技术领域Technical field
本发明涉及数据处理技术领域，具体涉及一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法。The invention relates to the technical field of data processing, and specifically relates to a method for calculating side friction resistance of coral reef sand foundation pipe piles.
背景技术Background technique
珊瑚礁地质勘测过程需要通过插入珊瑚礁砂内的基础管桩体现出的基本信息对地质数据进行采集评估。因为珊瑚礁地质环境复杂，地质数据采集难度较大，根据管桩采集到的地质数据具有较多的缺失数据或者异常数据，影响后续对管桩测摩擦阻力的计算。因此为了保证后续计算结果的准确性，需要对管桩采集到的地质数据进行插值处理，通过反距离权重插值等插值方法获得地质数据中需要插值位置处的插值数据。The coral reef geological survey process requires the collection and evaluation of geological data through the basic information reflected by the foundation pipe piles inserted into the coral reef sand. Because the geological environment of the coral reef is complex, it is difficult to collect geological data. The geological data collected from the pipe piles has a lot of missing data or abnormal data, which affects the subsequent calculation of the friction resistance of the pipe piles. Therefore, in order to ensure the accuracy of subsequent calculation results, it is necessary to interpolate the geological data collected by the pipe piles, and obtain the interpolation data at the required interpolation positions in the geological data through interpolation methods such as inverse distance weighted interpolation.
现有技术中对数据进行插值时，插值数据的结果和精度依赖于样本的密度和分布，如果样本分布不均匀或者数量较少，会导致插值数据的误差较大，准确性较低，而对于珊瑚礁砂中部署的基础管桩采集到的数据而言，为了保证数据采集的效率，采集到的地质数据的样本较少且密度较低，使用现有技术中的插值处理会导致插值数据的不准确，进而影响后续侧摩阻力的计算的准确度。When interpolating data in the prior art, the results and accuracy of the interpolated data depend on the density and distribution of the samples. If the samples are unevenly distributed or have a small number, the interpolation data will have larger errors and lower accuracy. For For data collected from foundation pipe piles deployed in coral reef sand, in order to ensure the efficiency of data collection, the geological data collected has fewer samples and a lower density. Using the interpolation processing in the existing technology will lead to inconsistencies in the interpolated data. Accurate, which in turn affects the accuracy of subsequent side friction calculations.
发明内容Contents of the invention
为了解决现有技术对珊瑚礁砂地质数据进行插值的结果不准确，进而影响后续侧摩阻力计算的技术问题，本发明的目的在于提供一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法，所采用的技术方案具体如下：In order to solve the technical problem of inaccurate interpolation results of coral reef sand geological data in the prior art, which in turn affects the subsequent side friction calculation, the purpose of the present invention is to provide a method for calculating the side friction of coral reef sand foundation pipe piles. The technical solutions adopted are as follows:
本发明提出了一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法，所述方法包括：The present invention proposes a method for calculating the side friction resistance of coral reef sand foundation pipe piles. The method includes:
获取珊瑚礁砂中部署的基础管桩上每个采样位置处的地质数据；所述地质数据至少包括砂粒直径和管桩接触面积；获得所述地质数据中的插值位置；Obtain geological data at each sampling position on the foundation pipe pile deployed in the coral reef sand; the geological data at least includes the diameter of the sand grain and the pipe pile contact area; obtain the interpolated position in the geological data;
以每个所述采样位置作为图结构节点构建初始图结构，所述初始图结构中每个所述图结构节点的节点值为对应所述采样位置的地质数据；在预设数量个不同聚类尺度下对所述初始图结构中的图结构节点进行的聚类，获得每个所述聚类尺度下的聚类类别；Construct an initial graph structure with each sampling position as a graph structure node, and the node value of each graph structure node in the initial graph structure is the geological data corresponding to the sampling position; in a preset number of different clusters Cluster the graph structure nodes in the initial graph structure under the scale to obtain the clustering category under each of the clustering scales;
获得每个所述聚类类别的节点结构特征；获得每个所述插值位置在不同所述聚类尺度下所处所述聚类类别之间的所述节点结构特征的匹配相似度；根据所述匹配相似度获取每个所述插值位置在不同所述聚类尺度之间的关联聚类类别集合；在所述关联聚类类别集合中，以中心类别与每个聚类类别之间的匹配相似度作为每个聚类类别对应所述聚类尺度下的权重；Obtain the node structure characteristics of each clustering category; obtain the matching similarity of the node structure characteristics between the clustering categories where each interpolation position is located at different clustering scales; according to the The matching similarity is used to obtain a set of associated clustering categories for each interpolation position between different clustering scales; in the set of associated clustering categories, the matching between the center category and each clustering category is The similarity is used as the weight of each clustering category corresponding to the clustering scale;
获得所述插值位置在所述关联聚类类别集合中每个聚类类别对应聚类尺度下的插值数据，根据对应所述聚类尺度下的所述权重和所述插值数据进行整合，获得对应所述插值位置的最终插值数据；获得所有所述插值位置的所述最终插值数据，获得增强地质数据；Obtain the interpolation data at the clustering scale corresponding to each clustering category in the associated clustering category set at the interpolation position, integrate the weight and the interpolation data corresponding to the clustering scale, and obtain the corresponding Final interpolation data of the interpolation positions; obtaining the final interpolation data of all the interpolation positions to obtain enhanced geological data;
根据所述增强地质数据中的数据获得珊瑚礁砂基础管桩的侧摩阻力。The side friction resistance of the coral reef sand foundation pipe pile is obtained based on the data in the enhanced geological data.
进一步地，所述以每个所述采样位置作为节点构建初始图结构，包括：Further, constructing an initial graph structure with each sampling position as a node includes:
根据所述采样位置的坐标，采用三角剖分法建立拓扑图结构，获得所述初始图结构。According to the coordinates of the sampling position, a triangulation method is used to establish a topological graph structure, and the initial graph structure is obtained.
