Active sound absorbing-insulating layouts controlled by algorithms based on neural network and genetic algorithm.

Project manager: Leszek Morzyński Ph.D. (Eng.)

Project summary:

The main objective of the project was to develop a control algorithm of active sound absorbing-insulating system was based on neural network and genetic algorithm. As a result control algorithm of active sound absorbing-insulating system based on neural sigmoide one-way network which is taught to be used with genetic algorithm. In genetic algorithm was used: floating-point encoding parameters, multiple crossing heuristic, uniform mutation and selection based on a turnover roulette.
Within the task following steps were completed:
- Simulation studies in order to develop control algorithm of active sound absorbing-insulating system;
- Modifications of the laboratory post;
- Development and implementation of the absorbing-insulating model and active control system;
- Implementation of the control algorithm in the control system;
- Research of developed control algorithm of active sound absorbing-insulating system;
- Prototype development of a sound absorbing-insulating cover with an active sound absorbing-insulating system;
- Verification tests of the control algorithm of active sound absorbing-insulating system for the real sources of noise.

Unit: Department of Vibroacoustic Hazards

Duration: 01.01.2011 – 31.12.2013