Optimized design of error sensors in active noise control window applying boundary control

• Main Author: Wang, Miqing
• Other Authors: Gan Woon Seng
• Format: Thesis-Master by Research
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/161733
• Institution: Nanyang Technological University

The increasing urbanization leads to the challenge for urban planners to alleviate the noise pollution caused by moving vehicles and construction sites located around the residential areas. Three strategies including control of the noise source, control along the propagation path, and control at the receiver side are applied in noise control engineering. Passive noise control (PNC) has been widely adopted in the first two strategies. However, if the frequency of the target noise is low, bulky PNC solutions with high cost will be needed to obtain expected noise reduction. Closing the window is an easy solution to achieve control at the receiver side. Nevertheless, the natural ventilation will be blocked. Active noise control (ANC) system is a promising solution due to its superiority in cancelling low frequency noise. Meanwhile, the ventilation function of the window can be maintained. Thus, attentions have been focused to the implementation of ANC window systems along an open aperture using boundary integrated speakers. Former research works focused on implementations of faster algorithms and more powerful hardware platforms to improve the noise control performance. However, these complicated strategies achieved limited improvement. Few attentions were paid to the physical arrangement of ANC systems, to our best knowledge. Hence, we focus on the influence of the practical system layout on the noise control performance in this research work. Specifically, the number and placement of error sensors are the research focuses. This thesis provides systematic analyses on the placement of error sensors of the ANC window system using boundary control to form a guideline for the placement of error microphones in practical implementations. Three main research restrictions are considered. The first restriction is the limitation on the number of error sensors in practical implementations due to the increased cost and computational complexity. The second restriction is that the distance between secondary sources and error sensors will affect the control performance. The third restriction is the distance between error sensors, which will affect the noise control performance of the targeted control area. To solve the above-mentioned research restrictions, a method based on numerical simulations to determine the optimal placement of error microphones of ANC window systems using boundary integrated speakers is proposed. Our aim is to give quick and accurate investigations on the noise control performance under different placements of error sensors. In the proposed method, COMSOL Multiphysics, which provides physical field simulation using finite element method (FEM), is applied to accurately model the acoustic environment of open apertures and interior space. MATLAB tool is used to calculate and analyze numerical simulation results from COMSOL. The performance of the ANC window system is evaluated based on the noise attenuation performance. The combined simulation results are used to analyze the optimal placement of error microphones in the ANC window system. The performance of the ANC system can be improved without consuming excessive computational resources. Subsequently, four factors which affect the optimal placement of error sensors are analyzed, including the boundary condition, the frequency and incident angle of primary noise, the size of the aperture, and the size of the target control area. Two boundary conditions of the target control area: free field and total reflection boundaries are considered to match with different acoustic environments in practical implementations. Primary noise from 100 Hz to 1000 Hz with three different incident angles are analyzed. The size of the aperture is proved to have the most significant influence on the controllable frequency range and the optimal placement of error sensors, while the size of the target control area barely has an influence. In the last chapter, an ANC window with boundary integrated speakers is designed, and the experiments are conducted in a residential room. A multiple channel ANC system is implemented on the PXIe-8135 platform. A YAMAHA speaker is adopted as the primary noise source. Six measurement microphones, whose placement is based on the ISO standard, are adopted to measure the sound pressure level (SPL) of the interior space. The ANC system is tested using primary noise in different frequency band and is able to achieve a desirable noise control performance inside the residential room. The comparison between simulation and experimental results is discussed, to prove the efficiency of the proposed COMSOL & MATLAB simulation method in guiding the design of a practical ANC window system with boundary integrated speakers.