Dissertation > Excellent graduate degree dissertation topics show

Paralleling Genetic Annealing Algorithm with OpenMP

Author: ChenSiCheng
Tutor: ZhuHongBing
School: Wuhan University of Science and Technology
Course: Computer Software and Theory
Keywords: Protein Structure Prediction Simulated annealing algorithm OpenMP Parallel technology
CLC: TP301.6
Type: Master's thesis
Year: 2011
Downloads: 59
Quote: 1
Read: Download Dissertation

Abstract


In the field of biology based on the amino acid sequence of protein structure prediction is a complex and challenging problem . Genetic annealing algorithm is a new algorithm of the advantages of the combination of genetic algorithms and annealing algorithm . It overcomes the precocious early as convergence of genetic algorithm , local search optimization capability shortcomings of poor , but also improve the simulated annealing algorithm efficiency is not high . Genetic annealing algorithm (GAA) is to be applied in the AB -lattice model protein structure prediction (PSP) , one of the most representative algorithms , genetic annealing algorithm requires a large-scale and long time of calculation . Therefore, looking for some way to reduce the computation time and calculated the scale of the problem of protein structure prediction has become an urgent task . The emergence of multi-core processors , gradually mature parallel language and can be run directly in the existing dual-core processor computer algorithm based on these shared memory programming provides a good prerequisite . OpenMP is based on the industry standard for shared memory programming , it has many advantages , such as simplicity , portability and better scalability sex . Therefore , most users prefer to use OpenMP to improve the efficiency of the algorithm . This paper presents a parallel simulated annealing algorithm (GAA), the purpose of this algorithm is to improve the computing speed of protein structure prediction problem . Parallel genetic annealing algorithm uses a coarse-grained parallel model , several sub- populations replaced the original single population , the independent evolution of each sub-population , each evolution after the completion of this sub - population the best individual in turn replace the worst individual in the other sub - populations to promote the evolution of the entire population . Experimental results show that this parallel algorithm greatly improve the computational efficiency of the genetic annealing algorithm .

Related Dissertations

  1. Civilian ship management system design and implementation,TP311.52
  2. Radar jamming and interference covering optimal allocation of resources,TN974
  3. Research and Application of hybrid genetic algorithm - based shop scheduling problem,TP18
  4. Based on the feature fusion membrane protein transmembrane helix prediction,Q51
  5. Optimization on Input Operations and Management Information System Development of a Steel Plate Stacking Yard in a Shipyard,F426.474
  6. FCM combines multi-resolution image cone multicore parallel image segmentation algorithm,TP391.41
  7. Predictions of Human Kinome-Small Molecule Interactions,Q55
  8. Study on Design and Tuning Methods of Sampled Fiber Gratings with Special Structure,TN253
  9. Application of Parallel FDTD and MPSTD Algorithm in EM Scattering,O441.4
  10. Research on Model and Optimization of Production Scheduling for Copper Sheet and Strip Oriented MES,TP391.7
  11. Study on Multi-Floor Facility Layout Problem Based on M/G/1 Lift Handling System,TU857
  12. Design and Implementation of Motor Train Sets Scheduling System for Passenger Dedicated Lines,U268
  13. Research on the Method of Resisting Co-channel Interference with LPI Radar,TN974
  14. Branch flight training scheduling system,V323
  15. Software optimization techniques in software development projects,TP311.52
  16. Theoretical Simulating and Experiment Research of Beam Shaping Based on Liquid Crystal Spatial Light Modulator,TN761
  17. Research on Heterogeneous Fixed Fleet Vehicle Routing Problem with Fuzzy Time Windows,F252
  18. The Research of Three-dimensional FDTD Parallel Algorithm Based on MPI and OpenMP,TN011
  19. Parameters Optimization of Micro-End-Milling Based on Simulated Annealing Genetic Algorithm,TG54
  20. Multi-stream heat exchanger channel arranged to optimize the design of,TK172

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > General issues > Theories, methods > Algorithm Theory
© 2012 www.DissertationTopic.Net  Mobile