جستجوی مقالات مرتبط با کلیدواژه "fuzzy inference system" در نشریات گروه "زمین شناسی"

The Iranian plateau is located in the high seismicity belt. Earthquake can inflict severe loss of life and property, especially when they occur in densely populated areas. Therefore, seismic hazard evaluation is very essential to prevent the harmful effects. The region of the study is located in the northwest of Iran, between 43°-50° E longitude and 35.5°-40.5° N latitude. This city which is located in the center of East Azerbaijan province, has been ruined by terrible earthquakes, which is due to the presence of active faults in the region. Seismic hazard assessment similar to other seismology researches is very complicated due to the effect of different parameters in an earthquake occurring with uncertainty. The amount of uncertainty should be considered in a rational way. The fuzzy method is a suitable method that is used as a decision-making method for solving problems and modeling uncertainties and ambiguities. We used a fuzzy inference system, as the practice is based on uncertainty estimation of seismic hazard for Tabriz region. Peak ground Acceleration value is estimated for fuzzy Logic System in deterministic method 0.55g which is obtained from a seismic source with a Mmax=8.0 at a distance of  36.98 km of Tabriz city.The contour map of the peak ground acceleration throughout Tabriz city can help in urban planning.

The purpose of this study was to develop a model for the estimation of rock mass classification of Sarvak limestone in the Bakhtiari dam site, south-west (SW) Iran. Q system had been used as the starting point for the rock mass classification. This method was modified for sedimentary rock mass which is known as Qsrm. Because Qsrm considers a wide range of rock mass properties, it has become a tool for rock mass classification that more correlates with geophysical parameters. This study tried to revise and empower the correlation between P-wave velocity (Vp) with Q and Qsrm in Sarvak limestone. By using data sets of Bakhtiari Dam Site (BDS) in SW Iran and multivariate regression and the Fuzzy Inference System (FIS), models were rendered for prediction of Q and Qsrm. About 700 sets of data were used for modeling and Vp was considered as the input parameter. The regression equations showed the relationship between Vp with Q and Qsrm,under conditions of quadratic relations, obtained coefficients of determination (R2) of 0.49 and 0.66, respectively. The correlation coefficient was calculated as 0.82 for the Qsrm obtained from FIS models. Also, Variance Accounted For (VAF) and Root Means Square Error (RMSE) indexes were also used for evaluation of prediction accuracy of models. Results showed that Vp has better performance in prediction of Qsrm than Q and theFIS model showed the best prediction results. Because these models have accuracy, they could be used in similar conditions.