powercast.ai

Our numbers

6 of 7 days

better average accuracy than NESO during intraday

up to 22%

better average accuracy than NESO during intraday

7 of 7 days

better average accuracy than NESO during day-ahead

up to 33%

better accuracy than NESO during day-ahead

Our mission

at powercast.ai is to deliver state-of-the-art AI forecasting solutions which empower Analysts and Traders to make better trading decisions.

With backgrounds in data science, data engineering and the energy industry, our team came together to build powercast.ai because we recognized a shift in technology, enabling a transformation of the way power forecasting can be developed, managed and delivered.

A shift away from rigid legacy company structures, traditional regression analysis as the only tool in the toolbox, local physical hardware, software installation requirements, and cumbersome data integration processes.

Towards decentralization, exploitation of artificial intelligence, security, consistency, flexibility, and a user-focused approach.

The technological shift was primarily driven by 3 key advancements. First, improvements in compute capability; it’s simply become easier to crunch a lot of data. Second, availability of cloud-based IT architecture, making it possible to build a very flexible, robust and secure platform for structuring and managing data. And third, a proliferation of data availability, especially within the area of weather data, which can be taken advantage of through careful and deliberate data engineering and a high level of domain knowledge.

We are data and energy forecasting specialists, who strive not only to capture the complexity of the demand and supply side of electricity systems - but also to make this business-critical data easily accessible and digestible.

To achieve our mission

we apply the latest machine learning techniques to data based on deep fundamental insights and an obsession with data quality. We take advantage of cloud-enabled technology to compute, store and automate our processes, and provide a seamless user experience through an intuitive API.

State-of-the-art AI and machine learning

Data quality and fundamental insights

Cloud-based automation

Easy data integration

Our products

We currently offer 3 forecast outputs for Great Britain (GB) as standard; demand, embedded solar generation and embedded wind generation.

All forecasts express an average MW value for each settlement period and are generated at national level for the GB market.

The time horizon covers the short-term trading cycle, optimized for trading in the intraday and day-ahead trading windows.

The forecast outputs are made available through an API endpoint and can be retrieved in your preferred data format, such as csv or json.

Demand

The national demand forecast includes transmission losses, but excludes interconnector flows and demand from station transformers and pumped storage units. As such, it is a standardized data point which can be directly interpreted and compared to standardized industry and public data, such as the Initial National Demand Out-turn (INDO) and National Demand Forecast (NDF) from the National Energy System Operator (NESO).

Embedded Solar

National embedded solar generation is also considered "invisible" as it reflects generation from solar pv panels, whether domestic or commercial, connected to the distribution network. Both individual embedded generation forecasts are used as data input for the national demand forecast.

Embedded Wind

The national embedded wind generation forecast estimates electricity output generated from wind turbines connected to the distribution grid. This generation does not have transmission system metering installed and is therefore often considered "invisible". The effect of embedded generation is a suppression of demand at the level of the transmission grid.

info@powercast.ai

+44 (0) 7748646622