自由声场自适应有源消声中误差传感器的最优布放

A GA Optimal Arrangement of Error Sensors in Free Acoustic Field AANC

有源噪声控制中，误差传感器布放的合理与否，直接关系到自适应有源噪声控制系统的降噪效果。本文利用遗传算法研究了自由场ＡＡＮＣ中误差传感器的布放问题，较好地解决了传统的梯度搜索算法难以解决的位置优化问题。仿真结果表明，用遗传算法寻找到的误差传感器的最优位置与理论分析结果吻合。

We used objective function in a previous paper to study the effect of arrangement of error sensors on attenuation level in free acoustic field adaptive active noise control (AANC). But when the number of variants is large, using ref., which follows traditional gradient search method, to find global optimal solution is almost impossible. We present a genetic algorithm (GA) method as a better way of obtaining optimal arrangement of error sensors. The mathematical formulas needed for finding this GA optimal arrangement are eqs. (1), (2), (4), and (11) derived by us. Figs.4 through 7 present computer simulation results. The number of variants is small for Figs.4 through 6 and accurate global optimal solutions can be found by traditional gradient search method (solid curves in Figs.4 through 6). The GA method gives the top dots of the solid curves. The number of variants in Fig.7 is large, but the optimal arrangement can still be easily found with our GA method.