Marcus Voss

AI Expert and Intelligence Architect at Birds on Mars, Ph.D. student at TU Berlin. At CCAI working for the CCAI Wiki (https://wiki.climatechange.ai/)

Datasets on urban trees and irrigation?

admin
We, Birds on Mars, the CityLabBerlin, and the authority responsible for Berlin's trees are just kicking off the project Quantified Trees (Q-Trees) in Berlin on using AI for improving the irrigation of urban trees. The developed system/algorithms will be targeted towards both the authorities, but will also include citizen science aspects (e.g., citizens reporting on the status of a tree). We will also collect some ground truth data on water supply (e.g., usable field capacity) and will have access to the data of project Giessdenkietz.

For the initial SotA, we are now looking for any interesting datasets (some are within this great list) or initiatives that may have data to collaborate with. Β 
Like Comment

Session "Data Science, Climate Change and the Environment" at Data4Good conference CorrelCon (NOVEMBER 12-13)

admin
The CorrelCon is a weekend on data and data science for good organized by CorrelAid. Check out the program and register here: https://correlaid.org/events/2021-11/correlcon/
Like Comment

11/11/2021 - Western Power Distribution's Data Science Challenge Kick off

admin
Following on from the success of the Presumed Open Data, Data Science Challenge, Western Power Distribution partners with Energy Systems Catapult to develop a series of three new data science challenges.

More details and registration here:
https://es.catapult.org.uk/event/western-power-distributions-data-science-challenge-series-kick-off/
Like Comment

List of datasets and review paper on load forecasting in the distribution grid

admin
We recently published our review paper on load forecasting in the (low voltage) distribution grid in Applied Energy (freely available via share link here until November 24, 2021, otherwise here as a preprint).

We provide an overview of load forecasting applications, typical methods used (time series/statistical methods and machine learning), datasets with load data, and trends we found. We conclude the paper with identified problems that we see after the literature review and provide recommendations for action.

In the paper, there is a list of datasets on consumption data at the household level, but also on transformers/substations. We host this list online here with the hope to add more data sets in the future.

So feel free to comment if there is still a dataset missing that you know, or for any feedback and discussion!
Like Comment