Transmission Line Audible Noise Research Articles

Transmission Line Audible Noise Prediction Based on CNN-BiLSTM-Attention Method

High-voltage transmission lines’ audible noise parameters are impacted by a variety of multidimensional elements. in order to better the accuracy of audible noise prediction and effectively utilize the time-series properties in the observed audible noise data. In this paper, we propose a combined model of a convolutional neural network (CNN) and a bidirectional long and short-term memory network (BiLSTM)-based attention mechanism based on feature filtering for transmission line audible noise prediction. Firstly, using the transmission line real-world audible noise data as the dataset, the multidimensional factor time series parameters are optimally filtered, and high correlation feature vectors are extracted by using CNN. Secondly, the extracted feature vectors are fed into the BiLSTM for training and prediction, and the prediction performance is further improved by introducing an attention mechanism at the BiLSTM end so that the model focuses on learning more important data features. Finally, the prediction analysis using actual recorded audible noise data from a 500 kV AC transmission line in Sichuan Province demonstrates that the combined CNN-BiLSTM-Attention model suggested in this paper has a higher prediction accuracy than the BiLSTM, CNN-BiLSTM, and BiLSTM-Attention models.

Invalid Data Rejection of Audible Noise on AC Transmission Lines Based on Moving Window Kernel Principal Component Analysis

The statistical characteristics of the nighttime noise data of 1000 kV AC transmission lines were investigated, the noise data of the Huainan-Shanghai 1000 kV AC transmission line collected at night (0:00 to 6:00) from September 25, 2015, to February 16, 2016, were statistically analyzed using the nonparametric statistical K-S test, and the outliers were detected using the moving window kernel principal component analysis (MWKPCA). The results show that after the ineffective data are removed by MWKPCA, the 5, 50, and 95% values of the data are basically unchanged. To a certain extent, the method proposed in this paper can remove the invalid audible noise (AN) data of 1000 kV AC transmission lines without affecting the subsequent study of AN, we use various machine learning algorithms to predict the A weight sound level (Awsl) before and after the invalid data rejection, and the results show that the invalid data rejection has contributed to the improvement of the transmission line AN Awsl prediction accuracy.

Long-term evaluation of HVDC transmission line audible noise and its correlation with background noise

The audible noise (AN) generated by corona discharge from a full-scale, high-voltage direct-current (HVDC) transmission line was evaluated over a one-year period from December 2017 to November 2018. The double bipolar HVDC transmission line using a metallic return conductor was demonstrated using a full-scale tower having a height of 21 m in Gochang, Korea. The measured AN in an outdoor environment was statistically evaluated and compared in different seasons and weather conditions. The A-weighted spectrum of the HVDC corona noise exhibited a greater distribution than the background noise above 1.0 kHz in the frequency domain. Various background noises were considered and eliminated to improve the measurement accuracy. The highest AN levels were exhibited during the spring season. Individuals in the age range of 18–60 years were randomly selected to conduct a human-perception test. The results indicated that 15% of the participants became aware of the AN from the corona discharge when it was >5 dBA greater than the background noise. The AN was <36 dBA when the conductor surface gradient was <18 kV/cm with a six-bundle sub-conductor.

Audible Noise Characters Analysis of 220kV Xiangtang Transmission Line

In order to know the corona audible noise characters of the 220kV transmission line and its influences on nearby residents’ life, we measured the corona discharge UV-light, the environmental noise and sound energy of the transmission line respectively. The study results showed that the corona discharge occurred at the transmission line. The audible noise sound level on the roof was 46.2dB(A) at night, that was slightly beyond the limit value of the first zone of Sound Environment Quality Standard. At night, the transmission line audible noise like ‘SiSi’ was so obvious that disturbs residents. The sound energy analysis results showed that the transmission line audible noise energy focused on the frequency range below 200Hz. The directivity study of sound intensity showed that the highest sound energy happened at 100 Hz in the direction from the transmission line to the top floor of the residents’ house, in which explains that the transmission line noise, slightly beyond the limit, is obvious.

