AI4CZC platform
AI4CZC platform is the Inceptive platform for energy forecasting.
About AI4CZC platform
AI4CZC platform is the Inceptive platform for energy based on Igloo. It was developed in collaboration with CGES, Montenegro TSO, during the AI4CZC project.
This page explains the use case around which the platform was built (this section) and the potential further uses are detailed in the section "Beyond Cross Zonal Capacity".
Development use case
In high voltage electricity networks, the Cross Zonal Capacity (CZC) is the amount of transmission capacity. That is the maximum energy that is allowed to be transferred between two bidding zones. Its management is a critical aspect of the European TSOs' efforts to harmonize electricity balancing services, ensuring security of supply and fostering efficiency in the energy market. However, the lack of precise CZC estimations leads to suboptimal allocation and utilization of available cross-zonal capacity.
To address this challenge, we built AI4CZC platform, a web application that ships a machine learning based model to provide the net position, a key element for a precise estimation of CZC, 48h in advance.
More details on the model built during the project, can be found here.
Architecture & data sources
AI4CZC platform is a web application that ships a machine learning based model to provide the net position, a key element for a precise estimation of CZC, 48h in advance. The platform automatically retrieves pertinent data from its sources and performs net position predictions. These predictions are accessible through a GUI and a REST API.
At the moment, the platform can recover, store, process, and forecast data from ENTSOE platform. But it's flexibility allows it to aggregate many more data sources.
Overview video
AI4CZC platform is build on top of an easy-to-use Angular Graphical User Interface. Take a look at our explanatory video of the interface here.
This platform is part of a project that has received funding from the European Union’s Horizon 2020 research and innovation program under I-NERGY grant agreement No 101016508.
Beyond Cross Zonal Capacity
Platform use cases
AI4ZCZ platform is a flexible tool that can be adapted to wide range of use cases and situations. As the platform is structured by users and projects, it is possible to have many models, dashboards and API endpoints for a wide range of use cases in the domain of energy. This includes use cases around energy generation, load, transfer, prices, and carbon emissions to name a few. The granularity of the predictions, starting from hour basis is also flexible.
As ENTSOE transparency platform is already connected to the platform, retrieving energy data from Europe in real time is an easy task. But from our experience, accurate energy predictions rely on multiple data sources. That's why our platform is designed with flexibility to be able to be integrated to many sources.
How we work
We believe that there is already plenty of tools and platforms that allow data scientists to build their own platforms. But not all the actors on the energy sector have this knowledge. This platform is designed to help us support these companies. In AI4CZC we build and deploy models for you, help you to customize your dashboards and provide a comprehensive API to connect the platform to your Information Systems. Or just upload the data manually to start using the model.
GUI Functionalities overview
- Secure interface behind a login screen
- Organization in Projects
- Projects are containers for models, that can have different versions.
- Explore model inputs
- Dashboards associated to a project, showing forecast and expected values, along error metrics.
- Possibility to download a sample of the stored data.
- Possibility to upload a CSV dataset to make predictions
- An alarms system based on model metrics.
- Reset your API key used to make predictions with the REST API
You can see what functionalities look like in this video
API overview
All the functionalities described here can be performed with a REST API based on the exchange of JSON files. Here is an overview of these functionalities:
- Retrieve the projects of an user
- Retrieve the models of a project
- Generate and regenerate the api key
- Get the alarms associated to an user
- Create and delete alarms
- List the needed descriptors for a model
- Make predictions uploading a csv file
- Download a sample of data
- Make predictions using the main model of a project
- Feedback a prediction
Infrastructure and deployment
AI4CZC is designed to be both a Saas platform or be delivered as On Premise software with an on-board model.
Work with us
Interested in the platform? Reach us at contact@inceptive.tech to find out the plan and terms that best suit your problem.