Identification of Potential Strong Earthquake Source Areas in Western South Tianshan-Pamir Region of Xinjiang Based on MATLAB Radial Basis Function Neural Network
The study area is located in the frontal zone of the collision between the Indian Plate and the Eurasian Plate, characterized by extremely complex geological structures and frequent seismic activity. Accurate identification of potential strong earthquake source zones in this region and quantification of earthquake occurrence probabilities are crucial for enhancing earthquake prevention and disaster reduction capabilities, as well as for optimizing seismic hazard assessments. Based on the MATLAB radial basis function neural network classification algorithm, and following the principles of seismic tectonic analogy and historical earthquake recurrence, eight geological structural features related to seismic activity were selected as input variables, with the historical maximum magnitude as the output variable, to construct a neural network model. A total of 142 samples were used for model training and testing. By adjusting key parameters such as target error, spread of the radial basis function, and maximum number of neurons, a stable and feasible network model was achieved.This model was then applied to 395 samples to be classified. The output results, combined with manual verification, ultimately formed a zoning scheme for potential strong earthquake sources. Compared with schemes derived from Kriging interpolation and kernel density estimation methods, the proposed zoning scheme shows strong consistency with Seismic Ground Motion Parameter Zonation Map of China (GB18306—2015).This validates the reliability of the method for seismic hazard assessment in complex tectonic regions and its adaptability to small sample datasets. The results provide valuable references for the planning of earthquake prevention and disaster reduction, as well as for seismic risk management.
FAN Wen-long, YAO Yuan, LI Gui-rong, LI Dong, SUN Zhao-jie, , .
Identification of Potential Strong Earthquake Source Areas in Western South Tianshan-Pamir Region of Xinjiang Based on MATLAB Radial Basis Function Neural Network[J].
Science Technology and Engineering, 2026, 26(14): 5844-5854 DOI:10.12404/j.issn.1671-1815.2506312
强震潜在震源区识别的核心是通过分析构造特征与地震活动的关联性,预测未来可能发生强震的潜在区域及对应震级概率。当前,学术界普遍使用的识别方法包括地质构造解析法、历史地震统计法及数值模拟法等,但由于该区域构造复杂,断裂空间展布多样,且部分隐伏断裂的地表活动证据不足,仅依靠定性分析方法难以准确刻画潜在震源区的空间分布[6-7]。近年来,随着机器学习技术的发展,神经网络因其具备强大的非线性映射能力和数据挖掘优势,在地震预测与震源区识别等领域的应用逐渐深入。径向基神经网络(radial basis function neural network,RBFNN)具备非线性逼近能力强、收敛速度快等优势,在小样本、高维度数据的分类问题中表现出良好的适应性,为潜在震源区的量化识别提供了新的技术手段[8-10]。
ZhengWenjun, ZhangPeizhen, YuanDaoyang, et al. Basic characteristics of active structures in Chinese mainland and their control over regional dynamic processes[J]. Chinese Journal of Geomechanics, 2019, 25(5): 699-721.
WuYuxiang, ZhangMin, XiaLiuqing. Compositional characteristics of tricyclic terpanes in crude oil from the Dayanqi Oilfield, Tarim Basin and their geological significance[J]. Science Technology and Engineering, 2025, 25(26): 11029-11041.
HuZongkai. Late Quaternary crustal shortening of the Bole Basin and its role in the tectonic deformation of the Tianshan Mountains[D]. Beijing: Institute of Geology, China Earthquake Administration, 2023.
NieWenyu, FanXiwei, LiHuayue, et al. Risk analysis of casualties under the recurrence of the 1902 Artux MS8.0 earthquake: a case study of Kashgar City and Artux City[J]. Seismology and Geology, 2020, 47(2): 627-648.
XuJiajing, HanGuangjie, XiNan, et al. Preliminary analysis of focal parameters of the 7.1-magnitude Wushi earthquake in Xinjiang on January 23, 2024[J]. Earthquake in China, 2024, 40(1): 260-270.
ZhengXingqun, ZhaoBoming. Research on the attenuation relationship of peak acceleration of ground motion in Southwest China based on seismology[J]. Journal of Vibration and Shock, 2025, 44(5): 191-196.
ChenTao, ZhouBengang, MinWei, et al. Analysis of the segmented nature and maximum potential seismic Force of the Talwan-Dengshan-Chijia Ciwo Fault[J]. Advances in Earthquake Science, 2021, 51(7): 315-319.
ZhouGuanqun, LuoShiling, ZhangWeixin, et al. Identification and classification of microseismic signals based on multimodal EWT-CNN[J]. Science Technology and Engineering, 2025, 25(30): 12786-12794.
JiaQiang, LuoQiyao, ZhaoChao, et al. Application of the sheorey model combined with PSO-BP neural network for high-precision prediction of in-situ stress state in hot spring mine: a case study of the Qinghe Hot Spring Mine Area[J]. Science Technology and Engineering, 2025, 25(26): 11042-11050.
ShangYue, MaPengfei, LüYiqing, et al. Evaluation of geological disaster susceptibility based on coupling of information value and machine learning[J]. Science Technology and Engineering, 2025, 25(23): 9713-9724.
XieChenghao, LiYong, LiuWei, et al. Geological structural characteristics and ore-controlling structure research of the Xili Mine west of 509 Highway Maintenance Station in Hetian County, Xinjiang[J]. Northwestern Geology, 2025, 58(4): 234-249.
ZhangJian, KongYanlong. Thermal-rheological characteristics of the lithospheric mantle in the Pamir Plateau and thermal structure analysis of its northeastern margin: one of the geothermal research achievements of the Third comprehensive scientific expedition to Xinjiang[J]. Chinese Journal of Geophysics, 2020, 68(1): 199-212.
JiaYuheng, LiuXifeng, LiuYan, et al. Study on the genesis of the Qiemo Jade Deposit in Xinjiang[J]. Chinese Journal of Petrology and Mineralogy, 2018, 37(5): 824-838.
DongYunpeng, HuiBo, WangBo, et al. Transform faults and transfer faults: tectonic dynamic transformation from plate boundaries to intracontinental settings[J]. Science China Earth Sciences, 2025, 55(12): 4014-4063.
XuJianhong, ChenJie, LiTao, et al. Activity characteristics and formation mechanism of bending moment normal faults in the Qianguman Anticline Zone of the West Kunlun Mountains[J]. Seismology and Geology, 2020, 47(2): 405-428.
[31]
PowellM J D. Radial basis functions for multivariable interpolation: a review[J]. Algorithms for Approximation, 1987(1): 143-167.
MaQiang, ZhangYang, YangKe. Adaptive sliding mode control for second-order uncertain systems based on RBF neural networks[J]. Modern Defence Technology, 2026, 54(1): 156-164.
ZengRui, ShaoYanlin, HuangYu, et al. Intelligent lithology identification from well logs based on deep learning[J]. Science Technology and Engineering, 2025, 25(30): 12804-12812.
LengChongbiao, ZhangHui, KangBo, et al. Denoising of fixed-point deformation signal based on ICEEMDAN-SST: a case study of Yichang Seismic Station[J]. Science Technology and Engineering, 2025, 25(25): 10579-10585.