|IMPORTANT LEGAL NOTICE - The information on this site is subject to a disclaimer and a copyright notice.|
3A-3 TECHNOLOGIES FOR IMPROVED AIRCRAFT EFFICIENCY
Network on industrial design and control applications using genetic algorithms and evolution strategies - INGENET
The increasing complexity of design in Aeronautic, Automotive and Energy Industries requires more and more robust optimisation and control tools for solving simultaneously discrete, continuous and combinatorial problems.
Evolutionary Algorithms (EAs) are artificial intelligence techniques which mimic nature according to Darwin's "survival of the fittest" principle. They randomly encode physical (quantitative or qualitative) variables via digital DNA inside computers and are known for their robustness to better explore search spaces and find near-global optima than traditional optimization methods. For this reason Evolutionary Algorithms are becoming widely used in many engineering and scientific problems.
The main goal of the INGENET Network is to bring together academic and industrial partners in order to evaluate and compare performances of recent evolution based methodologies and (parallel) softwares on selected test problems of industrial interest.
The Network includes the following tasks:
Fluids Dynamics, Acoustics, Structure Mechanics, Electromagnetics, Automation Control and Energy are the featured multidisciplinary areas in Engineering and Applied Sciences targeted by INGENET where evolution approaches work impressively well.
The ultimate goal of INGENET is to stimulate and disseminate widely and electronically in Europe innovative evolution based methodologies and software, providing an interactive dynamic bridge between Artificial Intelligence, Computer Sciences and Engineering for solving real life problems.
European dimension and partnership
The objective of the INGENET project is to set up a network of expertise in the field of Evolutionary Algorithms allowing universities, laboratories and industries to collaborate in the development and validation of methodologies involving EAs as well as offering a benchmark platform for applications of industrial interest.
The INGENET network intends to tandem node partners from academia and Industry with the INGENET kernel.
Industrial nodes will participate in the definition of test cases and their computations with Evolutionary Algorithms. They will participate during workshops in data evaluation and interpretation, giving views on the results from an industrial perspective. They will work in tandem with one or several universities or institutes of their choice to favour stimulation between teams and encourage the cross-fertilization of evolution methodologies with the industrial environment.
The interest of universities and research institutes to participate in INGENET consists mainly in using Evolutionary Algorithms in practical industrial applications in order to highlight the particular strength of these optimization algorithms. As an outcome of INGENET these academic centres will get into contact with European industries or SMEs represented in INGENET to look for potential applications of EAs beyond the test cases discussed in the network.
This new optimization technology was initially developed in the United States. Nowadays industrial problems require these global optimization techniques and there is an increasing interest in Industry to apply these robust algorithms to their complex multi-objective multidisciplinary design and control problems.
Contract Brite-Euram Nr: BRRT-CT97-5034
989 500 ECU
15 November 1997
14 November 2000
EC scientific officer
Jyrki KALEVI SUOMINEN
Dassault Aviation SA
Etablissement de Saint-Cloud
British Aerospace Defence Ltd - Preston-Lancashire (GB)