Energy consumption readings for a sample of 5,567 London Households that took part in the UK Power Networks led Low Carbon London project between November 2011 and February 2014.

Readings were taken at half hourly intervals. Households have been allocated to a CACI Acorn group (2010). The customers in the trial were recruited as a balanced sample representative of the Greater London population.

The dataset contains energy consumption, in kWh (per half hour), unique household identifier, date and time, and CACI Acorn group. The CSV file is around 10GB when unzipped and contains around 167million rows.

Within the data set are two groups of customers. The first is a sub-group, of approximately 1100 customers, who were subjected to Dynamic Time of Use (dToU) energy prices throughout the 2013 calendar year period. The tariff prices were given a day ahead via the Smart Meter IHD (In Home Display) or text message to mobile phone. Customers were issued High (67.20p/kWh), Low (3.99p/kWh) or normal (11.76p/kWh) price signals and the times of day these applied. The dates/times and the price signal schedule is availaible as part of this dataset. All non-Time of Use customers were on a flat rate tariff of 14.228pence/kWh.

The signals given were designed to be representative of the types of signal that may be used in the future to manage both high renewable generation (supply following) operation and also test the potential to use high price signals to reduce stress on local distribution grids during periods of stress.

The remaining sample of approximately 4500 customers energy consumption readings were not subject to the dToU tariff.

More information can be found on the Low Carbon London webpage

Some analysis of this data can be seen here.

The Household Electricity Disaggregation dataset provides detailed, appliance-level insights into how households use electricity. Each record corresponds to a single household (user_id) and captures the energy consumed in a specific category during a given period, measured in kilowatt-hours (kWh) and as a percentage of total household electricity.

Key features include: 1. user_id: Unique anonymised identifier for each household, enabling cross-linkage with other datasets. 2. created_at: Timestamp indicating when the disaggregation record was created. 3. id: Unique record identifier. 4. period_type & period: Aggregation period (e.g., month) and specific month of observation. 5. type: Electricity consumption (elec). 6. category: Appliance or usage type, including lighting, cooking, washing, hot water, entertainment, refrigeration, always-on devices, and heating. 7. energy (kWh): Absolute electricity consumption for the category. 8. percentage (%): Share of household electricity consumed by that category; summing all categories per household equals 100%.

The dataset is exclusively electricity-focused and provides insights into household energy behaviour, enabling: - Appliance-level analysis: Understand which categories drive consumption in different households. - Segmentation & profiling: Group households based on dominant energy uses, e.g., EV charging, always-on devices, or high entertainment load. - Energy efficiency evaluation: Identify opportunities to reduce consumption in specific categories. - Behavioural insights: Study patterns like time-of-year changes in usage or appliance adoption trends. - Demand forecasting & modelling: Integrate with half-hourly electricity consumption datasets to enhance predictive models.

This dataset is anonymised and suitable for commercial, research, and policy applications, providing a rich resource for understanding residential electricity consumption at a granular, category-specific level.

Context

The dataset featured below was created by aggregating hourly energy consumption data from individual London homes provided by UK Power Networks. The dataset keeps track of the energy consumption of 5,567 randomly selected households in London from November 2011 to February 2014.

-> This energy dataset is a great addition to this London Weather Dataset. You can join both datasets on the 'date' attribute, after some preprocessing, and perform some interesting data analytics regarding how energy consumption was impacted by the weather in London.

Content

The size for the file featured within this Kaggle dataset is shown below — along with a list of attributes and their description summaries: - london_energy.csv - 3510433 observations x 3 attributes

1. LCLid - individual household unique identifier - (str)
2. Date - date of recorded measurement - (date)
3. KWH - energy consumption measurement in kWh - (float)

Source

Energy Data - https://data.london.gov.uk/dataset/smartmeter-energy-use-data-in-london-households

This dataset contains standardised household electricity consumption records measured at 30-minute intervals, reflecting the industry standard resolution used across the UK and many other countries for smart meter reporting. Each record provides consumption in kilowatt-hours (kWh), alongside the associated fuel type, ensuring compatibility with regulatory frameworks and analytical practices used in energy markets. Key Features

Standardised Half-Hourly Resolution: Consumption is recorded every 30 minutes, aligning directly with the settlement periods used in electricity markets, grid balancing, and billing systems. This makes the dataset particularly useful for modelling, forecasting, and compliance tasks.

