A system and method for thunder signal recognition based on convolutional neural network
A convolutional neural network and signal recognition technology, applied in the field of lightning signal monitoring, can solve the problem of inability to meet the strong randomness of thunder recognition, and achieve the effect of meeting real-time and high-efficiency requirements, improving robustness and accuracy, and overcoming interference.
Active Publication Date: 2022-04-01
WUHAN NARI LIABILITY OF STATE GRID ELECTRIC POWER RES INST
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the thunder signal will vary with the form, strength, and distance of the lightning, and the traditional waveform, frequency and other characteristics can no longer satisfy the identification of strong random thunder
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0023] Below in conjunction with accompanying drawing and specific embodiment the present invention is described in further detail:
[0024] The thunder signal identification system based on the convolutional neural network designed by the present invention, such as figure 1 As shown, it includes a manual labeling module 1, a training set data preprocessing module 2, a feature extraction module 3, a model learning module 4, a sound data preprocessing module 5 to be identified, and a classification recognition module 6; The sound training set data is manually marked with the type of thunder, and the thunder classification label positioning file is established, and the thunder training set data is in one-to-one correspondence with the thunder classification label positioning file; the training set data preprocessing module 2 is aimed at by the resample function The thunder training set data output by the manual labeling module 1 is subjected to down-sampling processing to genera...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a thunder signal recognition system and method based on a convolutional neural network. The thunder type is manually marked for the thunder training set data, the thunder classification label positioning file is established, and the Mel of the thunder training set data is extracted. The frequency cepstral coefficient is based on the convolutional neural network to train the Mel frequency cepstral coefficient to obtain the thunder classification recognition model, and finally realize the classification and recognition of the sound data to be recognized. The method of the present invention uses a deep learning method to improve the robustness and accuracy of the traditional thunder signal detection method, so as to meet the real-time and high-efficiency requirements of the thunder positioning system for the identification results, and can not only overcome the impact of various environmental noises on the thunder signal identification process. interference, but also suitable for thunder signals with different characteristics originating from different channel positions.
Description
technical field [0001] The invention relates to the technical field of lightning signal monitoring, in particular to a convolutional neural network-based thunder signal recognition system and method. Background technique [0002] Lightning is one of the serious disasters affecting human activities in nature. It will not only cause casualties, but also cause immeasurable economic losses to my country's aerospace, electronics industry, petrochemical, transportation, forestry and other industries. In recent years, frequent accidents caused by lightning have made people from all walks of life pay more and more attention to the real-time monitoring and protection of lightning. [0003] Real-time monitoring of lightning is the basis of lightning protection and disaster mitigation. China began to study lightning positioning technology in the late 1980s, and has established lightning positioning systems in more than 30 provinces since the 1990s. It has now been realized. National n...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F2218/08G06F2218/12
Inventor 章涵谷山强李健严碧武吴敏苏杰雷梦飞陈扬许远根王宇李涛
Owner WUHAN NARI LIABILITY OF STATE GRID ELECTRIC POWER RES INST