Introduction

Within the frame of IRIS Utrecht, data regarding energy consumption is being collected from a number of apartment buildings. Purpose of collecting these data is to be able to evaluate the result of energy saving measures.

To be able to calculate energy consumption, the following variables are needed:

• electricity delivered (from net to connection)
• electricity generated (by PV panels)
• electricity returned (from connection to net)

Using these variables, electricity consumption can be calculated using the following formula:

electricity consumed = electricity delivered + electricity generated - electricity returned

Electricity delivered, produced and generated explained

We are collecting energy consumption from different sources.

Toon smart thermostat

Every night, Quby uploads a file containing energy consumption data from Alexander de Grotelaan I, II, III and IV, Columbuslaan II and Rooseveldlaan I and II to a Amazon S3 bucket. Theses data contain average electricity delivered (from net to connection) and electricity received (from connection to net) for two tariffs (low and normal tariff) at 15 minute intervals. In addition to the average, the files contain the standard deviation and the number of connections involved in calculating the average and standard deviation values. The dataset does not contain energy generated, and can thus not be used to calculate energy consumption. To work around this issue, energy generated by Columbuslaan I (downloaded from the myBeNext API) is used to estimate energy generation by the other apartment buildings.

The S3 file should contain four rows per hour for each apartment building/variable combination. In the picture below, hours for which this is not the case have been marked red. This is probably due to the fact that there are no households producing data for that timestamp/variable. If all variables are missing for a timestamp, this will lead to a missing data point in this dataset. If at least one variable is present for a timestamp, the missing ones will be zero.

myBeNext API

Every hour, quarterly data is being downloaded from the myBeNext API for two apartment buildings: Henriettedreef and Columbuslaan I. For Henriettedreef, data regarding electricity delivered, electricity returned and electricity produced is collected. This information is available at the level of the individual connections, but this information is never stored in the dataset. The dataset contains the sum for the entire apartment building.

Access the data

CKAN datastore API

The data can be downloaded from the CKAN resource included in this package. Because of the size of the dataset, downloading the entire dataset in one go will probably fail due to time outs. The CKAN datastore API can be used to download the dataset in parts.Example code how that can be done can be found here. To be able to run this code, you will need a resource identifier (in this case energy_consumption_observed_cepontod) and (in case of a private dataset) an API key. You can find your API key by logging in to CKAN and clicking on your username in the top right corner of the page. You will then find your API key on the left.

FIWARE ContextBroker

The get entities gets the current state of the different enties. It provides a list of a) the entities and b) the attributes which can be used to compile requests for the Short Term History API.

Short Term History API

The Short Term History (STH) API provides access to historical information for the different entities in both raw and aggregated form.

Columbuslaan II, electricity consumed, raw values

Number of connections corresponding to the previous API call

Columbuslaan II, electricity delivered, aggregated values per month between January 1, 2022 and January 1, 2023