进一步地，采用拉普拉斯图聚类的聚类方法进行迭代聚类，获得不同所述聚类尺度下的所述聚类类别。Further, a clustering method of Laplacian graph clustering is used to perform iterative clustering to obtain the clustering categories at different clustering scales.
进一步地，所述获得每个所述聚类类别的节点结构特征包括：Further, obtaining the node structure characteristics of each cluster category includes:
对所述聚类类别进行图因子分解，获得分解子图；通过最小哈希法获得每个所述分解子图的二进制码，以所有所述分解子图的所述二进制码的集合作为所述节点结构特征。Perform graph factor decomposition on the clustering categories to obtain decomposed subgraphs; obtain the binary code of each decomposed subgraph through the minimum hash method, and use the set of binary codes of all decomposed subgraphs as the Node structural characteristics.
进一步地，所述获得每个所述插值位置在不同所述聚类尺度下所处所述聚类类别之间的所述节点结构特征的匹配相似度包括：Further, obtaining the matching similarity of the node structural features between the clustering categories where each interpolation position is located at different clustering scales includes:
将所述插值位置在不同所述聚类尺度下所处所述聚类类别之间的所述节点结构特征进行KM匹配；所述KM匹配过程中以两个所述聚类类别之间的每个所述分解子图作为匹配节点，以所述分解子图之间的所述二进制码的余弦相似度作匹配值，将所述匹配值大于预设匹配阈值的匹配节点组作为保留节点组；以所述保留节点组的数量与两个所述聚类类别中的最大分解子图数量的比值作为所述匹配相似度。KM matching is performed on the node structural characteristics between the clustering categories where the interpolation position is located at different clustering scales; in the KM matching process, each node between the two clustering categories is used. Each of the decomposed subgraphs is used as a matching node, the cosine similarity of the binary codes between the decomposed subgraphs is used as a matching value, and the matching node group whose matching value is greater than the preset matching threshold is used as a retained node group; The ratio of the number of retained node groups to the maximum number of decomposed subgraphs in the two clustering categories is used as the matching similarity.
进一步地，所述根据所述匹配相似度获取每个所述插值位置在不同所述聚类尺度之间的关联聚类类别集合包括：Further, obtaining a set of associated clustering categories for each interpolation position between different clustering scales based on the matching similarity includes:
若所述匹配相似度大于预设相似度阈值，则对应的两个所述聚类类别存在关联关系；获得所述插值位置在不同所述聚类尺度下对应的所有聚类类别之间的关联关系，将具有关联关系的所述聚类类别作为关联聚类类别，获得关联聚类类别集合。If the matching similarity is greater than the preset similarity threshold, then there is an association between the two corresponding clustering categories; the association between all clustering categories corresponding to the interpolation position at different clustering scales is obtained relationship, use the clustering categories with associated relationships as associated clustering categories, and obtain a set of associated clustering categories.
进一步地，所述获得所述插值位置在所述关联聚类类别集合中每个聚类类别对应聚类尺度下的插值数据包括：Further, obtaining the interpolation data of the interpolation position at the clustering scale corresponding to each clustering category in the associated clustering category set includes:
根据所述插值位置所处所述聚类类别中的所述图结构节点，利用反距离插值法获得所述插值位置在对应聚类尺度下的所述插值数据。According to the graph structure node in the clustering category where the interpolation position is located, the inverse distance interpolation method is used to obtain the interpolation data of the interpolation position at the corresponding clustering scale.
进一步地，所述最终插值数据的获取方法包括：Further, the method for obtaining the final interpolation data includes:
将所述插值位置对应的每个所述插值数据与对应所述权重相乘，获得对应聚类尺度下的调整插值数据；将所述调整插值数据求平均，获得所述最终插值数据。Multiply each interpolation data corresponding to the interpolation position by the corresponding weight to obtain the adjusted interpolation data at the corresponding clustering scale; average the adjusted interpolation data to obtain the final interpolation data.
本发明具有如下有益效果：The invention has the following beneficial effects:
本发明考虑到珊瑚礁地质数据在单一尺度上具有样本少分布离散的问题，通过设置多尺度的聚类过程获得每个尺度下的聚类结果，通过分析插值位置在每个聚类尺度下对应的聚类类别的相关性，提高插值结果的准确性和稳定性。本发明具体通过构建图结构的方法将每个采样位置下的地质数据进行参数空间转换，进而方便后续的多尺度聚类。进一步通过每个分析不同聚类尺度下聚类类别之间的节点结构特征的匹配相似度表征不同尺度之间的自相关性，获得每个插值位置对应的关联聚类类别集合，因为关联聚类类别集合中每个关联聚类类别都对应一个聚类尺度，因此可通过关联聚类类别集合的中心类别与其他类别之间的匹配相似度获得其他类别对应聚类尺度下的权重。通过权重即可对每个尺度下的插值数据进行加权求和，获得准确的最终插值数据，实现对地质数据的增强。通过增强地质数据即可实现准确的侧摩阻力的计算。This invention takes into account the problem that coral reef geological data has a small number of discrete samples at a single scale, obtains clustering results at each scale by setting up a multi-scale clustering process, and analyzes the interpolation positions corresponding to each clustering scale. The correlation of clustering categories improves the accuracy and stability of interpolation results. Specifically, the present invention converts the geological data at each sampling location into parameter space by constructing a graph structure, thereby facilitating subsequent multi-scale clustering. Further, the autocorrelation between different scales is represented by the matching similarity of the node structure characteristics between clustering categories under each analysis of different clustering scales, and the set of associated clustering categories corresponding to each interpolation position is obtained, because associated clustering Each associated clustering category in the category set corresponds to a clustering scale, so the weights of other categories corresponding to the clustering scale can be obtained through the matching similarity between the central category of the associated clustering category set and other categories. Through weighting, the interpolation data at each scale can be weighted and summed to obtain accurate final interpolation data and enhance geological data. Accurate side friction calculations can be achieved by augmenting geological data.
