Analysis_of_Smart_meter_London

The goal of this analysis is to understand energy consumption pattern in UK. The dataset consisted of 5,566 London households and their energy consumption between November 2011 and February 2014. The readings for each household were taken every half hour. Households in UK have been allocated to CACI ACORN(Classification of Residential Neighborhoods) groups, which is a geo-demographic segmentation of UK's population. Broadly, there are 6 such groups. Out of these 6 groups, 'Affluent', 'Comfortable', 'Adversity' groups were considered. More information about ACORN groups can be found here.

UK's energy tariff plans are categorized into two groups: Dynamic Time of Use and Standard. The Dynamic Time of Use plan is set up in such a way that each household is informed in advance of the specific times when their electricity tariff would be higher or lower than normal price– High (67.20p/kWh), Low (3.99p/kWh) or normal (11.76p/kWh). The Standard plan has a constant flat rate(14.228p/kWh) throughout the day.

This dataset was used to understand energy consumption patterns across different ACORN groups. Initial analysis consisted of understanding the different ACORN groups and the difference between them. The next steps of analysis consisted of a deep dive into the dataset which contains the records of household energy consumption (monitored every half hour) to try to establish different consumption patterns based on tariff plans and the ACORN groups. Analysis of weather impacts on energy consumption across groups was also conducted. The resulting output comprised of visualizations of various parameters which exhibit energy consumption trends within and across ACORN groups.

Dependencies:

Python Packages used:

1. pandas
2. numpy
3. matplotlib
4. seaborn
5. datetime
6. plotly
7. calendar

Datasets used:

Note: Due to huge data size, we havent kept a dataset folder. Datasets can be found here:

• Dataset from Kaggle
• Tariff rates

1. Informations_households.xls : It contains all the information on the households including their ACORN group classification and type of tariff plan to which they are subscribed.
2. Halfhourly_dataset.zip : This zip file contains the block files with the half-hourly smart meter measurement. For this analysis, only a few blocks representing each ACORN group( Affluent/ Comfortable/ Adversity) were used. Blocks used : Block_0, Block_2, Block_4, Block_62, Block_78, Block_79, Block_80, Block_95, Block_96, Block_105
3. Acorn_details.xls : It contains detail on acorn groups and profile of individuals in each group based on social, demographic, economic and other differentiating aspects.The first three columns are the attributes studied, the ACORN-* is the index of the attribute for each acorn group. At a national scale, the index is 100. If for one column the value is 150, it means that there are 1.5 times more people with this attribute in the ACORN group than at the national scale. You can find an explanation on the CACI website .
4. Weather_hourly_darksky.csv : It contains hourly data from darksky api. It consists of hourly weather information across UK including minimum-maximum temperature and wind data.
5. Tariff:It contains half hourly tariff rates(High, Normal, Low) for the households subscribed to Dynamic Time of Use plan for the year 2013.

Table Of Contents:

Data preparation

Exploratory data analysis and visualizations

Inline Visualization

Summary

Next Steps

Inline Visualizations

Energy usage is very high during evening hours, less during day time and mean energy usage dips down to the lowest during late night hours. This pattern remains same irrespective of their tariff rates (high, normal, low). Also, it is seen that less energy is consumed when high rates are charged.

The average energy consumption by different tariff plan subcribers for each ACORN classification over a span of roughly 2 years. It was seen that mean energy consumption for DToU subcribers were lower than the standard tariff plan subscribers in the case of Affluent and Adversity ACORN groups whereas for the comfortable ACORN group it was exactly opposite.

Energy consumption is high from December to March. Temperature (green Line) is plotted on Y-axis located on the right side, the average temperature for the same period was below 5°C and thus, this explains higher energy consumption during this period as heating systems would have been used extensively. Thus, temperature change plays a major role in energy consumption irrespective of different ACORN groups.

Summary

To summarize, energy consumption depends on:

• Temperature
• Applied tariff rates
• Different tariff plans
• ACORN classification

Next Steps

some visualizations could be plotted for the following :

1. Energy consumption pattern during weekday v/s weekend
2. How holiday season impacts energy consumption
3. forecasting energy consumption