Active noise control using circular barrier and array speakers

Abstract

Many workers in the industry have been exposed to loud noise for a long time, and noise-induced hearing loss of workers is gradually intensifying. A noise barrier, which is a conventional noise reduction method, can be used, but it is not effective to low-frequency noise reduction due to diffracted noise. In addition, there is a problem in that the size of the barrier is limited because it must be movable to respond to a change in the workplace. Active noise control, which is another noise reduction method, reduces noise by generating a sound field with an opposite phase to noise through a control speaker in a manner based on destructive interference. An adaptive algorithm is performed through the data measured by the reference sensor and the error microphone to obtain the optimal gain of the control speaker to generate the control sound field. Due to the property of the noise reduction method, a large number of speakers and microphones are required to maintain a controllable space size for high-frequency noise, resulting in cost and equipment placement problems. Accordingly, an active noise barrier in which active noise control is introduced into the noise barrier has been proposed. By arranging speakers around the noise barrier to conduct active noise control for the diffracted noise, the required size of the barrier can be reduced, and by limiting the propagation of noise through the barrier, active noise control can be performed with fewer speakers than in the case of a free sound field. Since the effective frequency bands of the two methods are different, the frequency band of reducing noise can be expanded. However, conventional active noise control methods using error microphones are not suitable for active noise control systems using compact noise barriers aimed at mobility and quick installation and calibration of control filters. Error microphones and real-time digital signal processing devices for performing active noise control of the target control space not only increase the cost required for hardware setting but also increase the system complexity which increases the stabilization time of the noise control system. In addition, there is a problem in that the arrangement of error microphones in the workspace is limited because the workers' activity should not be disturbed. A virtual microphone technique can be introduced that estimates the measurement value of the target position through microphones placed elsewhere, but requires measurement of the transfer function between the placed microphones and the control points. Measurement in a large space requires a lot of time, and there is a problem with re-measuring when the environment changes. As a result, when active noise control is performed using error microphones, the system becomes complicated and the installation and calibration time of the noise control device becomes longer. Therefore, in this study, a compact hybrid noise control system based on a theoretical model is proposed as a movable noise reduction method. Based on an appropriate theoretical model, the control filter required for active noise control is calculated without a real-time digital signal processing device and error microphones. Control filters can be calculated with only some information, such as distance, or pre-calculated values can be used, allowing simple installation and calibration after equipment movement. So, the proposed method is not affected by equipment noise or other external noise, and the size of the control space does not matter. The proposed method can be applied to various structures, but this study deals with a structure consisting of a circular barrier and control speakers placed at the edge of the barrier. The addressed noise control system can estimate the sound field through a simplified two-dimensional theoretical model through several assumptions, and can greatly reduce the amount of calculation. Assuming that the noise source and target control space are determined by the user and the noise control system is arranged on the same axis, an axisymmetric noise problem is assumed. The noise source was considered as a monopole source by assuming a far-field condition, and the barrier was assumed to be a thin and acoustically hard boundary condition. In the simplified noise problem, the sound fields are theoretically calculated, and the insertion loss and noise reduction performance in the control space of the hybrid noise control are compared with the case of using barrier and active noise control respectively. It is confirmed that noise reduction can be achieved in a wide frequency band because the hybrid method generates a control sound field similar to the noise field. In addition, the noise reduction tendency according to the size and placement position of the active barrier, which is a design variable, and the influence of errors that may occur during the installation process are investigated. Based on the noise reduction performance results confirmed in various cases, error factors to be paid attention to are explained, and design guidelines for the appropriate placement position and size of active barriers are suggested. Finally, to verify the proposed noise control method, the theoretically calculated control filter is applied to the constructed experimental system to confirm whether noise is reduced in the target space. In the case of the experiment, control speakers were placed around the barriers to replace the ring control source, and the number without performance degradation is described. Since the performance may be deteriorated due to the difference between the theoretical model simplified through various assumptions and the experimental system, the main factors are identified by investigating the performance degradation factors through FEM simulation. The noise control system was built in a way to minimize the effect of the identified major degradation factors, and a closed baffle ring piston structure was introduced instead of a ring source as the control source in the theoretical model to compensate for the interaction between the speaker and the barrier. It is verified that noise reduction similar to the maximum performance of the experimental system can be achieved by applying the theoretically calculated control filter through the modified theoretical model and the hybrid noise control system. In conclusion, an appropriate theoretical model and system structure of a theoretical model-based compact hybrid noise control system are suggested. Also, guidelines for sufficient performance are provided, and it is verified that noise reduction can be achieved in an actual system.