Smart Meter Compatible: The structure and frequency of the data comply with smart meter data standards, making it suitable for testing, validating, or demonstrating smart meter infrastructure and analytics workflows.

Clean, Consistent Format: The dataset is provided in a CSV format (halfhourly_readings_elec.csv), with straightforward fields that can be easily ingested into analytical tools, energy modelling platforms, or regulatory reporting systems.

This dataset is a specialised subset of Half-Hourly Electricity Consumption Data | UK Coverage, focusing on households with heat pumps. It currently includes 1,460+ households, with 880+ having 12 months or more of continuous readings, and the number increasing monthly.

All data fields and structure are identical to the main dataset (30-minute intervals, kWh values, smart-meter compatible). This subset supports focused analysis of electrified heating demand, seasonal usage patterns, and the impact of heat pump adoption on household energy consumption and grid load.

This dataset is a specialised subset of Half-Hourly Electricity Consumption Data | UK Coverage, containing data only for households with solar panels. It includes 5,100+ households, with 3,600+ having 12+ months of data, and the number increasing monthly.

All fields, structure, and standards are identical to the main dataset (30-minute intervals, kWh values, smart meter compatible). This subset enables focused analysis of solar household consumption, self-generation impact, and grid interaction within the wider UK electricity dataset.

Department for Environment, Food & Rural Affairs Head Office Building (Nobel House) Half Hourly Electricity Data.

Data shows electricity consumption in kWh.

This dataset is a specialised subset of Half-Hourly Electricity Consumption Data | UK Coverage, focusing exclusively on households with electric vehicles (EVs). It includes 5,500+ households, with 3,600+ having 12 months or more of continuous half-hourly readings, and the number increasing monthly.

All fields and structure mirror the main dataset (30-minute intervals, kWh values, smart-meter standard). This subset enables targeted analysis of EV charging behaviour, load profiles, and the impact of EV adoption on household and grid electricity demand.

The Household Gas Disaggregation dataset provides appliance-level insights into how households use gas across key categories such as heating, hot water, cooking, and other uses. Each record corresponds to a single household (user_id) and reports monthly gas consumption in kilowatt-hours (kWh), along with the percentage contribution of each category to total household gas usage.

Currently, the dataset covers 26,000+ households, with coverage expanding monthly as new data becomes available. This growing dataset enables comprehensive analysis of domestic gas consumption patterns and appliance-level energy use.

Key attributes include:

1. user_id: Unique anonymised household identifier (can be linked with other datasets).
2. created_at: Timestamp for record creation.
3. id: Unique record identifier.
4. period_type & period: Aggregation period (e.g., month) and corresponding date.
5. type: Fuel type (gas).
6. category: End-use category ( heating, hot water, cooking, other.)
7. energy (kWh): Absolute gas consumption for each category.
8. percentage (%): Proportion of total household gas usage represented by each category (sums to 100% per household).

Ideal for: - Energy analysis: Understand household gas consumption patterns at the category level. - Demand forecasting: Support predictive models for heating and hot water demand. - Energy efficiency & decarbonisation research: Identify opportunities for reducing gas consumption and improving building performance. - Behavioural insights: Explore how different households allocate gas use between heating, hot water, and cooking. - Segmentation & policy design: Profile households based on gas use intensity or category distribution. - All data is anonymised to protect household privacy while offering high-value analytical insights.

When linked with other datasets, such as Household Profiles, Property Characteristics, or Half-Hourly Gas Consumption, it supports a wide range of applications in energy analytics, policy design, and data-driven innovation.

