Global optimization of a dryer by using neural networks and genetic algorithms

✍ Scribed by A. Hugget; P. Sébastian; J.-P. Nadeau

• Publisher: American Institute of Chemical Engineers
• Year: 1999
• Tongue: English
• Weight: 162 KB
• Volume: 45
• Category: Article
• ISSN: 0001-1541
• DOI: 10.1002/aic.690450609

✦ Synopsis

For many optimum design problems, the objecti®e function is the result of a complex numerical code and may not be differentiable and explicit. The first aim is to propose a way of sol®ing such complexity on an example problem. A no®el and global strategy in®ol®ing artificial neural networks and a genetic algorithm is presented and ®alidated for an industrial con®ecti®e dryer. To begin with, a method to represent a drying model using artificial neural networks is defined. This method is tested and the results are compared with those obtained with classic numerical methods. This approach drastically reduces simulation times and maintains good accuracy and generalization properties. Second, the associated optimal design problem is considered. This optimization appears as a difficult combinatorial problem with a complex objecti®e function that in®ol®es different economical criteria. The second aim is to present the methodology to sol®e this problem using genetic algorithms. Final results illustrate the efficiency of this global approach.