Forecasting interlocks at HIPA

by Melissa Zacharias (PSI)

Monday 11 Jan 2021, 16:00 → 17:00 Europe/Zurich

Zoom (PSI)

The interlocks of particle accelerators, despite being necessary safety measures, lead to abrupt operational changes and a substantial loss of beam time. A novel time series classification approach was devised to decrease beam time loss in the High Intensity Proton Accelerator (HIPA) complex by forecasting interlock events. The forecasting is performed through binary classification of windows of multivariate time series. The time series are transformed into Recurrence Plots which are then classified by a Convolutional Neural Network, which not only captures the inner structure of time series but also utilities the advances of image classification techniques. The forecasting model was adapted to data from the current HIPA run and new performance evaluation methods were devised. A software package was developed and interfaced with the existing GUI for live predictions to allow for direct evaluation and continuous training of the forecasting model on new data.

For details, contact Jochem Snuverink

 

Forecasting_interlocks_at_HIPA_MZ.pdf