附图说明Description of the drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案和优点，下面将对实施例或现有技术描述中所需要使用的附图作简单的介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它附图。In order to more clearly explain the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description The drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts.
图1为本发明一个实施例所提供的一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法流程图。Figure 1 is a flow chart of a method for calculating the side friction resistance of coral reef sand foundation pipe piles provided by one embodiment of the present invention.
具体实施方式Detailed ways
为了更进一步阐述本发明为达成预定发明目的所采取的技术手段及功效，以下结合附图及较佳实施例，对依据本发明提出的一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法，其具体实施方式、结构、特征及其功效，详细说明如下。在下述说明中，不同的“一个实施例”或“另一个实施例”指的不一定是同一实施例。此外，一或多个实施例中的特定特征、结构或特点可由任何合适形式组合。In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended purpose of the invention, the following is a calculation method for the side friction resistance of coral reef sand foundation pipe piles proposed according to the present invention in conjunction with the drawings and preferred embodiments. Its specific implementation, structure, characteristics and efficacy are described in detail as follows. In the following description, different terms "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Additionally, the specific features, structures, or characteristics of one or more embodiments may be combined in any suitable combination.
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which the invention belongs.
下面结合附图具体的说明本发明所提供的一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法的具体方案。A specific scheme for calculating the side frictional resistance of coral reef sand foundation pipe piles provided by the present invention will be described below in detail with reference to the accompanying drawings.
请参阅图1，其示出了本发明一个实施例提供的一种适用珊瑚礁砂基础管桩侧摩阻力的计算方法流程图，该方法包括：Please refer to Figure 1, which shows a flow chart of a method for calculating the side friction of coral reef sand foundation pipe piles provided by one embodiment of the present invention. The method includes:
步骤S1：获取珊瑚礁砂中部署的基础管桩上每个采样位置处的地质数据；地质数据至少包括砂粒直径和管桩接触面积；获得地质数据中的插值位置。Step S1: Obtain geological data at each sampling position on the foundation pipe pile deployed in the coral reef sand; the geological data at least includes the diameter of the sand grain and the contact area of the pipe pile; obtain the interpolated position in the geological data.
珊瑚礁砂中部署的基础管桩上包含具有多种传感器的探头，通过探头可采集多种地质数据，通过管桩的深入可采集珊瑚礁上不同深度地层的地质数据。探头中传感器采集到的数据通过传输以及量化即可完成对地质数据的采集，具体数据采集方式和传输量化方式为本领域技术人员熟知的技术手段，在此不做赘述。The foundation pipe piles deployed in the coral reef sand contain probes with multiple sensors. Various geological data can be collected through the probes, and geological data of strata at different depths on the coral reef can be collected through the depth of the pipe piles. The data collected by the sensor in the probe can complete the collection of geological data through transmission and quantification. The specific data collection method and transmission quantification method are technical means well known to those skilled in the art, and will not be described in detail here.
因为珊瑚礁砂对于管桩的侧摩阻力能够体现出当前地质的紧致程度、砂砾形态特点等多种特征，而侧摩阻力的计算主要通过砂粒直径和管桩接触面积进行计算，因此所采集到的地质数据至少包括砂砾直径和管桩接触面积。在本发明一些实施例中还可包括土层厚度、堆积密度、摩擦角等多种参数，可通过具体实施场景具体进行采集。需要说明的是，每个采样位置处均可获得多种地质数据，每种地质数据的处理方法均是相同的，因此在后续表述中均以一种地质数据进行举例说明，且统称为地质数据。Because the side friction resistance of coral reef sand to pipe piles can reflect the tightness of the current geology, the characteristics of the gravel shape and other characteristics, and the calculation of side friction resistance is mainly based on the diameter of the sand grains and the contact area of the pipe piles, so the collected The geological data include at least gravel diameter and pipe pile contact area. In some embodiments of the present invention, various parameters such as soil layer thickness, packing density, friction angle, etc. may also be included, which can be specifically collected through specific implementation scenarios. It should be noted that a variety of geological data can be obtained at each sampling location, and the processing method of each geological data is the same. Therefore, in subsequent statements, one type of geological data is used as an example, and is collectively referred to as geological data. .
采集到的地质数据因为采集环境以及采集操作的原因会导致数据中出现缺失数据，缺失数据的位置即为需要进行插值处理的插值位置。需要说明的是，插值位置的获取可直接通过数据采集时数据的时间戳或者坐标信息的连续性体现出来，也可通过现有异常数据检测算法获得，具体方法为本领域技术人员熟知的技术手段，在此不做赘述。The collected geological data will cause missing data in the data due to the collection environment and collection operations. The location of the missing data is the interpolation location that needs to be interpolated. It should be noted that the interpolation position can be obtained directly through the time stamp of the data during data collection or the continuity of the coordinate information, or through the existing abnormal data detection algorithm. The specific method is a technical means well known to those skilled in the art. , will not be described in detail here.
步骤S2：以每个采样位置作为图结构节点构建初始图结构，初始图结构中每个图结构节点的节点值为对应采样位置的地质数据；在预设数量个不同聚类尺度下对初始图结构中的图结构节点进行的聚类，获得每个聚类尺度下的聚类类别。Step S2: Construct an initial graph structure with each sampling location as a graph structure node. The node value of each graph structure node in the initial graph structure is the geological data corresponding to the sampling location; perform the initial graph at a preset number of different clustering scales. Clustering is performed on the graph structure nodes in the structure to obtain clustering categories at each clustering scale.
