Analysis of the characteristics of noise from substations in buildings

The urbanization has resulted in an increasing number of transformer stations, which has resulted insignificant building noise problems. However, noise problems persist because of inadequate noise characterization and the use of imperfect noise evaluation indexes for centralized substations. Based on this problem, a transformer vibration noise coupling analysis method based on empirical mode decomposition (EMD) and spectrum analysis is proposed in this study. The proposed method accurately and effectively screens and characterizes transformer noise and provides a theoretical basis for transformer noise reduction. To verify the effectiveness of the proposed method, a transformer was reconstructed as an example. It was found that the low-frequency noise from the transformer was mainly caused by vibrations with a frequency below 500 Hz, particularly frequencies of 300 Hz and 100 Hz and 50 Hz. Through the calculation and analysis of eigenvalues, the noise reduction measures focusing on vibration reduction were proposed. In the end, a noise reduction of 10 dB was achieved, which meets the comfort requirements. This method can accurately and effectively identify the characteristics of transformer noise, which makes up for the insufficiency of transformer characteristics analysis in the past. Provide guidance for perfecting transformer noise evaluation index. Practical implication: The noise problem caused by substations is getting more and more serious. Conventional noise detection and noise reduction methods can no longer meet people’s requirements for sound comfort. The coupling analysis method of vibration and noise based on EMD and spectrum analysis proposed in this study can effectively extract the characteristics of transformer noise. It provides theoretical support for the noise reduction transformation of transformers, and solves the problem that the current engineering noise reduction transformation has no theoretical basis. Noise characteristic analysis can make up for the shortcomings of existing acoustic comfort indicators that only use sound pressure level as the evaluation indicator.