【摘要】 本文选取1974年以来我国华北、西南地区15个发生过6级以上地震的地区时段和20个没有发生过6级以上地震的地区时段作为模式识别研究的对象,用地震空区、条带、b值、集中度、密集、平静、安全域等地震活动性方法构成问题征询表。采用修改的CORA-3算法进行模式识别。消除等效特征与弱特征,进行广泛的K阈值、对象及参数变化的控制试验。计算结果表明,识别较稳定,地震条带、b值、集中度、安全域等组成的地震活动性图象特征具有重要的前兆意义。综合模式识别能力明显高于任一单项方法。本文改进了一般的识别准则,采用综合决策方案,提高了模式识别结果的可信度。更多还原

【Abstract】 We have selected region-time intervals from 15 strong earthquakes ( M≥6) since 1974 in North and Southwest China and 20 earthquakes with M < 6 as the case study, using seismicity parameters to constitute the questionaire for pattern recognition. The calculation of recognition is carried out using the VAX 11/780 computer on modified CORA - 3 clustering algorithm. The equivalent and weaker traits have been removed and some control experiments have been completed.The results of pattern recognition using recent seismicity data show that the seismic band, b -value, concentration, safety area and so on implicate some important precursory meaning. We have adopted the synthetic decision method, so the responsibility of recognition results has been improved. The recognition results obtained in this paper are stable and successful.更多还原