Brief annotation and scientific rationale:

The project is aimed at organizing and providing computational support for physics research programmes implemented with JINR’s participation, at developing mathematical methods and software for modelling physical processes and experimental facilities, processing and analysing experimental data in the field of elementary particle physics, nuclear physics, neutrino physics, condensed matter, radiobiology, etc. The particular attention will be paid to the creation of systems for the distributed processing and analysis of experimental data, as well as information and computing platforms to support research at JINR and other world centres.

The main areas of work are mathematical and computational physics to support JINR’s large research infrastructure projects, first of all, the experiments at the NICA accelerator complex and the Baikal-GVD neutrino telescope. Further cooperation with experiments at the largest world accelerator centres (CERN, BNL, etc.), experiments in the field of neutrino physics and astrophysics, radiobiological research programmes will also be continued. The possibility of using the developed methods and algorithms within other megascience projects is being considered.

Expected results upon completion of the project:

1. Revision of interaction generators and their development for modelling the processes of interactions of light and heavy nuclei, including those at NICA energies (FTF, QGSM, DCM-QGSM-SMM, etc.), and processes beyond the Standard Model, such as the production of candidate particles for the role of dark matter, additional Higgs bosons and processes that violate the lepton number, etc. (QBH, Pythia, MadGraph, etc.) for LHC conditions at a nominal energy and a total integrated luminosity up to 450 fb^-1.
2. Development of algorithms for the reconstruction of charged particle tracks for experimental facilities, including those at NICA and the LHC, creation of appropriate software and its application for data processing and analysis, the study of the physical and technical characteristics of detector systems.
3. Development of scalable algorithms and software for processing multi-parameter, multi-dimensional, hierarchical data sets of exabyte volume, including those based on recurrent and convolutional neural networks, for machine and deep learning tasks, designed primarily for solving various problems in particle physics experiments, including for the NICA megaproject and neutrino experiments.
4. Creation and development of data processing and analysis systems and modern research tools for international collaborations (NICA, JINR neutrino programme, experiments at the LHC).
5. Development of algorithms and software for JINR’s research projects in the field of neutron physics.
6. Development of algorithms, software and computing platforms for radiobiological research, applied research in the field of proton therapy and ecology.

Expected results of the project current year: 

− revision of FTF and QGSM models and development of software modules for modelling nuclear interactions of charmed hadrons, light hyper-nuclei;

− development of the DCM-QGSM-SMM generator: considering the dependence of the lifetime of resonances on the density of the nuclear medium, the suppression of the production cross section of pseudoscalar mesons, and the enhancement of the production of hyperons in a dense nuclear medium, including the deformation of nuclei;

− development of software for simulating the events indicated in the previous paragraph, taking into account the performance of the NICA SPD facility;

− evaluation of cross sections and modelling of the processes of production of dark matter particles within extended two-doublet Higgs models (MadGraph generator);

− debugging the procedure for testing sensitive elements of the high-granularity calorimeter of the CMS experiment, including track reconstruction and the evaluation of the efficiency of each detector cell;

− development and adjustment of algorithms and methods for reconstructing muon trajectories in the Cathode-Strip Chambers (CSCs) of the muon system of the CMS experiment for the comparison of the continuous and discrete approaches of wavelet analysis for separating overlapping signals; estimation of the CSC spatial resolution and the aging effect on data obtained in 2024 at the GIF++ facility at CERN and in proton-proton beam collisions at the LHC;

− optimization of algorithms for local track reconstruction in the DCH and CSC detectors of the BM@N experiment, their fitting with scintillation detectors for global reconstruction and particle identification, detector alignment and estimation of their operation parameters with experimental data of 2022-2023.

− finding and checking correction parameters for the CSC and GEM detectors of the BM@N experiment, development and implementation of software for modelling and data processing methods, as well as their development and adaptation for the current configurations of a number of GEM and Silicon Profilometer tracking detectors in 2023-2024;

− study of the effectiveness of the application of machine learning methods based on decision trees for the particle identification task in the MPD experiment;

− optimization of the software platform of the MPD experiment: development and implementation in MPDRoot on the basic rules of OOP, unified tests of algorithms and interaction of classes, etc.;

− development and training of a neural network for searching and restoring tracks in the vertex detector and tracker of the SPD facility, for restoring clusters in the electromagnetic calorimeter and in the SPD muon system;

− development of a data processing and storage model: specification of data types and formats, estimation of computational costs for processing at each stage of data transformation, formulation of technical requirements for a real-time data selection system, a distributed data processing and storage system, and offline processing software;

− creation of a prototype of a system that provides multi-stage data processing on a real-time event filtering cluster, SPD OnLine Filter;

− creation of prototypes of the SPD task management system based on the PanDA package and the data management system based on the RUCIO DDM package;

− development of a prototype system for data processing for the Baikal telescope;

− development of a test package for the primary processing of small-angle experimental data from the YuMO spectrometer for a multi-detector system with a position-sensitive detector;

− development of a C++ library for converting online condition data into JSON, implementation of DCS data conversion to CREST. Modification of Athena package algorithms using COOL for CREST. Development and maintenance of the operation of information systems for the BM@N and MPD experiments to describe the geometry of facilities, configuration of detectors, management process;

− study of the background from cosmic protons for the TAIGA observatory, estimation of the number of evaporation neutrons and investigation of their interaction in the OLVE-HERO detector;

− analysis of test data from the prototype of a digital calorimeter for proton therapy, development of an algorithm based on a cellular automaton for track recognition and reconstruction;

− application of piecewise polynomial approximation based on the high-order basis element method for processing and analysing neutron noise from the IBR-2M reactor;

− development of a behavioral analysis module that will automate the analysis of video data obtained during the testing of laboratory animals in various test systems;

− application of algorithms for the automatic selection of optimal data augmentation policies, testing of various loss minimization functions, determination of the most effective methods for classifying images with plant diseases;

− improving the existing functionality and providing new opportunities for monitoring and predicting the state of the environment. Automation of the monitoring process using simulation.

