München 2006 – scientific programme

Parts | Days | Selection | Search | Downloads | Help

HK: Physik der Hadronen und Kerne

HK 21: Postersitzung

HK 21.88: Poster

Tuesday, March 21, 2006, 15:30–17:00, P

Neural Net for WASA^* — •Mikhail Bashkanov, Heinz Clement, Eugene Doroshkevich, Olena Khakimova, Florian Kren, Tatiana Skorodko, and Gerhard J. Wagner — Uni. Tuebingen

Neural Networks have established as a powerful tool for solving complex tasks in nuclear and particle physics. Here we report on applications for measurements with the WASA detector regarding the calibration of the detector and its ability for particle identification.

The WASA detector is a complex device consisting of thousands of detector elements. E.g., it has 11 layers of scintillating detectors in forward direction. Each pair of these planes can be used for particle ID via dE−E plots which gives 55 possible dE−E plots for the whole forward detector. For good ID in presence of hadronic interactions in the detector material the use of a neural net is shown to be very successful. The neural net also allows to account for all possible conditions at once.

Another application is calibration. Most of the detector elements have position dependent light output. Together with nonlinearities of photomultipliers and aging effects of detector elements the calibration procedure gets quite complicated. Also here the use of neural nets is shown to be advantageous.

The implementation as hardware neural net will be discussed in particular with respect to the future WASA program.

* - supported by BMBF(06TU201) and DFG (Europ. Grad. Kolleg.)