Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Seismic array performance analysis using a hybrid expert system

Technical Report ·
OSTI ID:6863570
Several factors contribute to the performance of a seismic array in locating events; configuration, frequency response, and distribution of sensor types are all important. Statistical analysis of the effect of these multiple variables on event location accuracy requires tedious repeated runs of analysis codes, using different input data sets and parameters. A hybrid expert system is an ideal tool to aid in processing and analyzing massive amounts of data and creating and manipulating a data base. This relatively new concept is a coupling of artificial intelligence (AI) programming techniques with analysis codes. Our expert system combines the individual strengths of an AI machine (Symbolics 3610) and conventional microcomputer (Micro VAX II). Code written in Lisp on the Symbolics provides the user interface, maintains the data base and runs the desired FORTRAN codes, which reside on the Micro VAX II. Functions are provided for entering data into the data base, analyzing data sets with respect to multiple variables, and inspecting the results in graphical form. The system is easily expandable to include new analysis methods. We demonstrate the capabilities of the hybrid expert system, ASSISTANT, using teleseismic data from the NORESS and ARCESS seismic arrays. The effects of array configuration, event set, frequency band, and location algorithm on the event location accuracy are illustrated using several examples. 11 refs., 18 figs.
Research Organization:
Sandia National Labs., Albuquerque, NM (USA)
DOE Contract Number:
AC04-76DP00789
OSTI ID:
6863570
Report Number(s):
SAND-88-2410; ON: DE89004665
Country of Publication:
United States
Language:
English