A foundation pit wireless safety monitoring and early warning method based on strain gauge array vibration feature extraction

Abstract

The application discloses a kind of based on strain gauge array vibration feature extraction's foundation pit wireless safety monitoring early warning method, it is related to foundation pit engineering safety monitoring technical field.The method includes: through double passage parallel acquisition node vibration signal;The kurtosis value in the node vibration signal and the difference between high-frequency energy ratio and background model benchmark value are calculated;Gateway calculates and reports real-time kurtosis value, the difference between high-frequency energy ratio and background model benchmark value is marked as suspected structure rupture event;For suspected structure rupture event, the original vibration waveform of corresponding node and adjacent node is obtained;Based on the original vibration waveform of corresponding node and adjacent array node, direction angle discriminant factor and rupture impact index are calculated, and dangerous level is determined;According to the determined dangerous level, the early warning information of corresponding level is output, and the acquisition cycle, acquisition length and reporting strategy of corresponding node are dynamically adjusted.The application can provide long-term reliable safety warning for deep foundation pit engineering.

Claims

1. A method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction from strain gauge arrays, characterized in that, include: The vibration signals of the nodes are acquired in parallel through dual channels, namely a static channel for monitoring long-term strain accumulation and a dynamic channel for capturing high-frequency vibrations of micro-fractures in the structure. Calculate the kurtosis value and high-frequency energy ratio in the node vibration signal, and report based on the change in the kurtosis value and high-frequency energy ratio and the forced reporting time; The difference between the reported real-time kurtosis value, high-frequency energy ratio and the background model baseline value is calculated. When the difference between the kurtosis difference and the high-frequency energy ratio both exceed the preset trigger threshold, it is marked as a suspected structural rupture event. For suspected structural rupture events, the original vibration waveforms of the corresponding node and adjacent nodes are obtained; Based on the original vibration waveforms of the corresponding nodes and adjacent array nodes, the orientation angle discrimination factor and the fracture impact index are calculated to determine the hazard level; Based on the determined level of danger, corresponding warning information is output, and the collection cycle, collection duration, and reporting strategy of the corresponding nodes are dynamically adjusted.

2. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The formula for calculating the kurtosis value is: in, The waveform is a vibration waveform. The mean, Standard deviation, This represents the number of sampling points; The formula for calculating the high-frequency energy ratio is: in, The power spectrum is obtained by fast Fourier transform of the vibration waveform. For frequency.

3. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The reporting is based on the change in the kurtosis value and the high-frequency energy ratio, and the forced reporting time, specifically as follows: The node saves the kurtosis value and high-frequency energy ratio of the last successfully reported value, and calculates the absolute change of the current kurtosis value from the last reported kurtosis value and the absolute change of the current high-frequency energy ratio from the last reported high-frequency energy ratio, respectively. When the absolute change in the kurtosis value is less than a preset first change threshold, the absolute change in the high-frequency energy ratio is less than a preset second change threshold, and the mandatory reporting period has not been reached, the node will not report. When the absolute change of any feature exceeds the corresponding preset threshold or reaches the mandatory reporting period, the node will report the real-time kurtosis value and high-frequency energy ratio.

4. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The background model baseline values ​​are the moving average of the kurtosis and the moving average of the high-frequency energy ratio during normal periods when the corresponding node does not trigger any warnings. The calculation formula is as follows: in, The moving average of the kurtosis. Let be the moving average of the high-frequency energy ratio, and α be the forgetting factor. Indicates the current moment. This indicates the last updated value.

5. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The number of corresponding nodes and adjacent nodes is 3.

6. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The distance between the corresponding node and its adjacent node satisfies the formula: in, The propagation velocity of elastic longitudinal waves in the steel structure supporting the foundation pit. The sampling frequency of the dynamic path vibration signal at the node. This is the main period of the vibration signal.

7. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The formula for calculating the orientation angle discrimination factor is as follows: in, , , The arrival time of the vibration waveform at the node is denoted as . This is the main period of the vibration signal; The formula for calculating the fracture impact index is as follows: in, The peak amplitudes of the waveforms at the three nodes are given. The average power spectral density in the high-frequency band of 100–500 Hz. The average power spectral density in the low-frequency range of 10–100 Hz. It is a constant.

8. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 6, characterized in that, The main period of the vibration signal is: in, The number of zero crossings of the waveform. The number of sampling points. This is the main period of the vibration signal.

9. The method for wireless safety monitoring and early warning of foundation pits based on vibration feature extraction of strain gauge array as described in claim 1, characterized in that, The specific steps for outputting corresponding level warning information based on the determined danger level, and dynamically adjusting the collection cycle, collection duration, and reporting strategy of the corresponding nodes are as follows: If a Level II high-risk warning is issued, an on-site audible and visual alarm will be triggered immediately, and detailed alarm information including node location, discrimination indicators, and estimated seismic source location will be sent to the monitoring backend. At the same time, the acquisition cycle of the corresponding array node will be shortened, the acquisition time will be extended, and characteristic data will be reported after each acquisition to strengthen the monitoring status. If it is determined to be a Level 1 warning, only a pop-up notification will be sent to the monitoring backend, and the collection cycle of the corresponding node will be shortened to half of the normal value. If no new suspected structural rupture events are triggered during the warning period, the corresponding node will automatically return to the normal low-power acquisition mode.

Images