Paper  Title  Page 

TUAAI2  A Massively Parallel General Purpose Multiobjective Optimization Framework, Applied to Beam Dynamic Studies  62 


Particle accelerators are invaluable tools for research in basic and applied sciences. The successful design, commissioning, and operation of accelerator facilities is non trivial. We implemented a framework for general simulationbased multioptimization methods automating the investigation of optimal sets of machine parameters. In order to solve the emerging, huge problems we propose a massivelyparallel master/slave approach. We employ the framework to identify optimal parameters of existing and new accelerators at PSI.  
Slides TUAAI2 [0.694 MB]  
TUABI1  The TRIUMF Optimization Platform and Application to the Elinac Injector  67 


Funding: Funding is received from Natural Sciences and Engineering Research Council of Canada for this research. Multiobjective genetic algorithms (MOGA) have demonstrated their usefulness for the global optimization of accelerator design using Elegant and Astra. A MOGA platform developed at TRIUMF seeks to expand the capabilities of such tools by allowing multiple simulation engines to be used. The TRIUMF optimization software platform was applied to the transport design of an injection line leading from a cryomodule to the beam dump. The optimization involves two simulation engines, Astra and MADX, and demonstrates the ability for the platform to handle multiengine optimization for a realistic problem. Results of the optimization are shown. 

Slides TUABI1 [1.132 MB]  
TUABC2  Global Optimization of the ANKA Lattice Using Multiobjective Genetic Algorithms (MOGA)  72 


Funding: This work has been supported by the Initiative and Networking Fund of the Helmholtz Association under contract number VHNG320. The optimization of a storage ring lattice is a multiobjective problem, since the parameter space of possible solutions can be very large and a high number of constraints have to be taken into account during the optimization process. In this paper we used Genetic Algorithms (GA) and MultiObjective Genetic Algorithms (MOGA), which can solve such problems very efficiently and rapidly, to find the optimized settings for the ANKA storage ring lattice. 

TUABC3  MultiObjective Genetic Optimization of Linac Beam Parameters for a seeded FEL  75 


The optimization of the beam dynamics in a Linac for free electron lasers (FELs) can be a very time consuming process, in which several parameters of the acceleration and compression sections need to be varied simultaneously. The optimization procedure is required to tackle different and often opposite goals at a time, depending on the adopted FEL scheme. As such, multiobjective genetic algorithms are an interesting choice, given their ability to targeting several, often conflicting objectives. We have studied an optimization strategy based on a combination of multiobjective optimization with a fast parallel computation of the FEL performance and, for the specific case of the proposed UK’s New Light Source, we illustrate the benefits of this method for the optimization of the average gain length and its variation along the beam pulse. The method can be extended to other sets of objectives, such as power and bandwidth of the FEL.  
Slides TUABC3 [5.567 MB]  
WEP13  ModelBased Analysis of Digital Signal Processing Blocks in a Beam Phase Control System  164 


Funding: This work is supported by the German Federal Ministry of Education and Research under grant number 06DA9028I. A beam phase control system comprises digital phase detectors and band pass filters to detect coherent longitudinal dipole and higher order bunch oscillations. These digital signal processing functions can be implemented in several ways, e. g. in software or on a programmable logic device. In this paper, we consider different possible implementations and compare them in terms of their realtime performance and their system resource consumption. For the phase detectors, a software implementation is compared against different (e. g. lookup table and CORDICbased) hardware implementations. For the band pass filters, we consider software, hardware and mixed implementations. 

WEP14  Design of SRF Cavities with Cell Profiles Based on Bezier Splines  167 


Funding: This work is funded by BMBF under contract 05K10PEA. Elliptical cavities have been a standard in SRF linac technology for 30 years. In this work, we present a novel approach [1] using Bezier spline profile curves. By using different degrees of spline curves, the number of free parameters can be varied to suit a given problem (endcell tuning, basecell figures of merit), thus leading to a high flexibility of the spline approach. As a realistic example, a cubic spline SRF multicell cavity geometry is calculated and the figures of merit are optimized for the operational mode. We also present an outline for HOM endcell optimization that can be realized using available 2D solvers. [1] B. Riemann et al., "SRF multicell cavity design using cubic and higher order spline cavity profiles", T 80.9, Verhandlungen DPG Göttingen 2012 

WEP15  Tools for Analysis and Improvement of Linac Optics Design for High Brightness Electron Beams  170 


The optics design of single pass high brightness electron linacs usually aims at the preservation of the transverse emittance. Collective effects mainly impose constraints to the optics design such as at the lowbeta interaction points in colliders and magnetic compressors in FELs. Other constraints are from the trajectory correction scheme, performance of diagnostics, collimation systems and physical space limitations. Strong focusing is typically prescribed for all the aforementioned cases, although it may hamper the main goal of emittance preservation through the excitation of optical aberrations. Strong focusing also potentially leads, through focusing errors, to large beam optics mismatch. Based on these sometimes conflicting requirements, we have developed tools for the analysis and improvement of electron linac optics. They are based on the Elegant [1] code and allow the user to identify:
[1] M. Borland, Advanced Photon Source LS287 (2000). 