珊瑚礁地质数据具有样本少密度低的数据特征，在单一尺度上分析容易造成插值数据的不准确。考虑到珊瑚礁地质数据中的一个位置处的数据在不同尺度上的聚类结果可能具有自相关性，即小尺度下的数据结构类似与大尺度下的数据结构，通过获取自相关性的聚类类别能够增加插值过程中的参考数据，增强插值结果的准确性和稳定性。因此需要对采集到的地质数据进行多尺度下的聚类分析，聚类尺度越多，则对应的聚类结果中包含的样本数量越多。考虑到珊瑚礁地质数据的检测需要重点关注数据的采样位置，即在聚类过程中应同时考虑到数据采样位置以及对应的地质数据，因此以每个采样位置作为图结构节点，构建初始图结构，在初始图结构中每个图结构节点的节点值为对应采样位置的地质数据。在后续聚类过程中在初始图结构上进行聚类操作，能够同时考虑到地质数据的具体数据值以及采样位置，使得获得的聚类结果中的样本点之间参考性更强，所获得的插值结果的参考性更强。Coral reef geological data has the characteristics of small samples and low density, and analysis at a single scale can easily lead to inaccuracies in interpolation data. Considering that the clustering results of data at a location in coral reef geological data at different scales may have autocorrelation, that is, the data structure at a small scale is similar to the data structure at a large scale, by obtaining the clustering of autocorrelation Categories can add reference data during the interpolation process and enhance the accuracy and stability of the interpolation results. Therefore, it is necessary to perform cluster analysis on the collected geological data at multiple scales. The more clustering scales, the more samples are included in the corresponding clustering results. Considering that the detection of coral reef geological data needs to focus on the sampling location of the data, that is, the data sampling location and the corresponding geological data should be considered at the same time during the clustering process. Therefore, each sampling location is used as a graph structure node to construct the initial graph structure. In the initial graph structure, the node value of each graph structure node is the geological data corresponding to the sampling location. In the subsequent clustering process, the clustering operation is performed on the initial graph structure, and the specific data values and sampling locations of the geological data can be taken into consideration at the same time, making the sample points in the obtained clustering results more referential. The interpolation results are more informative.
优选地，在本发明一个实施例中，以每个所述采样位置作为节点构建初始图结构，包括：根据采样位置的坐标，采用三角剖分法建立拓扑图结构，获得初始图结构。Preferably, in one embodiment of the present invention, using each sampling position as a node to construct an initial graph structure includes: using the triangulation method to establish a topological graph structure according to the coordinates of the sampling position to obtain the initial graph structure.
需要说明的是，三角剖分法为本领域技术人员熟知的技术手段，具体过程不再赘述。在本发明其他实施例中也可通过泰森多边形构建算法等其他图结构构建算法获得初始图结构。It should be noted that the triangulation method is a technical method well known to those skilled in the art, and the specific process will not be described again. In other embodiments of the present invention, the initial graph structure can also be obtained through other graph structure construction algorithms such as Thiessen polygon construction algorithm.
通过对初始图结构中的图结构节点进行多次不同聚类尺度下的聚类即可获得每个聚类尺度下的聚类类别。需要说明的是，聚类尺度的具体数量选择可根据地质数据具体数据状态进行选取，在本发明一个实施例中将聚类尺度的数量设置为5个，即通过5个不同大小的聚类尺度分别对初始图结构进行聚类，每个聚类尺度下均对应一组聚类类别。The clustering category at each clustering scale can be obtained by clustering the graph structure nodes in the initial graph structure multiple times at different clustering scales. It should be noted that the specific number of clustering scales can be selected according to the specific data status of the geological data. In one embodiment of the present invention, the number of clustering scales is set to 5, that is, through 5 clustering scales of different sizes. The initial graph structure is clustered separately, and each clustering scale corresponds to a set of clustering categories.
优选地，在本发明一个实施例中，采用拉普拉斯图聚类的聚类方法进行迭代聚类，获得不同聚类尺度下的所述聚类类别。拉普拉斯图聚类为本领域技术人员熟知的技术手段，在此不做赘述及说明。Preferably, in one embodiment of the present invention, the clustering method of Laplacian graph clustering is used to perform iterative clustering to obtain the clustering categories at different clustering scales. Laplacian graph clustering is a technical method well known to those skilled in the art, and will not be described in detail here.
步骤S3：获得每个聚类类别的节点结构特征；获得每个插值位置在不同聚类尺度下所处聚类类别之间的节点结构特征的匹配相似度；根据匹配相似度获取每个插值位置在不同聚类尺度之间的关联聚类类别集合；在关联聚类类别集合中，以中心类别与每个聚类类别之间的匹配相似度作为每个聚类类别对应聚类尺度下的权重。Step S3: Obtain the node structure characteristics of each clustering category; obtain the matching similarity of the node structure characteristics between the clustering categories of each interpolation position at different clustering scales; obtain each interpolation position based on the matching similarity. A set of associated clustering categories between different clustering scales; in a set of associated clustering categories, the matching similarity between the central category and each clustering category is used as the weight of each clustering category at the corresponding clustering scale. .
因为聚类过程是基于图结构进行聚类，因此每个聚类尺度下所得到的聚类类别可视为图结构中的一片节点区域，分析不同尺度之间聚类类别的自相关性可视为分析节点区域之间的结构相似性，因此为了更准确分析不同尺度之间聚类类别的自相关性，需要首先获得每个聚类类别的节点结构特征，以节点结构特征作为基础数据进行后续的相似性计算。Because the clustering process is based on the graph structure, the clustering categories obtained at each clustering scale can be regarded as a node area in the graph structure. Analyzing the autocorrelation of the clustering categories between different scales can be seen In order to analyze the structural similarity between node areas, and therefore to more accurately analyze the autocorrelation of cluster categories between different scales, it is necessary to first obtain the node structure characteristics of each cluster category, and use the node structure characteristics as basic data for subsequent similarity calculation.
