抄録
This paper describes psycho-acoustic experiments using online optimization in order to improve the performance of active noise control to modify noise into comfortable sound. Three comfortable sound parameters characterizing the artificially synthesized based on insect's singing are set as deign variables and the alpha wave components of the electroencephalogram measured after hearing the sound is set as the objective function. By using the simplex method the comfortable sound parameters are optimized during experiments. Since experimental results show that there exists optimum combination among the comfortable sound parameters, the comfortable sound plane is calculated as the optimum combination. After discussing multiple planes and dependence on frequency of sinusoidal function, active control experiments to modify machinery noise to comfortable sound are carried. It is confirmed that the proposed comfortable sound function improves the performance of the active noise control.
