Articles | Volume 7, issue 3
https://doi.org/10.5194/gi-7-235-2018
https://doi.org/10.5194/gi-7-235-2018
Research article
 | 
16 Aug 2018
Research article |  | 16 Aug 2018

Backpropagation neural network as earthquake early warning tool using a new modified elementary Levenberg–Marquardt Algorithm to minimise backpropagation errors

Jyh-Woei Lin, Chun-Tang Chao, and Juing-Shian Chiou

Cited articles

Abbena, E., Salamon, S., and Gray, A.: Modern Differential Geometry of Curves and Surfaces with Mathematica, Textbooks in Mathematics, 3rd Edn., 1016 pp., Chapman and Hall/CRC, 2006. 
Allen, R. M. and Kanamori, H.: The Potential for Earthquake Early Warning in Southern California”, Science, 300, 786–789, https://doi.org/10.1126/science.1080912, 2003. 
Arjun, C. R. and Kumar, A.: Artificial neural network-based estimation of peak ground acceleration, ISET Journal of Earthq. Tech., Paper No. 501, 46, 19–28, 2009. 
Bo, L., Wang, L., and Jiao, L.: Feature Scaling for Kernel Fisher Discriminant Analysis Using Leave-One-Out Cross Validation, Neural Computation, 18, 961–978, https://doi.org/10.1162/neco.2006.18.4.961, 2006. 
Böse, M., Wenzel, F., and Erdik, M.: PreSEIS: A Neural Network-Based Approach to Earthquake Early Warning for Finite Faults, B. Seismol. Soc. Am., 98, 366–382, https://doi.org/10.1785/0120070002, 2008. 
Download
Short summary
This BPNN approach with two alarms was well suited, and it was not necessary to consider the problems of characterising the wave phases and pre-processing, as stated previously. Furthermore, BPNN is a mature technology, which is expected to develop rapidly in the future, and does not require complex hardware. Determining an initial location and magnitude of the event was not necessary for this technique. An existing seismic monitoring network can be used.