002746988 001__ 2746988
002746988 005__ 20220114040809.0
002746988 0248_ $$aoai:cds.cern.ch:2746988$$pcerncds:FULLTEXT$$pcerncds:CERN:FULLTEXT$$pcerncds:CERN
002746988 0247_ $$2DOI$$9Elsevier B.V.$$a10.1016/j.cpc.2021.108204$$qpublication
002746988 037__ $$9arXiv$$aarXiv:2011.04335$$cphysics.acc-ph
002746988 035__ $$9arXiv$$aoai:arXiv.org:2011.04335
002746988 035__ $$9Inspire$$aoai:inspirehep.net:1829063$$d2022-01-13T01:25:35Z$$h2022-01-14T03:00:11Z$$mmarcxml$$ttrue$$uhttps://inspirehep.net/api/oai2d
002746988 035__ $$9Inspire$$a1829063
002746988 041__ $$aeng
002746988 100__ $$aAppleby, Robert B.$$uU. Manchester (main)$$uCockcroft Inst. Accel. Sci. Tech.$$vThe University of Manchester, Oxford Rd, Manchester M13 9PL, UK$$vThe Cockcroft Institute, Keckwick Ln, Daresbury, Warrington, WA4 4AD, UK
002746988 245__ $$9Elsevier B.V.$$aMerlin++, a flexible and feature-rich accelerator physics and particle tracking library
002746988 269__ $$c2020-11-09
002746988 260__ $$c2022-02
002746988 300__ $$a17 p
002746988 520__ $$9Elsevier B.V.$$aMerlin++ is a C++ charged-particle tracking library developed for the simulation and analysis of complex beam dynamics within high energy particle accelerators. Accurate simulation and analysis of particle dynamics is an essential part of the design of new particle accelerators, and for the optimization of existing ones. Merlin++ is a feature-full library with focus on long-term tracking studies. A user may simulate distributions of protons or electrons in either single particle or sliced macro-particle bunches. The tracking code includes both straight and curvilinear coordinate systems allowing for the simulation of either linear or circular accelerator lattice designs, and uses a fast and accurate explicit symplectic integrator. Physics processes for common design studies have been implemented, including RF cavity acceleration, synchrotron radiation damping, on-line physical aperture checks and collimation, proton scattering, wakefield simulation, and spin-tracking. Merlin++ was written using C++ object orientated design practices and has been optimized for speed using multicore processors. This article presents an account of the program, including its functionality and guidance for use. Program Title: Merlin++ CPC Library link to program files:https://doi.org/10.17632/4x4nsbhz37.1 Developer's repository link: 10.5281/zenodo.3700155 Licensing provisions: GPLv2+ Programming language: C++ Nature of problem: Complexity of particle accelerators beam dynamics over extensive tracking distances. Solution method: Long-term particle accelerator and tracking simulations utilizing explicit symplectic integrators. Additional comments including restrictions and unusual features: For further information see github.com/Merlin-Collaboration
002746988 520__ $$9arXiv$$aMerlin++ is a C++ charged-particle tracking library developed for the simulation and analysis of complex beam dynamics within high energy particle accelerators. Accurate simulation and analysis of particle dynamics is an essential part of the design of new particle accelerators, and for the optimization of existing ones. Merlin++ is a feature-full library with focus on long-term tracking studies. A user may simulate distributions of protons or electrons in either single particle or sliced macro-particle bunches. The tracking code includes both straight and curvilinear coordinate systems allowing for the simulation of either linear or circular accelerator lattice designs, and uses a fast and accurate explicit symplectic integrator. Physics processes for common design studies have been implemented, including RF cavity acceleration, synchrotron radiation damping, on-line physical aperture checks and collimation, proton scattering, wakefield simulation, and spin-tracking. Merlin++ was written using C++ object orientated design practices and has been optimized for speed using multicore processors. This article presents an account of the program, including its functionality and guidance for use.
002746988 540__ $$3preprint$$aarXiv nonexclusive-distrib 1.0$$uhttp://arxiv.org/licenses/nonexclusive-distrib/1.0/
002746988 542__ $$3publication$$dElsevier B.V.$$g2021
002746988 65017 $$2arXiv$$aphysics.acc-ph
002746988 65017 $$2SzGeCERN$$aAccelerators and Storage Rings
002746988 690C_ $$aCERN
002746988 690C_ $$aARTICLE
002746988 700__ $$aBarlow, Roger [email protected]$$uHuddersfield U.$$vThe University of Huddersfield, Queensgate, Huddersfield, HD1 3DH, UK
002746988 700__ $$aKruecker, Dirk$$uDESY$$vDESY, Notkestraße 85, D22607 Hamburg, Germany
002746988 700__ $$aMolson, James$$uCERN$$vCERN, CH1211 Genève 23, Switzerland
002746988 700__ $$aRowan, Scott$$uHuddersfield U.$$vThe University of Huddersfield, Queensgate, Huddersfield, HD1 3DH, UK
002746988 700__ $$aTygier, Sam$$uU. Manchester (main)$$vThe University of Manchester, Oxford Rd, Manchester M13 9PL, UK
002746988 700__ $$aRafique, Haroon$$uCERN$$vCERN, CH1211 Genève 23, Switzerland
002746988 700__ $$aWalker, Nicholas$$uDESY$$vDESY, Notkestraße 85, D22607 Hamburg, Germany
002746988 700__ $$aWolski, Andrzej$$uU. Liverpool (main)$$uCockcroft Inst. Accel. Sci. Tech.$$vThe University of Liverpool, Liverpool L69 7ZX, UK$$vThe Cockcroft Institute, Keckwick Ln, Daresbury, Warrington, WA4 4AD, UK
002746988 773__ $$c108204$$mpublication$$pComput. Phys. Commun.$$v271$$y2022
002746988 8564_ $$82268826$$s44886$$uhttp://cds.cern.ch/record/2746988/files/match_halo_x.png$$y00004 Bunch distribution profile in the $x$-$x'$ plane. Initially the bunch is created in normalized coordinates (top), then transformed by Courant-Snyder parameters (middle). A distribution from the halo option, set at one sigma, is also shown (bottom).
