Study Of Vibration Sensors Placing In The Estimations Of Structural Acoustic Radiation

The estimations of structural acoustic radiation include sound pressure estimation, acoustic power estimation and acoustical measure variation estimation. When the vibration sensors are evenly placed, the relationships between the estimation accuracy and the vibration sensor number requirement, the position of the excitation force and field points are investigated.The methods of acoustic radiation estimation and the sensor number requirement are introduced firstly. The sound pressure and acoustic power from a cylinder excited by a point force is estimated by employing the approximate Kirchhoff integration method. It is found that more sensors are needed to meet the given estimation precision requirement when the frequency increases in general. However, at the frequencies where the peaks occur in the field points’sound pressure level or sound power level curves, less sensors are needed. The position of the excitation force has little effect on the sensor number requirement.Then the required sensor number of the whole space sound pressure estimation is investigated. Compared with the curve of required sensor number when estimating a single field point sound pressure, the curve of required sensor number of the whole space sound pressure estimation is smoother. If every field point achieves the same estimation precision in the frequency band, the maximum demands on the number of sensors of all field points are needed at every frequency point. The position of the field point has little effect on this maximum, but has great influence on the frequency point at which this maximum is.Finally, vibration sensor number requirement for the acoustical measure variation estimation is investigated. It is found that the estimation of the acoustical measure variation is more accurate than the sound measure estimation at the frequencies where the sound measure is overestimated or underestimated simultaneously. For all field points, the estimation of the acoustical measure variation is more accurate at the eigenfrequencies which are excited in both work states when the sensors’placement is fixed.