THP06  An OpenMP Parallelisation of Realtime Processing of CERN LHC Beam Position Monitor Data  230 


SUSSIX is a FORTRAN program for the post processing of turnbyturn Beam Position Monitor (BPM) data, which computes the frequency, amplitude, and phase of tunes and resonant lines to a high degree of precision. For analysis of LHC BPM data a specific version run through a C steering code has been implemented in the CERN Control Centre to run on a server under the Linux operating system but became a real time computational bottleneck preventing truly online study of the BPM data. Timing studies showed that the independent processing of each BPMs data was a candidate for parallelization and the OpenMP package with its simple insertion of compiler directives was tried. It proved to be easy to learn and use, problem free and efficient in this case reaching a factor of ten reduction in realtime over twelve cores on a dedicated server. This paper reviews the problem, shows the critical code fragments with their OpenMP directives and the results obtained.  
THACC3  Preliminary Study of Single Spike SASE FEL Operation at 0.26 Nanometers Wavelength for the European XFEL  253 


The production of ultrashort (fs or subfs long), high power, radiation pulses in the Xray spectral region, showing a single spike spectrum, represents a challenge for many existent SASE FELs [1, 2]. In order to realize single spike operation the length of the electron bunch after compression must be extremely small [3] (less than a micrometer) and the consequent degradation of its emittance has not to suppress the radiation production. Several technical restrictions, such as limits of diagnostics for small charges, RF jitter and microbunching instabilities play an important role in the choice of the operation working point. In this paper we are going to study the feasibility of single spike or few spikes lasing for bunches with charge of tens of pC in the European XFEL facility giving some preliminary results concerning the choice of the working point.
[1] J.B. Rosenzweig et al., NIM A 593 (2008) 3944 [2] S. Reiche et al., NIM A 593 (2008) 4548 [3] R. Bonifacio et al., PRL vol. 73 n.1 (1994) 

Slides THACC3 [1.401 MB]  
THSCC2  Reconstruction of Velocity Field  256 


In this paper we suppose that the distribution density of particles in phase space is known. Using Liouville’s equations the problem of finding velocity field is considered as a minimization problem. Thus the problem of determination of velocity field is reduced to solving of elliptic system of EulerLagrange equations.  
Slides THSCC2 [8.701 MB]  
THSCC3  On Accelerator Driven Subcritical Reactor Power Gain  259 


The accelerator driven system (ADS) with subcritical reactor is considered. Such systems demonstrate high safety, due to the fact, that the reactor operates at subcritical level. The problem of the reactor power rate maximiztion on fixed values of effective multiplication factor and the external neutron supply (neutron generating target) intensity is studied. In this paper the main attention is paid to the reactor core optimization. Some ways of ADS power rate gain and optimized reactor core parameters are proposed.  
Slides THSCC3 [1.857 MB]  
FRAAC4  Astra based Swarm Optimizations of the BERLinPro Injector  281 


The Berlin Energy Recovery Linac Project BERLinPro is a compact ERL to develop the accelerator physics and technology required to generate and accelerate a 100mA, 1mm mrad normalized emittance beam. One of the project challenges is to generate a beam of this kind in the injector part of the machine. Extensive injector optimization studies have been done over the last years. A deep insight in the physics of high brilliance, low energy beams together with single parameter scans allowed for an efficient optimization, resulting in a layout, capable to deliver bunches of the needed charge and dimension. However, changes in the injector components' technical layout, as they are unavoidable in the current stage of the project, may require reoptimizations at any time, if necessary of the whole injector part. To support these work an ASTRA based 'swarm optimization' tool for massive parallel calculations on the institutes Linux computing cluster has been developed. Once the optimization wrapper code is written, results come for free and can help to extend the understanding of the underlying physics. Strategy, procedure and results of the 'swarm optimizations' will be presented in this paper.  
Slides FRAAC4 [7.286 MB]  