优选地，本发明一个实施例中，获得每个聚类类别的节点结构特征包括：Preferably, in one embodiment of the present invention, obtaining the node structure characteristics of each cluster category includes:
对聚类类别进行图因子分解，获得分解子图。每个分解子图都包含了所有节点，但是不同分解子图之间节点的连接方式不同，因此每个聚类类别的分解子图集合可以表示对应聚类类别中的节点结构分布信息。通过最小哈希法获得每个分解子图的二进制码，通过二进制码的转换可以将每个分解子图中的节点结构进行量化，方便后续的相似度计算。因此以所有分解子图的二进制码的集合作为对应聚类类别的节点结构特征。需要说明的是，图因子分解和最小哈希法均为本领域技术人员熟知的技术手段，具体过程不再赘述。Perform graph factorization on cluster categories to obtain decomposed subgraphs. Each decomposed subgraph contains all nodes, but the nodes are connected in different ways between different decomposed subgraphs. Therefore, the decomposed subgraph set of each clustering category can represent the node structure distribution information in the corresponding clustering category. The binary code of each decomposed subgraph is obtained through the minimum hash method. The node structure in each decomposed subgraph can be quantified through the conversion of the binary code to facilitate subsequent similarity calculations. Therefore, the set of binary codes of all decomposed subgraphs is used as the node structure feature of the corresponding clustering category. It should be noted that graph factorization and minimum hashing are both technical means well known to those skilled in the art, and the specific processes will not be described again.
在本发明其他实施例中，也可利用链码分析或者傅里叶变换等方法获取聚类类别中的节点结构特征，具体算法为本领域技术人员熟知的技术手段，在此不做赘述及说明。In other embodiments of the present invention, methods such as chain code analysis or Fourier transform can also be used to obtain the node structure characteristics in the clustering category. The specific algorithm is a technical means well known to those skilled in the art, and will not be described in detail here. .
通过步骤S2的聚类后，同一个插值位置在不同的聚类尺度下对应不同的聚类类别，因此获得插值位置在不同聚类尺度下所处聚类类别之间节点结构特征的匹配相似度，匹配相似度越大，说明两个聚类类别之间的节点结构特征越相似，相关性越大。通过匹配相似度即可获取每个插值位置在不同聚类尺度之间的关联聚类类别集合，即关联聚类类别集合中的每个关联聚类类别对应一个聚类尺度。After clustering in step S2, the same interpolation position corresponds to different clustering categories at different clustering scales. Therefore, the matching similarity of node structure features between the clustering categories of the interpolation position at different clustering scales is obtained. , the greater the matching similarity, the more similar the node structure characteristics between the two clustering categories are, and the greater the correlation. By matching the similarity, the associated clustering category set of each interpolation position between different clustering scales can be obtained, that is, each associated clustering category in the associated clustering category set corresponds to a clustering scale.
优选地，因为本发明一个实施例中通过构建分解子图集合的方法获得每个聚类类别的节点结构特征，所以可将插值位置在不同聚类尺度下所处聚类类别之间的节点结构特征进行KM匹配。KM匹配过程中以两个聚类类别之间的每个分解子图作为匹配节点，以分解子图之间的二进制码的余弦相似度作匹配值，将匹配值大于预设匹配阈值的匹配节点组作为保留节点组。保留节点组中的两个匹配节点分别隶属于不同聚类尺度下的聚类类别，保留节点可视为两个不同聚类类别之间具有相似特征的节点，因此以保留节点组的数量与两个聚类类别中的最大分解子图数量的比值作为匹配相似度，即最大分解子图数量表示匹配节点组的最大值，保留节点组的数量越多，说明具有相似特征的匹配节点占比越大，则对应的两个聚类类别匹配相似度越大。匹配相似度用公式表示为：Preferably, because in one embodiment of the present invention, the node structure characteristics of each clustering category are obtained by constructing a decomposed subgraph set, the interpolation position can be set at the node structure between the clustering categories at different clustering scales. Features are matched by KM. In the KM matching process, each decomposed subgraph between two clustering categories is used as a matching node, the cosine similarity of the binary codes between the decomposed subgraphs is used as the matching value, and the matching nodes whose matching value is greater than the preset matching threshold are group as a reserved node group. The two matching nodes in the retained node group belong to clustering categories at different clustering scales. The retained nodes can be regarded as nodes with similar characteristics between two different clustering categories. Therefore, the number of retained node groups is related to the two clustering categories. The ratio of the maximum number of decomposed subgraphs in each clustering category is used as the matching similarity, that is, the maximum number of decomposed subgraphs represents the maximum value of the matching node group. The greater the number of retained node groups, the greater the proportion of matching nodes with similar characteristics. larger, the greater the matching similarity between the two corresponding clustering categories. The matching similarity is expressed as:
其中，P(a,b)表示聚类类别a和聚类类别b之间的匹配相似度，N′表示聚类类别a和聚类类别b之间经过匹配后的保留节点组的数量，max()为最大值选取函数，Na为聚类类别a的分解子图数量，Nb为聚类类别b的分解子图数量。Among them, P(a, b) represents the matching similarity between cluster category a and cluster category b, N′ represents the number of retained node groups after matching between cluster category a and cluster category b, max () is the maximum value selection function,Na is the number of decomposed subgraphs of clustering category a, andNb is the number of decomposed subgraphs of clustering category b.
在本发明实施例中，匹配阈值设置为0.7。In this embodiment of the present invention, the matching threshold is set to 0.7.
需要说明的是，在本发明一些实施例中的节点结构特征并非为集合形式，因此可直接将节点结构特征进行相似度计算，获得匹配相似度。It should be noted that in some embodiments of the present invention, the node structural features are not in the form of a set. Therefore, the node structural features can be directly calculated for similarity to obtain the matching similarity.