Brief annotation and scientific rationale:

The project is aimed at the development and application of mathematical and computational methods for modelling complex physical systems studied within the JINR Topical Plan and described by systems of dynamic nonlinear, spatially multidimensional integral, integro-differential or differential equations that depend on the parameters of models. The evolution of solutions to such systems can be characterized by the occurrence of critical modes, bifurcations and phase transitions. Mathematical modelling is an inseparable part of modern scientific research. It entails an adequate mathematical formulation of problems within the models under study, the adaptation of known numerical approaches or elaboration of new ones to effectively take into account the features of the studied physical processes, the development of algorithms and software packages for high-performance simulation on modern computer systems, including the resources of the JINR Multifunctional Information and Computing Complex.  

Expected results upon completion of the project:

1. Development of methods, algorithms and software packages for conducting the numerical research of interactions of various types in complex systems of nuclear physics and quantum mechanics.
2. Methods for modelling multifactorial processes in materials and condensed matter under external actions.
3. Methods for solving simulation tasks in the design of experimental facilities and the optimization of their operating modes.
4. Methods for modelling complex processes in dense nuclear matter based on the equation of state.
5. Methods for modelling quantum systems using quantum information theory methods and hybrid quantum-classical programming methods.

Expected results of the project current year: 

 – Development of a mathematical formulation of the problem within the strong coupling channels method with the Woods–Saxon optical potential and regular boundary conditions for modelling sub-barrier heavy ion fusion and fission reactions.

– Development of methods and calculation of the energy of adsorption on the Au layer of heavy and superheavy atoms.

– Development and optimization of the method of self-similar approximations for solving nonlinear equations that do not contain small parameters and describe quantum mechanical systems, including spin ensembles and cold atoms in traps.

– Development of a method and a programme that initiates, within the transport-statistical approach, the initial state of colliding nuclei with nuclear potentials, which are used for further calculations in collision dynamics.

– Modelling of proton-nucleus interactions, based on a microscopic model of the optical potential, over a wide range of energies and for a large variety of atomic numbers of target nuclei with the aim at assessing the influence of the nuclear medium on the processes of proton scattering by intranuclear nucleons.

– Investigation of the dynamics of a shock wave in an irradiated material based on a model described by the combination of molecular dynamic equations, thermal conductivity equations and wave equations. Determination of the parameters of the wave equation based on the results of the numerical solution of molecular dynamics equations.

– Simulation of the interaction of amyloid beta and antimicrobial peptides with phospholipid membranes in vesicular and bicellar structures in the coarse-grained model; study of the dynamic properties of this interaction based on the calculation of the phonon spectra of systems; construction of the free energy profile of the process of pulling the peptide out of the membrane depending on the distance between the centres of mass and the conformation of the peptide (replica exchange umbrella sampling).

– Study of localized structures in systems described by nonlinear damped-driven equations. Investigation of the formation of a hydrated electron based on a modified polaron model that takes into account the Coulomb interaction, calculation of the observed characteristics of this process.

– Adaptation of the COMSOL Multiphysics® package to the HybriLIT heterogeneous platform in order to enhance the efficiency of computations and reduce the computational time through the use of a mixed vector-scalar formulation of magnetostatics and a hybrid finite and boundary element method. Development and software implementation of difference schemes for solving a boundary value problem for a 4^th order equation describing the distribution of physical fields in 2D and 3D regions of various configurations.

– Development of methods and study of the formation of magnetic fields of isochronous cyclotrons under various operating modes. Preparation of instructions and registration of the CORD (Closed ORbit Dynamics) programme in the JINRLIB library. CORD implements calculations to study the effect of betatron oscillations and the phase motion of beam particles on the magnetic field of the MSC230 cyclotron.

– Adaptation of the neural network approach to the approximate calculation of multiple integrals arising in the study of pion survival in heavy ion collisions; elaboration of methods aimed at extending to finite temperatures the previously developed model of the quark-hadron phase transition in cold nuclear matter.

– Modelling and calculation of cosmological redshift values based on the equation of state; investigation of the possibility of reconstructing the mass spectrum of an isolated neutron star from the data on the age and surface temperature of pulsars, based on simulations of their temperature evolution; simulation of the processes of scattering and production of particles in dense and hot nuclear matter.

– Development of an evolution operator trotterization algorithm for von Neumann and Lindblad equations and implementation of the corresponding quantum circuit on a quantum simulator in the QISKit environment. Improving the performance of the quantum circuit simulator by increasing the simulation speed using multiprocessing.

– Creation of a package of modules designed to decompose a quantum system into subsystems based on the use of the tensor products of representations of wreath products of finite cyclic groups.

– Determination of the relationship between the characteristics of the entanglement of composite quantum systems and the negativity of Wigner quasiprobability distributions. Development of a functional reduction method for two-loop Feynman integrals and its application to the calculation of integrals corresponding to diagrams with four and five external lines.