In order to promote the construction of eco-friendly power systems

bird damage prevention and control measures have gradually transformed into differentiated prevention and precise governance. It is urgent to realize the monitoring of bird activities in the near-line area of overhead transmission lines and guide the accurate deployment of bird-related fault prevention measures. This paper proposes a bird monitoring technology for transmission lines based on distributed acoustic sensing

which mainly includes bird acoustic event detection

sound source localization

and bird species identification. First

this paper carries out the line simulation test

and proposes a two-stage bird acoustic signal detection scheme combining the root mean square energy threshold and the kurtosis threshold of the intrinsic mode function to realize the detection of bird acoustic events. Then

the generalized cross-correlation phase transform weighted time delay estimation method is used to calculate the arrival time difference of bird song signals between different channels

and the error of sound source coordinate deduction is less than 1m. Finally

a bird voiceprint recognition model based on time-frequency joint attention and a convolutional neural network is proposed. The recognition accuracy of six high-risk birds is about 95%. The experimental results show that this technology provides a distributed technical scheme for bird activity monitoring on the whole transmission line.