Feature extraction and discrimination of blasting and natural earthquake waves

Abstract

The feature extraction of blasting and natural earthquake waves can effectively discriminate between the two, and is conducive to mastering the time–frequency distribution and failure mechanism. However, feature extraction and discrimination between the two are often subject to external interference. A complete ensemble empirical mode decomposition and multiscale permutation entropy and Hilbert–Huang transform method is proposed to extract features and discriminate for blasting and natural earthquake waves efficiently and accurately. Firstly, the simulation signals verify feasibility and applicability. Secondly, six scenarios for blasting and natural earthquakes are collected and compared. Thirdly, feature extraction and discrimination are performed. Results show that the proposed method can extract and discriminate features between the two effectively. Furthermore, the energy of the former is concentrated within 200 Hz, while the latter is within 3 Hz, and different scenarios or magnitudes have different characteristics. Moreover, the blasting earthquake has features such as short duration, peaks corresponding to the blasting initiation time and fast energy attenuation. The natural earthquake has features such as long duration, multiple peaks, and slow energy attenuation, resulting in more significant harm. This method can extract features and discriminate for blasting and natural earthquake waves from the perspective of time–frequency and energy effectively.

Keywords:

• Blasting earthquake
• natural earthquake
• time-frequency analysis
• feature extraction
• CEEMD-MPE-HHT

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable requests.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors gracefully acknowledge the Pacific Earthquake Engineering Research Center (PEER) in their work. This research was supported by the National Natural Science Foundation of China under Grant (No. 42174052).