Abstract: For large-scale logging data in petroleum mining areas,a small number of wells(i.e.,standard wells)can be effectively selected to conduct manual matching with high precision,then supervised automatic or semi-automatic matching of global wells can be performed,which is key step to obtain an accurate interpretation of geological structures.However,the selection of standard wells is a complex process,which is closely related to the spatial distribution and the geological characteristics of wells.Therefore,this paper synthetically considers the wells'geographical locations and various well-logging attributes,and proposes a visual analysis method of standard well selection based on blue noise sampling.First,according to the geographical locations of wells,a blue noise sampling model is used to adaptively determine sampling rate and range of standard wells.Second,within the local sampling range of the standard well,a dynamic planning based stratigraphic matching algorithm is designed to calculate the wells’multi-dimensional attribute difference,which measures the geological characteristic similarity between wells.MDS is further introduced to project the matching relationship of wells,further visualize the spatial distribution and multidimensional attribute differences,and support the automatic or interactive selection of standard wells.Then,an attribute view and a matrix view are designed to intuitively display the original multidimensional well-logging data and matching relationship of wells,and guide the domain experts to conduct in-depth exploration and iterative optimization of the standard well selection.Finally,a visual analysis system of multidimensional standard well selection integrating a convenient interaction model based on blue noise sampling is developed to help users interactively explore and analyze multidimensional well-logging data.On the basis of comprehensive consideration of the spatial distribution and multidimensional attribute information of wells,this system effectively selects a number of representative standard wells,and provides accurate and reliable data materials and experience support for subsequent geological structure interpretation.A large number of experimental results further validate the effectiveness and practicability of the proposed method.