- An advanced AI system that can isolate and identify a range of seismic signals from historical and continuous data.
- Convolutional neural networks will be used to extract features from seismographs and recurrent neural networks will then combine memory and inputs to improve the accuracy of its predictions. It learns the sequential characteristics of seismographs.
- To train and validate the earthquake-detecting AI system, this spatio-temporal data will be used.
- The robust network will be able to predict earthquake signals no matter whether the seismic event was large, small, local, or contained a high degree of background noise.
- Once the network is trained, it can be applied to a stream of seismic data in real time.
- False positive rates are minimal due to the high-resolution modeling of earthquake signals based on their spectral structure
We estimate based on the research work (stated below) that our model will be capable of detecting more than 700 microearthquakes induced far away from the training region.
This will be government enabled solution which will give warning via sms to all the residents of remote/non-remote areas. The mobile numbers are already with the government through Aadhar linkage.
The solution can be easily scaled to multiple sensors and could perform real-time monitoring in active tectonic zones or serve as the foundation of an early earthquake warning system.
The solution builds on the research work by Harvard and Google, which created an AI model capable of predicting the location of aftershocks up to one year after a major earthquake.