IIT Madras team comes up with new approach to accurately detect seismic activity


IIT Madras researchers have developed a new approach to accurately detect seismic activity and save lives.

Researchers at the Indian Institute of Technology (IIT) Madras have developed a new approach to accurately detect earthquakes and launch within 30 seconds to 2 minutes after receiving the first warning signals to save precious human lives and property.
They said that time, although it seems like a minute, is enough to shut down nuclear reactors, vehicles like the subway, or even elevators in high-rise buildings to save countless lives, they said Thursday.
They argued that this approach can also be accurately used for fault detection and isolation in other areas. “Information about the arrival of P-waves is critical for determining other parameters of the event source, such as the magnitude, depth and location of the epicenter.
Therefore, a solution to the problem of P-wave detection that is reliable, accurate and accurate is necessary to correctly assess the details of an event and reduce damage caused by an earthquake or other triggered events, ”says Kanchan Aggarwal, Ph.D. IIT, Madras, who conducted research under the direction of Professor Arun K. Tangural, Department of Chemical Engineering of the leading institute.
Highlighting the practical applications of this study, funded in part by the Nuclear Science Research Council, an advisory body to the Department of Atomic Energy, Professor Tangirala said the proposed structure is not necessarily limited to seismic event detection, but is general in nature and can also be used to detect and isolate errors in other domains. “In addition, the framework can include any predictive models, including machine learning and deep learning models, that will reduce human interference with detection,” said Tanguirala, an expert in time series analysis, multiscale signal analysis, system identification, and estimation theory among other fields. , is talking.
The results of their research have been published in open access scientific journals PLOS ONE. The proposed approach provides a wide range of possibilities for accurate detection and occurrence of the P-wave, especially in low signal-to-noise ratio (SNR) conditions, which cannot be achieved by all existing methods. The unique aspects of this new solution approach compared to existing approaches are that it is commensurate with noise performance, resulting in minimal sensitivity to outliers or reliable detection, offers more flexible bandwidth selection by decomposing both lower and higher frequencies. in each. level, which leads to accurate detection, the message says.
This allows the user to discard noise in unwanted time-frequency bands, resulting in an improved signal-to-noise ratio. Thus, an accurate determination of the beginning of the P wave is achieved. The paper proposes a new automatic detector and collector of P-waves in real time in a forecasting structure with a time-frequency localization function. This approach is not necessarily limited to detecting seismic events, but is rather general as it can be used for fault detection in other areas as well, added in the release.
Another positive offshoot of the proposed structure is that it facilitates the reconstruction of the P wave after its detection. This aspect has not been investigated by colleagues. “The earth is constantly in a state of unease, mainly due to waves in the ocean, changes in the earth’s crust, atmospheric fluctuations and human activities,” he added.
Seismic signals can be generated by both natural (earthquakes, volcanic eruptions, tsunamis, etc.) and artificial (heavy traffic, nuclear explosions, mining, etc.) Sources. Regardless of the source, all seismic events release energy proportional to their magnitude. This released energy propagates in all directions in the form of a wave and is recorded by a seismometer as a seismic signal. Seismograms are records of the vibrational motion of the ground taken with a seismometer. Analyzing these seismograms is vital to understanding earth activity and building earthquake early warning systems, locating sources, and detecting sources of other seismic events.
Detecting or predicting the destructive portion of an earthquake goes a long way in protecting lives and preventing property loss, they said.

Close history


Please enter your comment!
Please enter your name here