Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userApplication of machine learning techniques at BESIII experiment
54
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
BESIII is a currently running tau-charm factory with the largest samples of on threshold charm meson pairs, directly produced charmonia and some other unique datasets at BEPCII collider. Machine learning techniques have been employed to improve the performance of BESIII software. The studies for reweighing MC, particle identification and cluster reconstruction for the CGEM (Cylindrical Gas Electron Multiplier) inner tracker are presented.
rate research
Read More
The application of machine learning techniques to the reconstruction of lepton energies in water Cherenkov detectors is discussed and illustrated for TITUS, a proposed intermediate detector for the Hyper-Kamiokande experiment. It is found that applying these techniques leads to an improvement of more than 50% in the energy resolution for all lepton energies compared to an approach based upon lookup tables. Machine learning techniques can be easily applied to different detector configurations and the results are comparable to likelihood-function based techniques that are currently used.
The Beijing Electron Spectrometer III (BESIII) is a multipurpose detector that collects data provided by the collision in the Beijing Electron Positron Collider II (BEPCII), hosted at the Institute of High Energy Physics of Beijing. Since the beginning of its operation, BESIII has collected the world largest sample of J/{psi} and {psi}(2s). Due to the increase of the luminosity up to its nominal value of 10^33 cm-2 s-1 and aging effect, the MDC decreases its efficiency in the first layers up to 35% with respect to the value in 2014. Since BESIII has to take data up to 2022 with the chance to continue up to 2027, the Italian collaboration proposed to replace the inner part of the MDC with three independent layers of Cylindrical triple-GEM (CGEM). The CGEM-IT project will deploy several new features and innovation with respect the other current GEM based detector: the {mu}TPC and analog readout, with time and charge measurements will allow to reach the 130 {mu}m spatial resolution in 1 T magnetic field requested by the BESIII collaboration. In this proceeding, an update of the status of the project will be presented, with a particular focus on the results with planar and cylindrical prototypes with test beams data. These results are beyond the state of the art for GEM technology in magnetic field.
Gas detector are very light instrument used in high energy physics to measure the particle properties: position and momentum. Through high electric field is possible to use the Gas Electron Multiplier (GEM) technology to detect the charged particles
and to exploit their properties to construct a large area detector, such as the new IT for BESIII. The state of the art in the GEM production allows to create very large area GEM foils (up to 50x100 $mathrm{cm}^2$) and thanks to the small thickness of these foils is it possible to shape it to the desired form: a Cylindrical Gas Electron Multiplier (CGEM) is then proposed. The innovative construction technique based on Rohacell, a PMI foam, will give solidity to cathode and anode with a very low impact on material budget. The entire detector is sustained by Permaglass rings glued at the edges. These rings are used to assembly the CGEM, together with a dedicated Vertical Insertion System and moreover they host the On-Detector electronic. The anode has been improved w.r.t. the state of the art through a jagged readout that minimize the inter-strip capacitance. The mechanical challenge of this detector requires a precision of the entire geometry within few hundreds of microns in the whole area. In this contribution an overview of the construction technique, the validation of this technique through the realization of a CGEM, and its first tests will be presented. These activities are performed within the framework of the BESIIICGEM Project (645664), funded by the European Commission in the action H2020-RISE-MSCA-2014.
We apply deep learning methods as a track finding algorithm to the PANDA Forward Tracking Stations (FTS). The problem is divided into three steps: The first step relies on an Artificial Neural Network (ANN) that is trained as a binary classifier to build track segments in three different parts of the FTS, namely FT1,FT2, FT3,FT4, and FT5,FT6. The ANN accepts hit pairs as an input and outputs a probability that they are on the same track or not. The second step builds 3D track segments from the 2D ones and is based on the geometry of the detector. The last step is to match the track segments from the different parts of the FTS to form a full track candidate, and is based on a Recurrent Neural Network (RNN). The RNN is used also as a binary classifier that outputs the probability that the combined track segments are a true track or not. The performance of the algorithm is judged based on the purity, efficiency and the ghost ratio of the reconstructed tracks. The purity specifies which fraction of hits in one track come from the correct particle. The correct particle is the particle, which produces the majority of hits in the track. The efficiency is defined as the ratio of the number of correctly reconstructed tracks to all generated tracks.
A study on the use of a machine learning algorithm for the level 1 trigger decision in the JUNO experiment ispresented. JUNO is a medium baseline neutrino experiment in construction in China, with the main goal of determining the neutrino mass hierarchy. A large liquid scintillator (LS)volume will detect the electron antineutrinos issued from nuclear reactors. The LS detector is instrumented by around 20000 large photomultiplier tubes. The hit information from each PMT will be collected into a center trigger unit for the level 1 trigger decision. The current trigger algorithm used to select a neutrino signal event is based on a fast vertex reconstruction. We propose to study an alternative level 1 (L1) trigger in order to achieve a similar performance as the vertex fitting trigger but with less logic resources by using firmware implemented machine learning model at the L1 trigger level. We treat the trigger decision as a classification problem and train a Multi-Layer Perceptron (MLP)model to distinguish the signal events with an energy higher than a certain threshold from noise events. We use JUNO software to generate datasets which include 100K physics events with noise and 100K pure noise events coming from PMT dark noise.For events with energy higher than 100 keV, the L1 trigger based on the converged MLP model can achieve an efficiency higher than 99%. After the training performed on simulations,we successfully implemented the trained model into a Kintex 7FPGA. We present the technical details of the neural network development and training, as well as its implementation in the hardware with the FPGA programming. Finally the performance of the L1 trigger MLP implementation is discussed.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
