دانلود رایگان مقاله انگلیسی پیش بینی زلزله با استفاده از شبکه های عصبی: نتایج و کار آینده - اشپرینگر 2006

Abstract.

In 1994, a new earthquake forecasting method was developed, that integrated in a neural network several forecasting tools that had been originally developed for financial analysis. This method was tested with the seismicity of the Azores, predicting the July, 1998, and the January, 2004, earthquakes, albeit within very wide time and location windows. Work is beginning to integrate physical precursors in the neural network, in order to narrow the forecasting windows.

5. Future Work

A group including researchers from the universities of Coimbra, Porto and Tr´as-os-Montes e Alto Douro (Vila Real) is beginning work on project DESIRE (Dynamic Evaluation of Seismic Risk) this time aimed at seismic forecasting in Continental Portugal. Three stations will be deployed near the three universities and monitor microseismicity, seismic waves’ velocities, water piezometry, ground self-potential, EM piezoelectric emissions and magnetic susceptibility, known earthquake precursors [20, 21, 22]. This area was chosen due both to its moderate seismicity – around 35 events with M ≥ 3.0 per year in the last 30 years, of which 71% had epicentres North of latitude 40◦ – and also because of the locations of the monitoring stations (Figure 4). The collected physical data will be added as input nodes to a neural network that is similar to the one that was described above, together with the pre-processed seismic catalogue data, this time using magnitudes instead of intensities. It is expected that this mixed approach will succeed in narrowing the time forecast window – that is, to achieve the goal of mid- to short-term prediction.