Peer-to-peer (P2P) energy sharing involves novel technologies and business models at the demand-side of power systems, which is able to manage the increasing connection of distributed energy resources (DERs). In P2P energy sharing, prosumers directly trade energy with each other to achieve a win-win outcome. A research paper titled "Evaluation of peer-to-peer energy sharing mechanisms based on a multiagent simulation framework" has been published on Applied Energy regarding this topic. In the paper, a general multiagent framework was established to simulate P2P energy sharing, with two original techniques proposed to facilitate simulation convergence. Furthermore, a systematic index system was established to evaluate P2P energy sharing mechanisms from both economic and technical perspectives.In case studies of the paper, two sets of cases were conducted to validate the proposed simulation and evaluation methods and to give some practical implications on applying P2P energy sharing in Great Britain (GB) at present and in the future. The household demand dataset and electric vehicle (EV) dataset used in the paper has been provided for researchers to reproduce the results in the paper or to conduct further related studies. Also, the original numerical data of the results in the case studies of the paper have been provided, for researchers to better understand the results or to use the results for other purposes.The whole dataset includes 9 excel files in total. The detailed description for them are presented as follows:1. “CREST_Demand_Model_v2.2 (Great Britain).xlsm” is a high-resolution stochastic integrated thermal-electrical domestic demand simulation tool developed by Centre for Renewable Energy Systems Technology (CREST) of Loughborough University (refering to http://www.lboro.ac.uk/research/crest/demand-model/). It contains a lot of sheets and VBA codes, which are used to generate “fake” demand curves of domestic customers sampled from statistical distributions that are based on real-life data. In the “Main Sheet”, input parameters like “day of month”, “month of year”, “latitude”, “longitude”, etc. can be entered, and then the “Run simulation” button can be clicked to start the simulation. After the simulation, daily curves like “occupancy and activity”, “total electrical demand”, “total gas demand”, etc. are generated and visualized, with very high time resolution.2. “Electric_Vehicle_Dataset (Great Britain).xlsx” is a dataset based on the research conducted jointly by Centre for Integrated Renewable Energy of Cardiff University and Key Laboratory of Smart Grid of Ministry of Education of Tianjin University (referring to https://doi.org/10.1016/j.apenergy.2015.10.159). It contains two sheets, which provide the parameters of 1000 typical electric vehicles of Great Britain respectively. For each electric vehicle, the parameters include: (1) “Time starting charging / returning home (hour)”, (2) “Time finishing charging / leaving home (hour)”, (3) “Battery capacity (kWh)”, (4) “Energy consumption due to travel (measured by SOC)”, (5) “Lowerlimit of SOC”, (6) “Upperlimit of SOC”, (7) “Maximum charging/discharging power”, (8) “Charging efficiency”, and (9) “Discharging efficiency”.3. “Numerical results and figures _ Case 1-1.xlsx” provides the numerical results of Case 1-1 of the paper. It contains three sheets, providing the data behind Fig. 6, Fig. 7 and Fig. 8 of the paper respectively. In the “Fig. 6” sheet, the “Total Net Consumption (kWh)” and “Total PV Generation (kWh)” under “SDR mechanism” and “conventional paradigm” are provided. In the “Fig. 7” sheet, the “Net energy cost under SDR mechanism (£)” and “Net energy cost under conventional paradigm (£)” of each prosumer are provided. In the “Fig. 8” sheet, the “Internal selling price (£/MWh)”, “Internal buying price (£/MWh)” and “Total Net Energy Cost (£)” of each iteration are provided.4. “Numerical results and figures _ Case 1-2.xlsx” provides the numerical results of Case 1-2 of the paper. It contains two sheets, providing the data behind Fig. 9, Fig. 10 and Fig. 11 of the paper. In the “Fig. 9 and 10” sheet, for Fig. 9, the “The iteration at which the simulation stopped” given different ramping rates are provided; for Fig. 10, the “Overall Performance Index” with different ramping rates given different demand profiles are provided. In the “Fig. 11” sheet, the “Total net energy cost (ramping rate = 0.3) (£)” and “Total Net Energy Cost (ramping rate = 0.6) (£)” at each iteration are provided.5. “Numerical results and figures _ Case 1-3.xlsx” provides the numerical results of Case 1-3 of the paper. It contains only one sheets, providing the data behind Fig. 12 of the paper. In the “Fig. 12” sheet, the “Overall Performance Index” with different learning rates given different demand profiles are provided.6. “Numerical results and figures _ Case 1-4.xlsx” provides the numerical results of Case 1-4 of the paper. It contains two sheets, providing the data behind Fig. 13 and Fig. 14 of the paper. In the “Fig. 13” sheet, the “Overall Performance Index” with different ramping rates given different initial values are provided. In the “Fig. 14” sheet, the “Overall Performance Index” with different learning rates given different initial values are provided.7. “Numerical results and figures _ Case 1-5.xlsx” provides the numerical results of Case 1-5 of the paper. It contains only one sheet, providing the data behind Fig. 15 and Fig. 16 of the paper. In the “Fig. 15 and 16” sheet, for Fig. 15, the number of iterations when the simulation stopped given different maximum number of iterations and ramping rates are provided; for Fig. 16, the overall performance given different maximum number of iterations and ramping rates are provided.8. “Numerical results and figures _ Case 2-2.xlsx” provides the numerical results of Case 2-2 of the paper. It contains only one sheet, providing the data behind Fig. 17 of the paper. In the “Fig. 17” sheet, the overall performance scores of the three mechanisms (SDR, MMR and BS) and conventional paradigm in scenarios with different PV and EV penetration levels are provided.

