Digital energy

Our solutions will help you transform your digital energy capabilities. Here are some examples:

• Fault identification and predictive maintenance: increase uptime and reduce maintenance and spare parts cost for renewable energy assets
• Estimation and activation of flexibility: manage local grid congestions
• Forecasting: high temporal and spatial resolutions support local energy management, networks, and market operations
• Processing of time series: pre-processing, storage, analytics, and forecasting
• Graph machine learning: reconstruct faulty data of PV production and forecast future generation values

• Data-driven models: technical systems, such as complete buildings or wind turbines, including linear state-space models with non-linear regressors and architectures based on recurrent neural networks, such as long short-term memory (LSTM) cells, encoder-decoder architectures, and transformers
• Data collection architectures: based on a mixture of commercial solutions and custom developments, such as the highly available instrumentation of tens of radiator valves in operational buildings for state estimation and optimal control
• Customized numerical algorithms: improve the applicability, scalability, and robustness of demanding computations. Applications include the training of data-driven models and the resolution of optimization problems, such as our patent-pending solution for optimal control of PV inverters or the aggregation and dispatch of flexibility provided by buildings.