| Paper |
Title |
Page |
| WEPCH012 |
Comparison of Betatron Function Measurement Methods and Consideration of Hysteresis Effects
|
1945 |
| |
- O. Kopitetzki, D. Schirmer, G. Schmidt, K. Wille
DELTA, Dortmund
|
|
| |
Two methods for determining the betatron functions in a storage ring were used to survey the linear optics at Delta. The fast orbit response analysis is used to gain betatron functions at the beam position monitors (BPMs) and dipole correctors. These are compared to betatron functions measured by the tune scan method which gives the beta functions in the quadrupoles. To improve the accuracy of the betatron functions obtained by the tune scan method a measuring procedure is introduced which considers the hysteresis effects in the quadrupole magnets. Systematic deviations in the beta functions measured between the two methods have been observed. The calibration errors of the BPMs can explain the observed deviations. With the orbit response analysis also the betatron phase advances between the measurement points can be calculated. Because these do not depend on the calibration errors, unlike the betatron functions, the differences between measurement and model can be determined more precise. A comparison of both methods with the optics model will be presented.
|
|
| WEPCH013 |
Electron Transport Line Optimization using Neural Networks and Genetic Algorithms
|
1948 |
| |
- D. Schirmer, T. Buening, P. Hartmann, D. Mueller
DELTA, Dortmund
|
|
| |
Methods of computational intelligence (CI) were investigated to support the optimization of the electron transfer efficiency from the booster synchrotron BoDo to the electron storage ring DELTA. Neural networks and genetic algorithms were analysed alternatively. At first both types of methods were trained on the basis of a theoretical model of the transport line. After the training various algorithms were used to improve the magnet settings of the real transport line elements with respect to the electron transfer efficiency. The results of different strategies are compared and prospects as well as limitations of CI-methods to the application of typical optimization problems in accelerator operation are discussed.
|
|