Created by W.Langdon from gp-bibliography.bib Revision:1.7964
This thesis addresses the problem of actively control acoustic noise in ducts through the application of genetic algorithm - GA and genetic programming - GP (called genetic control - GC). Genetic programming obtain a self structured autonomous control model and genetic algorithms adapt model's parameters under real time.
Three different strategies were adopted with GA. In the Simple Genetic Algorithm (SGA) each individual of a generation represents a specific frequency, phase and amplitude used in cancellation of noise and the fitness function is the average energy of the signal. The Successive Approach Genetic Algorithm (SAGA) is a modification of SGA, where a first level procedure searches for candidate frequencies and a second level improves them between fixed limits, with phase and amplitude. To run in real time, a gain/delay model was coded into the chromosome.
A simulation model was developed to test the software and to analyses the behaviour of the genetic algorithm parameters.
The software was designed to work in a parallel DSP TMS320C44 architecture managing processors communication and shared memory with high performance. A mono processor version was developed to control the duct system under real time with noise reduction.
The acoustic feedback was removed through the microphone confinement, special sound boxes and through adaptive model approach.
Genetic programming applied to the system converges to the genetic algorithms gain/delay model as foreseen by the theory and experiment",
Genetic Programming entries for James Cunha Werner