temp_plain.tab

  • Number of cases: 8888

  • No. of variables per record: 3

  • Type of File: text/tab-separated-values

Notes:

UNF:6:fWB4cJqyVkkVX8GVm5hXPg==

File Description--f1929

File: vacation_periods.tab

  • Number of cases: 2

  • No. of variables per record: 4

  • Type of File: text/tab-separated-values

Notes:

UNF:6:aByXksokKcMOufKuk1uQoA==

Variable Description

List of Variables:

Variables

datetime

f1932 Location:

Variable Format: character

Notes: UNF:6:pK/xD1hrml42LOsOLh5+dw==

Consumption

f1932 Location:

Summary Statistics: Mean 0.04460555209483974; Valid 26263.0; Max. 1.4424109999999928; StDev 0.09096464687482603; Min. 0.0

Variable Format: numeric

Notes: UNF:6:cUrD92kYZwzqh+GlmHArMw==

entity_id

f1932 Location:

Variable Format: character

Notes: UNF:6:+UJM8Y9m5QquYuU+v/5HOw==

datetime

f1930 Location:

Variable Format: character

Notes: UNF:6:xmMkjykKeqXYU4daJhiAvw==

outdoor_temp

f1930 Location:

Summary Statistics: Min. -23.4; Max. 36.8; Valid 8888.0; Mean 9.221455895589663; StDev 10.950364493619386;

Variable Format: numeric

Notes: UNF:6:n4KoKUikha9xcURyNLCygA==

entity_id

f1930 Location:

Variable Format: character

Notes: UNF:6:cH2DcIjF6PuxlPQ4Ibs6mg==

vacation_name

f1929 Location:

Variable Format: character

Notes: UNF:6:nkohokprXb8ko7P3qgrIuA==

start_date

f1929 Location:

Variable Format: character

Notes: UNF:6:noIJdQezKIt+8IK7+vWpNQ==

end_date

f1929 Location:

Variable Format: character

Notes: UNF:6:3RzlmQT0RDM9W7uTjBdfjQ==

season

f1929 Location:

Variable Format: character

Notes: UNF:6:3oUru49PTD7dHzudn4VhaA==

Other Study-Related Materials

Label:

README.txt

Notes:

text/plain