smeisen / autopv Goto Github PK

I’m currently working on methods to automate time series forecasting in smart grid applications! ⚙️📈⚡

• AutoPV: Automated photovoltaic forecasts with limited information using an ensemble of pre-trained models
• Loss-Customised Probabilistic Energy Time Series Forecasts Using Automated Hyperparameter Optimisation

• Review of automated time series forecasting pipelines
• Multi-zone grey-box thermal building identification with real occupants

• A Lightweight User Interface for Smart Charging of Electric Vehicles: A Real-World Application
• Concepts for Automated Machine Learning in Smart Grid Applications
• Smart Charging of Electric Vehicles with Cloud-based Optimization and a Lightweight User Interface: A Real-World Application in the Energy Lab 2.0: Poster

Web application for bike and motorcycle riders to organize maintenance, spare parts, and tires, furthermore to log and analyze training sessions and chassis set-ups. Link to demo! (allocating server resources may take some time during the first start of the app)

Smart charging, i.e. the controlled and coordinated charging of Electric Vehicles (EVs) seems promising for efficient electric mobility. According concepts, however, require sufficient user acceptance. For this purpose, the Smart Charging Wizard provides a lightweight User Interface (UI) that makes smart charging more transparent. In particular, the utilization of energy and time flexibilities in EV charging events to reduce both EV operating cost and battery aging is illustrated.

The goals of pyWATTS are i) to support researchers in conducting automated time series experiments independent of the execution environment and ii) to make methods developed during the research easily reusable for other researchers.

📧 [email protected]
🤓 ResearchGate
🔗 LinkedIn