Description

In computational science and engineering there are many sophisticated, rigorous and realistic numerical models of physical systems. It is now commonplace in the engineering design process to use these models to find the combination of design parameters that minimise (or maximise) some specified design objective. Optimisation programs seek to apply mathematical techniques and heuristics to guide a computer to find solutions that satisfy this objective. The research described in this book is directed toward developing methods that use parallel computing to reduce total optimisation time. It explores gradient descent, direct search and evolutionary programming algorithms to find fast, effective methods for the solution of real-world problems. The work provides a snapshot of an emerging approach to automated design, improved solutions and reduced time-to-market. It should serve as a useful introduction to the use of parallel computing for professionals in engineering design considering the use of modeling and optimisation, and researchers tackling parameter estimation problems across a wide range of disciplines.