002746988 8564_ $$82268827$$s10656$$uhttp://cds.cern.ch/record/2746988/files/user_example_black.png$$y00015 A simplified example of a user program. The program is standard C++ (a). First an accelerator lattice is constructed (b) which can be parametrized (c). Then the beam parameters can be defined and a particle bunch constructed (d). The initial bunch is written to disk (e), then tracked though the lattice (f) and the resulting particle coordinates written to disk (g).
002746988 8564_ $$82268828$$s16812$$uhttp://cds.cern.ch/record/2746988/files/IR7.png$$y00000 Scattered particles in the collimation region of the LHC. Upper plot shows the arrangement of dipoles (blue), quadrupoles (green) and collimators (black). Lower plot shows the machine aperture (black) and proton tracks (blue).
002746988 8564_ $$82268829$$s75069$$uhttp://cds.cern.ch/record/2746988/files/architecture_v2.png$$y00001 High-level architectural view of Merlin++ lattice construction and particle bunch tracking.
002746988 8564_ $$82268830$$s36154$$uhttp://cds.cern.ch/record/2746988/files/third.png$$y00008 The effect on the horizontal parameters of adding a constant length/field sextupole to the end of a periodic FODO lattice cell. Shown are half, third, quarter and fifth integer phase advances $\mu_x$.
002746988 8564_ $$82268831$$s7447$$uhttp://cds.cern.ch/record/2746988/files/multi_proc_benchmark.png$$y00013 Performance of a benchmark test case with number of processes on a 16-core 32-thread Xeon processor. Dashed line shows ideal scaling.
002746988 8564_ $$82268832$$s2484395$$uhttp://cds.cern.ch/record/2746988/files/2011.04335.pdf$$yFulltext
002746988 8564_ $$82268833$$s947811$$uhttp://cds.cern.ch/record/2746988/files/lattice_modification.png$$y00016 An example of how modifying specific lattice element position and field strengths affect the beta function. Iterative modification can be implemented to achieve automated design of stable lattices.
002746988 8564_ $$82268834$$s58669$$uhttp://cds.cern.ch/record/2746988/files/collective_parallel.png$$y00014 Graphical depiction of how Merlin++ allows the combination of both distributed single particle tracking (MPI) and calculation concurrency (OpenMP) to maximise concurrency potential in taking into account collective effects.
002746988 8564_ $$82268835$$s179024$$uhttp://cds.cern.ch/record/2746988/files/norm_norm_x.png$$y00002 Bunch distribution profile in the $x$-$x'$ plane. Initially the bunch is created in normalized coordinates (top), then transformed by Courant-Snyder parameters (middle). A distribution from the halo option, set at one sigma, is also shown (bottom).
002746988 8564_ $$82268836$$s94876$$uhttp://cds.cern.ch/record/2746988/files/doxygen.png$$y00009 Example of a class inheritance structure within Merlin++, in this case the {\tt Aperture} class. The graphic is automatically produced by doxygen to be included in the documentation.
002746988 8564_ $$82268837$$s67541$$uhttp://cds.cern.ch/record/2746988/files/match_norm_x.png$$y00003 Bunch distribution profile in the $x$-$x'$ plane. Initially the bunch is created in normalized coordinates (top), then transformed by Courant-Snyder parameters (middle). A distribution from the halo option, set at one sigma, is also shown (bottom).
002746988 8564_ $$82268838$$s73641$$uhttp://cds.cern.ch/record/2746988/files/half.png$$y00005 The effect on the horizontal parameters of adding a constant length/field sextupole to the end of a periodic FODO lattice cell. Shown are half, third, quarter and fifth integer phase advances $\mu_x$.
002746988 8564_ $$82268839$$s22090$$uhttp://cds.cern.ch/record/2746988/files/compare_optics_ATLAS_IR1_betas.png$$y00010 Comparison of $\beta$ functions around the ATLAS interaction region in the LHC. The lower plot shows difference between MADX and Merlin++'s output in parts per million.
002746988 8564_ $$82268840$$s80041$$uhttp://cds.cern.ch/record/2746988/files/quarter.png$$y00007 The effect on the horizontal parameters of adding a constant length/field sextupole to the end of a periodic FODO lattice cell. Shown are half, third, quarter and fifth integer phase advances $\mu_x$.
002746988 8564_ $$82268841$$s20836$$uhttp://cds.cern.ch/record/2746988/files/lossmap_2016_Ring_B1H.png$$y00012 Beam 1 horizontal loss map from BLMs (top) and Merlin++ (bottom) at 6.5 TeV with $\beta$* of 50 cm.~\cite{tygier_performance_2019}
002746988 8564_ $$82268842$$s59523$$uhttp://cds.cern.ch/record/2746988/files/fifth.png$$y00006 The effect on the horizontal parameters of adding a constant length/field sextupole to the end of a periodic FODO lattice cell. Shown are half, third, quarter and fifth integer phase advances $\mu_x$.
002746988 8564_ $$82268843$$s26403$$uhttp://cds.cern.ch/record/2746988/files/CuCD_30cm_x.png$$y00011 Comparison of fixed target horizontal scattering results between Merlin++ and Geant4. Geant4 physics lists used: QGSP\_BERT and FTFP\_BERT. Material used: copper-diamond (CuCD). Material depth: 30~cm.
002746988 960__ $$a13
002746988 980__ $$aARTICLE