TWT magnetic focusing structure optimization by parallel evolutionary algorithm

• Salvatore Coco ^[1] ; Antonino Laudani ^[2] ; Giuseppe Pollicino ^[1] ; Giuseppe Pulcini ^[2] ; Francesco Riganti Fulginei ^[2] ; Alessandro Salvin ^[2]
  1. [1] University of Catania

     University of Catania

     Catania, Italia

     
  2. [2] University of RomaTre
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 31, Nº 5 (Special Issue: Electromagnetic Fields in Electrical Engineering), 2012, págs. 1338-1346
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – The purpose of this paper is to present the application of a novel hybrid algorithm, called MeTEO (Metric‐Topological‐Evolutionary‐Optimization), based on the combination of three heuristics inspired by artificial life to the solution of optimization problems of a real electronic vacuum device.

    Design/methodology/approach – The Particle Swarm Optimization (PSO), the Flock‐of‐Starlings Optimization (FSO) and the Bacterial Chemotaxis Algorithm (BCA) were adapted to implement a novel meta‐heuristic MeTEO the FSO has been powerfully employed for exploring the whole space of solutions, whereas the PSO is used to explore local regions where FSO had found solutions, and BCA to refine the solutions found by PSO, thanks its better performances in local search.

    Findings – The optimization of the focusing magnetic field of a Travelling Wave Tubes (TWT) collector is presented in order to show the effectiveness of MeTEO, in combination with COLLGUN FE simulator and equivalent source representation. The optimization of the focusing magnetic structure is obtained by using a maximum of 100 steps for each heuristic.

    Practical implications – The paper describes the development of a novel efficient parallel method for the solution of electromagnetic device optimization problems.

    Originality/value – The paper shows the capabilities of a novel combination of optimization methods inspired by “artificial life” which allows us to achieve effective solutions of multimodal optimization problems, typical of the electromagnetic device optimization, with an acceptable computational cost, thanks also to its natural parallel implementation.