通过匹配相似度的计算过程，可获得插值位置对应的所有聚类类别之间的匹配相似度，匹配相似度越大说明两个聚类类别相关性越大，进而筛选出插值位置对应的不同聚类尺度之间的关联聚类类别集合，即关联聚类类别集合中的每个关联聚类类别均为不同聚类尺度。Through the calculation process of matching similarity, the matching similarity between all cluster categories corresponding to the interpolation position can be obtained. The greater the matching similarity, the greater the correlation between the two cluster categories, and then the different clusters corresponding to the interpolation position are screened out. A set of associated clustering categories between class scales, that is, each associated clustering category in the set of associated clustering categories is a different clustering scale.
优选地，本发明一个实施例中根据匹配相似度获取每个插值位置在不同聚类尺度之间的关联聚类类别集合包括：若匹配相似度大于预设相似度阈值，则对应的两个聚类类别存在关联关系；获得插值位置在不同聚类尺度下对应的所有聚类类别之间的关联关系，将具有关联关系的聚类类别作为关联聚类类别，获得关联聚类类别集合。例如插值位置A经过匹配相似度的计算后，认定尺度1和尺度2对应的聚类类别存在关联关系，尺度1和尺度3对应的聚类类别不存在关联关系，而尺度2和尺度3对应的聚类类别存在关联关系，则通过关联关系整合后，插值位置A对应的关联聚类类别集合为尺度1、尺度2和尺度3对应的聚类类别集合。Preferably, in one embodiment of the present invention, obtaining the associated clustering category set of each interpolation position between different clustering scales based on the matching similarity includes: if the matching similarity is greater than the preset similarity threshold, then the corresponding two clusters There is an association relationship between the class categories; the association relationship between all clustering categories corresponding to the interpolation position at different clustering scales is obtained, the clustering categories with the association relationship are used as the associated clustering categories, and the associated clustering category set is obtained. For example, after calculating the matching similarity at interpolation position A, it is determined that the clustering categories corresponding to scale 1 and scale 2 are related, the clustering categories corresponding to scale 1 and scale 3 are not related, and the clustering categories corresponding to scale 2 and scale 3 are not related. If there is an association relationship between the clustering categories, then after integration through the association relationship, the set of associated clustering categories corresponding to the interpolation position A is the set of clustering categories corresponding to scale 1, scale 2 and scale 3.
在本发明一个实施例中，考虑到匹配像素点均为0到1之间的数值，因此将相似度阈值设置为0.7。In one embodiment of the present invention, considering that all matching pixel points have values between 0 and 1, the similarity threshold is set to 0.7.
每个插值位置对应一个关联聚类类别集合，关联聚类类别集合为获取插值位置的最终插值数据的基础，因为关联聚类类别集合中的聚类类别之间均存在相关性，因此可通过赋予关联聚类类别集合中每个关联聚类类别权重，通过后续的加权求和获得准确的最终差值数据。因为关联聚类类别集合中的中心类别为匹配过程中匹配相似度最大的类别，因此可通过中心类别作为基础类别，获得集合中每个聚类类别与中心类别之间的匹配相似度，并将对应的匹配相似度作为聚类类别对应聚类尺度下的权重。需要说明的是，权重获取过程中需要计算中心类别与集合中每个聚类类别之间的匹配相似度，而中心类别与自身的匹配相似度为1，因此中心类别所对应的聚类尺度下的权重也为1。Each interpolation position corresponds to an associated clustering category set. The associated clustering category set is the basis for obtaining the final interpolation data of the interpolation position. Because there is a correlation between the clustering categories in the associated clustering category set, it can be obtained by assigning The weight of each associated clustering category in the associated clustering category set is used to obtain accurate final difference data through subsequent weighted summation. Because the central category in the associated clustering category set is the category with the greatest matching similarity during the matching process, the matching similarity between each clustering category in the set and the central category can be obtained by using the central category as the basic category, and The corresponding matching similarity is used as the weight of the clustering category corresponding to the clustering scale. It should be noted that during the weight acquisition process, the matching similarity between the central category and each cluster category in the set needs to be calculated, and the matching similarity between the central category and itself is 1, so under the clustering scale corresponding to the central category The weight of is also 1.
步骤S4：获得插值位置在关联聚类类别集合中每个聚类类别对应聚类尺度下的插值数据，根据对应聚类尺度下的权重和插值数据进行整合，获得对应插值位置的最终插值数据；获得所有插值位置的最终插值数据，获得增强地质数据。Step S4: Obtain the interpolation data at the clustering scale corresponding to each clustering category in the associated clustering category set at the interpolation position, integrate it according to the weight and interpolation data at the corresponding clustering scale, and obtain the final interpolation data corresponding to the interpolation position; Final interpolated data are obtained for all interpolated locations, resulting in enhanced geological data.
因为需要根据关联聚类类别集合对插值位置进行插值数据分析，因此可首先获得插值位置在关联聚类类别集合中每个聚类类别对应聚类尺度下的插值数据。进而根据对应聚类尺度下的权重和插值数据进行整合，即可获得对应插值位置下的最终差值数据。因为一个聚类尺度下的插值数据和权重一一对应，而关联聚类类别集合中的聚类类别对应的尺度之间存在关联性，因此通过权重和插值数据的整合所获得的最终插值数据更准确，稳定性更强。通过对所有插值位置进行同样方法的分析，即可获得不存在缺失数据的增强地质数据。Because it is necessary to perform interpolation data analysis on the interpolation position according to the associated clustering category set, the interpolation data of the interpolation position at the clustering scale corresponding to each clustering category in the associated clustering category set can first be obtained. Then, by integrating the weights and interpolation data at the corresponding clustering scale, the final difference data at the corresponding interpolation position can be obtained. Because the interpolation data and weights under a clustering scale have a one-to-one correspondence, and there is a correlation between the scales corresponding to the clustering categories in the associated clustering category set, the final interpolation data obtained by integrating the weights and interpolation data is more accurate. Accurate and more stable. By performing the same analysis on all interpolated locations, enhanced geological data without missing data can be obtained.
