Students who need it can contact them via private message. It is recommended to pay attention to the subscription account platform in the lower right corner of the above picture to download it.
The 4th Jiangsu Big Data Algorithm Competition in 2023 includes 5 tracks, including medical and health track, new energy track, smart transportation track, smart health care track and digital government application track. This data set is the new energy competition Dao's vertical domain data set will provide historical power consumption data of charging stations and other related auxiliary information. Based on these data, contestants need to use machine learning, deep learning, time series and other related technologies to establish prediction models. To predict the power demand in the future, help station managers make wise decisions, optimize the operation and efficiency of charging stations, and promote the sustainable development of the electric vehicle industry.
1.1 Description of competition questions:
In the operation and management of electric vehicle charging stations, accurately predicting the power demand of charging stations is critical to improving charging station operation service levels and optimizing regional power grid supply capabilities. The purpose of this competition is to establish a site charging capacity prediction model, and accurately predict the charging capacity demand of the charging station in a certain period of time in the future based on the relevant information and historical power data of the charging station.
In the competition data, basic information such as the station number, location information, and historical power of the electric vehicle charging station are provided. Contestants are encouraged to supplement or construct additional features based on existing data to obtain better prediction performance.
Based on these data, contestants need to use artificial intelligence-related technologies to build prediction models to predict the demand for electricity in the future, help managers improve the operating efficiency and service levels of charging stations, and promote the overall development of the electric vehicle industry.
1.2 Competition tasks:
Based on the multi-dimensional desensitized data of electric vehicle charging stations provided by the competition question, reasonable features and algorithm models are constructed to estimate the daily charging capacity (in hours) of the station in the next week.
1.3 Data description:
The data set provided for this competition contains 3 data tables. Among them, power_forecast_history.csv is the site operation data, power.csv is the site charging data, stub_info.csv is the site static data, the training set is the data of the historical year, and the test set is the data of the next week.
Remark:
(1) h3 encoding is a system for hierarchical geocoding that divides the earth into different hexagonal grids. Players can try to use h3 encoding to construct additional features related to geographical location.
(2) Desensitized fields do not provide field business descriptions for players to explore freely.
1.4 Evaluation indicators
The evaluation index of this track is RMSE:
Conclusion
The above is all the contents of the new energy vehicle charging data set. To download the data set, please pay attention to the platform in the lower right corner of the article picture to obtain it.