Results from AC transmission line audible noise studies at the Anneberg EHV test station

Results from single-phase test line audible noise measurements on various conductor configurations for transmission lines in the 400 kV range are presented. The predicted L50 audible noise levels are given as functions of the rain rate. After single-phase to three-phase conversion, a comparison is made with results from long-term measurements on an operating 400 kV line. It is shown that the long-term L50 audible noise level in rain can be predicted with sufficient accuracy from the test line measurements.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Methods for Predicting ac Transmission Line Audible Noise by Short-term Single-Pahse Tests

The problems of converting short-term single-phase audible noise test results into three-phase long-term levels are discussed in the paper, and methods used at the Anneberg EHV test station for these conversions are presented in detail. The methods are applied to an EHV three-phase line configuration, and the calculated audible noise levels during rain are presented for different rates of rainfall.

Methods for Predicting AC Transmission Line Audible Noise by Short-Term Single-Phase Tests

Methods for Predicting AC Transmission Line Audible Noise by Short-Term Single-Phase Tests

A Methematical Model of Transmission Line Audible Noise Part 2: Comparison with Experiment

This part of the paper compares the model developed in Part 1 with experimental results obtained by a number of investigators. The shape of the curves predicted by the model is shown to be in good agreement with results showing audible noise as a function of gradient or rainfall rate. Simple curve fitting is used to find values of the constants in the model using published audible noise data. A single equation predicting audible noise for multiconductor lines in saturated conditions is presented. Implications in the areas of interpretation of existing data, conductor testing and conductor design are discussed.

A Mathematical Model of Transmission Line Audible Noise Part 1: Background and Model Development

A Mathematical Model of Transmission Line Audible Noise Part 1: Background and Model Development

A Mathematical Model of Transmission Line Audible Noise Part 2: Comparison with Experiment

A Mathematical Model of Transmission Line Audible Noise Part 2: Comparison with Experiment

A Mathematical Model of Transmission Line Audible Noise Part 1: Background and Model Development

A mathematical model is developed for transmission line corona audible noise (AN). The noise is described as the sum of the noises from a number of uncorrelated sources on the surface of the conductor. The intensity of each source is assumed to be a function of gradient and the number of sources is a function of rain rate, voltage gradient and the surface aging condition. Townsend's first ionization coefficient is used to relate the number of sources to the gradient. The steady state solution of the AN equation is plotted to show the variation of noise with gradient for various rain rates, and the variation of noise with rain rate for various gradients.

Modification of Transmission Line Audible Noise Spectra to Reduce Environmental Impact

Modification of Transmission Line Audible Noise Spectra to Reduce Environmental Impact

Modification of Transmission Line Audible Noise Spectra to Reduce Environmental Impact

The aversiveness of audible noise from EHV electric transmission lines was evaluated for spectral modifications of 5 different tape-recorded samples. The relative contributions of low-frequency humming and buzzing and of high-frequency hissing and crackling components of the noise were determined. The high-frequency components of the hoise proved more aversive for all five samples to a group of 25 listeners. Therefore, for outdoor or open window listening, engineering efforts should be directed at reducing the high-frequency portion of the audible noise spectrum.

Use of the ’’acoustic menu’’ in assessing human response to audible (corona) noise from electric transmission lines

A behavioral preference procedure, the ’’acoustic menu,’’ was used to assess human aversion to the sounds from two samples of transmission line audible noise and from samples of other environmental noises. The audible (corona) noise produced by extra-high voltage (EHV), overhead transmission lines was tape-recorded during moist weather. Reproduced samples of corona noise, other common environmental noises, and artificial reference stimuli were compared in two experiments. The two corona noise samples were found to be equally preferred to: (1) a 1000 Hz octave band of noise about 11 dB SPL higher than the corona noises and (2) a collection of other environmental noises about 8 dB SPL higher than the corona noises. The corona noise samples were more aversive than the ambient sounds (including rainfall) occurring near rural transmission lines and were roughly equivalent in aversiveness to the noise from a room air conditioner (indoor recording). Knowledge of the source of the corona noise (from photographs and a brief description) did not affect its aversiveness. Of the simple frequency-weighting scales, the A-weighted sound level reduced the difference between the two corona noise samples and the set of other environmental sounds from about 8 dB SPL to about 3 dB. The D-weighted sound level further reduced this discrepancy to about 2 dB. These results were obtained with only two samples of corona noise; other samples may give different results.