优选地，本发明一个实施例中获得插值位置在关联聚类类别集合中每个聚类类别对应聚类尺度下的插值数据包括：根据插值位置所处聚类类别中的图结构节点，利用反距离插值法获得插值位置在对应聚类尺度下的插值数据。需要说明的是，反距离插值法为本领域技术人员熟知的技术手段，在此不做赘述。Preferably, in one embodiment of the present invention, obtaining the interpolation data at the clustering scale corresponding to each clustering category in the associated clustering category set at the interpolation position includes: using the inverse graph structure node according to the graph structure node in the clustering category where the interpolation position is located. The distance interpolation method obtains the interpolation data of the interpolation position at the corresponding clustering scale. It should be noted that the inverse distance interpolation method is a technical method well known to those skilled in the art, and will not be described in detail here.
优选地，本发明一个实施例中，最终插值数据的获取方法包括：Preferably, in one embodiment of the present invention, the method for obtaining final interpolation data includes:
将插值位置对应的每个插值数据与对应权重相乘，获得对应聚类尺度下的调整插值数据；将调整插值数据求平均，获得最终插值数据。通过加权求平均的方法能够使得不同尺度下的插值数据进行准确整合，获得参考性强的最终插值数据。Multiply each interpolation data corresponding to the interpolation position by the corresponding weight to obtain the adjusted interpolation data at the corresponding clustering scale; average the adjusted interpolation data to obtain the final interpolation data. Through the weighted averaging method, the interpolation data at different scales can be accurately integrated to obtain the final interpolation data with strong reference.
步骤S5：根据增强地质数据中的数据获得珊瑚礁砂基础管桩的侧摩阻力。Step S5: Obtain the side friction resistance of the coral reef sand foundation pipe pile based on the data in the enhanced geological data.
最终插值数据中的数据具有良好的完整性，即每个采样位置均存在对应地质数据，进而可进行准确的侧摩阻力的计算，通过现有技术公式即可计算侧摩阻力，具体公式包括：The data in the final interpolation data have good integrity, that is, there is corresponding geological data at each sampling location, so that accurate side friction can be calculated. The side friction can be calculated through existing technical formulas. The specific formulas include:
Fs＝t×AFs＝t×A
其中，Fs为侧摩阻力，t为珊瑚礁砂与管桩之间的静摩擦力，A为管桩侧表面积。Among them, Fs is the side friction resistance, t is the static friction force between the coral reef sand and the pipe pile, and A is the side surface area of the pipe pile.
需要说明的是，对于珊瑚礁砂而言，静摩擦力主要有砂粒直径和接触面积两个参数共同决定，可通过现有数据记载的摩擦系数结合两个参数进行计算获取。具体获取侧摩阻力的方法为本领域技术人员熟知的技术手段，在此不做赘述及限定。It should be noted that for coral reef sand, the static friction force is mainly determined by the two parameters of sand particle diameter and contact area, which can be calculated and obtained by combining the friction coefficient recorded in existing data with the two parameters. The specific method for obtaining side friction resistance is a technical means well known to those skilled in the art, and will not be described in detail or limited here.
综上所述，本发明实施例获取不同聚类尺度下的聚类类别，以每个聚类类别的节点结构特征作为基础数据，分析插值位置在不同聚类尺度下所处聚类类别之间的节点结构特征的匹配像素点，进而获得每个插值位置对应的关联聚类类别集合。赋予关联聚类类别集合中每个聚类类别权重，并分析对应尺度下的插值数据，通过权重和插值数据的整合，获得对应插值位置的最终差值数据。通过增强地质数据进行准确的侧摩阻力计算。本发明通过对多尺度对应的类别之间的自相关性进行分析，提高了地质数据的质量，进而提高了侧摩阻力计算的准确度。To sum up, the embodiment of the present invention obtains clustering categories at different clustering scales, uses the node structure characteristics of each clustering category as basic data, and analyzes the interpolation position between the clustering categories at different clustering scales. The matching pixels of the node structure characteristics are then obtained, and the associated clustering category set corresponding to each interpolation position is obtained. Give each cluster category in the associated cluster category set a weight, and analyze the interpolation data at the corresponding scale. Through the integration of the weight and interpolation data, the final difference data corresponding to the interpolation position is obtained. Accurate side friction calculations with enhanced geological data. The present invention improves the quality of geological data by analyzing the autocorrelation between categories corresponding to multiple scales, thereby improving the accuracy of side friction calculation.
需要说明的是：上述本发明实施例先后顺序仅仅为了描述，不代表实施例的优劣。在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中，多任务处理和并行处理也是可以的或者可能是有利的。It should be noted that the above-mentioned order of the embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the figures do not necessarily require the specific order shown, or sequential order, to achieve desirable results. Multitasking and parallel processing are also possible or may be advantageous in certain implementations.