1. “Numerical results and figures _ Appendix B.xlsx” provides the numerical results of the cases in Appendix B of the paper. It contains two sheets, providing the data behind Fig. B1, Fig. B2, Fig. B3 and Fig. B4 of the paper. In the “Fig. B1 and B2” sheet, for Fig. B1, the EWH power consumption (kW) at t=1 and t=2 for each iteration without any techniques for convergence are provided; for Fig. B2, the Internal buying price (pence/kWh) at t=1 and t=2 without any techniques for convergence are provided. In the “Fig. B3 and B4” sheet, for Fig. B1, the EWH power consumption (kW) at t=1 and t=2 for each iteration with a limitation for its power change are provided; for Fig. B2, the Internal buying price (pence/kWh) at t=1 and t=2 with a limitation for its power change are provided.Research results based upon these data are published at https://doi.org/10.1016/j.apenergy.2018.02.089

The Half-Hourly Gas Consumption dataset provides granular energy usage data for UK households, enabling detailed analysis of gas consumption patterns over time. Each record corresponds to a single household and includes a unique anonymised identifier (user_id), making it possible to link with other Monda datasets such as Household Profiles or Property Characteristics for richer insights.

Key attributes include: 1. user_id: Unique anonymised identifier for each household. 2. consumption_kWh: Gas consumption in kilowatt-hours for each half-hour period. 3. reading_timestamp: Timestamp for each half-hour reading. 4. fuel_type: Type of fuel (gas) associated with the reading.

The dataset currently covers 29,600+ households, with over 20,100 households having data spanning 12 months to 3 years. The number of households included continues to grow weekly, ensuring a constantly expanding dataset for analysis.

This dataset is ideal for: - Energy analysis: Understanding household gas consumption patterns at a granular level. - Demand forecasting: Supporting predictive models for gas usage and peak load management. - Energy efficiency research: Linking consumption with household attributes to identify opportunities for efficiency improvements. - Grid and network planning: Informing infrastructure planning and gas supply management. - Data monetisation: Enabling creation of value-added products when combined with other household or property datasets.

All data is anonymised to protect household privacy while providing accurate, high-resolution information for research, commercial, and policy applications.