回顾了岩溶塌陷危险性评价研究进展,分析了鞍山市区岩溶塌陷的主要影响因素,选取了“盖层结构、盖层厚度、含砾石粘土厚度、构造影响距、构造组合、灰岩富水情况、降落漏斗影响距、地下水径流强度、基岩类型、岩溶填充率、线岩溶率、居民点分布和人口密度”作为岩溶塌陷危险性评价指标;结合实际情况阐述了人工神经网络的应用过程,得到了每个指标的权重;利用地理信息系统作为评价手段对鞍山市岩溶塌陷危险性进行了区划。通过与历史岩溶塌陷点分布图的比对,验证了区划结果是可信的。 Reviewing the research progress of karst collapse risk assessment, the main influencing factors of karst collapse in Anshan urban area are analyzed. The structure of cover layer, thickness of cover layer, thickness of gravel clay, structural influence distance, structural combination, , The influence of the falling funnel, the groundwater runoff intensity, the type of bedrock, the karst filling rate, the line karst, the residential distribution and the population density were evaluated as the evaluation indexes of karst collapse risk. According to the actual situation, the application process of artificial neural network was expounded, Get the weight of each index; using geographic information system as an evaluation means to carry out the zoning of Anshan karst collapse risk. By comparing with the historical karst collapsible point distribution map, it is verified that the zoning results are credible.