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。Each embodiment in this specification is described in a progressive manner. The same and similar parts between the various embodiments can be referred to each other. Each embodiment focuses on its differences from other embodiments.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310881301.9ACN116842416B (en) | 2023-07-18 | 2023-07-18 | A calculation method suitable for the side friction of coral reef sand foundation pipe piles |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310881301.9ACN116842416B (en) | 2023-07-18 | 2023-07-18 | A calculation method suitable for the side friction of coral reef sand foundation pipe piles |
| Publication Number | Publication Date |
|---|---|
| CN116842416Atrue CN116842416A (en) | 2023-10-03 |
| CN116842416B CN116842416B (en) | 2024-03-15 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310881301.9AActiveCN116842416B (en) | 2023-07-18 | 2023-07-18 | A calculation method suitable for the side friction of coral reef sand foundation pipe piles |
| Country | Link |
|---|---|
| CN (1) | CN116842416B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095743A (en)* | 2023-10-17 | 2023-11-21 | 山东鲁润阿胶药业有限公司 | Polypeptide spectrum matching data analysis method and system for small molecular peptide donkey-hide gelatin |
| CN119249546A (en)* | 2024-08-28 | 2025-01-03 | 中交第二航务工程局有限公司 | Calculation method and device for pile foundation lateral friction and exertion coefficient considering interface shape |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113503156A (en)* | 2021-07-12 | 2021-10-15 | 山东科技大学 | Reservoir vertical heterogeneity quantitative determination method, system, medium, equipment and terminal |
| JP2021167533A (en)* | 2020-04-10 | 2021-10-21 | 株式会社オムテック | Foundation pile bearing capacity calculation device and its method considering liquefaction |
| CN113550288A (en)* | 2021-06-17 | 2021-10-26 | 中交二航局第三工程有限公司 | Clay stratum side friction resistance measuring device and measuring method |
| KR102452877B1 (en)* | 2022-06-02 | 2022-10-11 | 채수근 | Calculation method of surface friction of SDA pile constructed in weathered rock and bedrock |
| CN115935477A (en)* | 2022-12-14 | 2023-04-07 | 大连理工大学 | Design method of energy pile green low-carbon building based on machine learning |
| CN116245013A (en)* | 2022-12-29 | 2023-06-09 | 中国地质大学（武汉） | Geological prediction model construction method, modeling method, equipment and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2021167533A (en)* | 2020-04-10 | 2021-10-21 | 株式会社オムテック | Foundation pile bearing capacity calculation device and its method considering liquefaction |
| CN113550288A (en)* | 2021-06-17 | 2021-10-26 | 中交二航局第三工程有限公司 | Clay stratum side friction resistance measuring device and measuring method |
| CN113503156A (en)* | 2021-07-12 | 2021-10-15 | 山东科技大学 | Reservoir vertical heterogeneity quantitative determination method, system, medium, equipment and terminal |
| KR102452877B1 (en)* | 2022-06-02 | 2022-10-11 | 채수근 | Calculation method of surface friction of SDA pile constructed in weathered rock and bedrock |
| CN115935477A (en)* | 2022-12-14 | 2023-04-07 | 大连理工大学 | Design method of energy pile green low-carbon building based on machine learning |
| CN116245013A (en)* | 2022-12-29 | 2023-06-09 | 中国地质大学（武汉） | Geological prediction model construction method, modeling method, equipment and storage medium |
| Title |
|---|
| 张永涛 等: "珊瑚礁灰岩地层灌注桩侧摩阻力特征及发挥系数取值研究与应用", 岩石力学与工程学报, vol. 42, no. 3, 31 March 2023 (2023-03-31), pages 748 - 759* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095743A (en)* | 2023-10-17 | 2023-11-21 | 山东鲁润阿胶药业有限公司 | Polypeptide spectrum matching data analysis method and system for small molecular peptide donkey-hide gelatin |
| CN117095743B (en)* | 2023-10-17 | 2024-01-05 | 山东鲁润阿胶药业有限公司 | Polypeptide spectrum matching data analysis method and system for small molecular peptide donkey-hide gelatin |
| CN119249546A (en)* | 2024-08-28 | 2025-01-03 | 中交第二航务工程局有限公司 | Calculation method and device for pile foundation lateral friction and exertion coefficient considering interface shape |
| Publication number | Publication date |
|---|---|
| CN116842416B (en) | 2024-03-15 |
| Publication | Publication Date | Title |
|---|---|---|
| CN116842416B (en) | A calculation method suitable for the side friction of coral reef sand foundation pipe piles | |
| EP2761340B1 (en) | Method for extracting a thumbnail image from a training image so as to constrain the multipoint geostatistical modeling of the subsoil | |
| Huang et al. | Predictive modelling of seabed sediment parameters using multibeam acoustic data: a case study on the Carnarvon Shelf, Western Australia | |
| CN107807387B (en) | Acquisition Method of First Arrival Travel Time of Earthquake Based on Neural Network | |
| CN105353412A (en) | Calculating method and system of well-to-seismic integration average speed field | |
| CN114707376B (en) | Distribution method for simulating occurrence of multi-metal nodules at sea bottom | |
| CN109298464B (en) | Tight sandstone reservoir diagenetic facies logging identification method and device | |
| CN116976705B (en) | Deep oil gas accurate navigation sand shale stratum physical property evaluation method and system | |
| CN113360587B (en) | Land surveying and mapping equipment and method based on GIS technology | |
| CN120012027B (en) | Intelligent fusion and optimization processing method for ocean physical data | |
| CN116957364A (en) | Method and system for evaluating lithology of deep oil gas accurate navigation sand shale stratum | |
| Bai et al. | A fast edge-based two-stage direct sampling method | |
| CN108535777B (en) | A method and system for detecting the first arrival of seismic waves | |
| CN119443418A (en) | A multi-dimensional fusion prospecting prediction method based on multi-source data | |
| Chen et al. | A robust multiquadric method for digital elevation model construction | |
| Lloyd et al. | Geostatistics and spatial structure in archaeology | |
| CN118966306A (en) | A method for predicting spatial distribution of soil heavy metal content based on genetic algorithm and BP neural network | |
| Magyar et al. | The Wasserstein distance as a hydrological objective function | |
| CN119251422B (en) | Three-dimensional geological model construction method and system for geothermal resource exploration area | |
| CN120147308B (en) | A method, device, electronic device and storage medium for monitoring the distribution and evolution characteristics of seabed sediments | |
| CN119089106B (en) | Meteorological element nearest weighted interpolation method based on terrain importance degree | |
| Wu et al. | A New Method for Automatic Extraction and Analysis of Discontinuities Based on TIN on Rock Mass Surfaces. Remote Sens. 2021, 13, 2894 | |
| CN120472102A (en) | Geological body layered display method, device, computer equipment and storage medium | |
| Prathumchai et al. | Elevation Surface Interpolation of Point Data Using Different Techniques-a GIS Approach | |
| HU | Comparison of the interpolation methods on digital terrain models |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |