Insertion Devices(IDs) are one of the key components of a modern Synchrotron Light Source, the sorting and shimming of imperfect permanent magnets to enable the production of optimised IDs is a well explored problem which we look to extend here. There are currently two main computational approaches to finding a solution: Simulated Annealing and Genetic Algorithms, which are implemented in FORTRAN77 and C++ respectively. This presentation discusses the redevelopment of this code in Python, making use of more modern and better suited algorithms, such as Artificial Immune Systems, and how it compares to the Fortran code in terms of performance, maintainability and user interaction. Finally we will discuss the advantages of using the Dawn IDE for this project, specifically the variety of tools it gives the developer through to the visual and data browsing features which allow for an excellent end user experience.
YOU MAY ALSO LIKE:
- Python for Programmers (in London on 11th - 13th November 2019)
- Advanced Python (in London on 14th - 15th November 2019)
- Practical ML 2020 (in London on 2nd - 3rd July 2020)
- Knowledge Graph Applications in the Legal Profession & Fashion Industry (in London on 22nd October 2019)
- Device Tree Driver (in London on 29th October 2019)
- Hiring a Personal Investigator for Your App (SkillsCast recorded in October 2019)
- Data Transparency on Django with yourData (SkillsCast recorded in October 2019)
Optimisation of Complex Magnetic Systems at Diamond Light Source with Artificial Immune Systems
Dr Mark Basham is a Senior Software Scientist at Diamond Light Source, the UK’s national synchrotron facility located on the RAL campus in South Oxfordshire. Mark was awarded his Physics PhD in surface science simulation from Reading University, he then moved to data analysis of synchrotron data after working with experimental colleagues in the field.