建筑网站资料, 163com箱登录,高质量的高密网站建设,电商运营怎么做如何从零开始地球科学家需要对地质环境进行最佳估计才能进行模拟或评估。 除了地质背景之外#xff0c;建立地质模型还需要一整套数学方法#xff0c;如贝叶斯网络、协同克里金法、支持向量机、神经网络、随机模型#xff0c;以在钻井日志或地球物理信息确实稀缺或不确定时定义哪些可能是…地球科学家需要对地质环境进行最佳估计才能进行模拟或评估。 除了地质背景之外建立地质模型还需要一整套数学方法如贝叶斯网络、协同克里金法、支持向量机、神经网络、随机模型以在钻井日志或地球物理信息确实稀缺或不确定时定义哪些可能是岩石类型/属性。 推荐用 NSDT编辑器 快速搭建可编程3D场景 我们已经用 Python 和最新强大的库Scikit Learn完成了一个教程以根据宝藏谷美国爱达荷州钻探的岩性创建地质模型。 本教程生成钻井岩性的点云并针对神经网络进行转换和缩放。 所选的神经网络分类器是多层感知器分类器在 Scikit Learn 库上实现为 sklearn.neural_network.MLPClassifier。 对神经网络的混淆进行分析。 本教程还包括 Paraview 中 Vtk 格式的井岩性和插值地质学的地理参考 3D 可视化。
首先导入必要的库
#import required libraries
%matplotlib inline
import os
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import pyvista as pv
import vtk1、井位置和岩性
数据来自来自公开发表论文选定的单位为
粗粒河流和冲积矿床上新世-更新世和中新世玄武岩细粒湖相沉积流纹岩和花岗岩基岩
wellLoc pd.read_csv(../inputData/TV-HFM_Wells_1Location_Wgs11N.csv,index_col0)
wellLoc.head()东向北向高度ft东向UTM北向UTM高程mA. Isaac2333140.951372225.653204.0575546.6288344.820355e06976.57920A. Woodbridge2321747.001360096.952967.2564600.3665824.807827e06904.40256A.D. Watkins2315440.161342141.863168.3558944.8434044.789664e06965.69784A.L. Clark; 12276526.301364860.742279.1519259.0061594.810959e06694.66968A.L. Clark; 22342620.871362980.463848.6585351.1502704.811460e061173.05328
2、岩性点云
litoPoints []for index, values in wellLito.iterrows():wellX, wellY, wellZ wellLoc.loc[values.Bore][[EastingUTM,NorthingUTM,Elevation_m]]wellXY [wellX, wellY]litoPoints.append(wellXY [values.topLitoElev_m,values.hydrogeoCode])litoPoints.append(wellXY [values.botLitoElev_m,values.hydrogeoCode])litoLength values.topLitoElev_m - values.botLitoElev_mif litoLength 1:midPoint wellXY [values.topLitoElev_m - litoLength/2,values.hydrogeoCode]else:npoints int(litoLength)for point in range(1,npoints1):disPoint wellXY [values.topLitoElev_m - litoLength*point/(npoints1),values.hydrogeoCode]litoPoints.append(disPoint)
litoNpnp.array(litoPoints)
np.save(../outputData/litoNp,litoNp)
litoNp[:5]
array([[5.48261389e05, 4.83802316e06, 7.70442960e02, 1.00000000e00],[5.48261389e05, 4.83802316e06, 7.70138160e02, 1.00000000e00],[5.48261389e05, 4.83802316e06, 7.70138160e02, 3.00000000e00],[5.48261389e05, 4.83802316e06, 7.68614160e02, 3.00000000e00],[5.48261389e05, 4.83802316e06, 7.69376160e02, 3.00000000e00]])3、坐标变换和神经网络分类器设置
from sklearn.neural_network import MLPClassifier
from sklearn.metrics import confusion_matrix
from sklearn import preprocessing
litoX, litoY, litoZ litoNp[:,0], litoNp[:,1], litoNp[:,2]
litoMean litoNp[:,:3].mean(axis0)
litoTrans litoNp[:,:3]-litoMean
litoTrans[:5]#setting up scaler
scaler preprocessing.StandardScaler().fit(litoTrans)
litoScale scaler.transform(litoTrans)#check scaler
print(litoScale.mean(axis0))
print(litoScale.std(axis0))
[ 2.85924590e-14 -1.10313442e-15 3.89483608e-20]
[1. 1. 1.]
#run classifier
X litoScale
Y litoNp[:,3]
clf MLPClassifier(activationtanh,solverlbfgs,hidden_layer_sizes(15,15,15), max_iter2000)
clf.fit(X,Y)
C:\Users\Gida\Anaconda3\lib\site-packages\sklearn\neural_network\_multilayer_perceptron.py:470: ConvergenceWarning: lbfgs failed to converge (status1):
STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.Increase the number of iterations (max_iter) or scale the data as shown in:https://scikit-learn.org/stable/modules/preprocessing.htmlself.n_iter_ _check_optimize_result(lbfgs, opt_res, self.max_iter)MLPClassifier(activationtanh, alpha0.0001, batch_sizeauto, beta_10.9,beta_20.999, early_stoppingFalse, epsilon1e-08,hidden_layer_sizes(15, 15, 15), learning_rateconstant,learning_rate_init0.001, max_fun15000, max_iter2000,momentum0.9, n_iter_no_change10, nesterovs_momentumTrue,power_t0.5, random_stateNone, shuffleTrue, solverlbfgs,tol0.0001, validation_fraction0.1, verboseFalse,warm_startFalse)4、混淆矩阵的确定
numberSamples litoNp.shape[0]
expectedlitoNp[:,3]
predicted []
for i in range(numberSamples):predicted.append(clf.predict([litoScale[i]]))
results confusion_matrix(expected,predicted)
print(results)输出如下
[[1370 128 377 0][ 67 2176 10 0][ 274 33 1114 0][ 1 0 0 151]]5、研究领域和输出网格细化
xMin 540000
xMax 560000
yMin 4820000
yMax 4840000
zMax int(wellLito.topLitoElev_m.max())
zMin zMax - 300
cellH 200
cellV 206、岩性基质的测定
vertexCols np.arange(xMin,xMax1,cellH)
vertexRows np.arange(yMax,yMin-1,-cellH)
vertexLays np.arange(zMax,zMin-1,-cellV)
cellCols (vertexCols[1:]vertexCols[:-1])/2
cellRows (vertexRows[1:]vertexRows[:-1])/2
cellLays (vertexLays[1:]vertexLays[:-1])/2
nCols cellCols.shape[0]
nRows cellCols.shape[0]
nLays cellLays.shape[0]
i0
litoMatrixnp.zeros([nLays,nRows,nCols])
for lay in range(nLays):for row in range(nRows):for col in range(nCols):cellXYZ [cellCols[col],cellRows[row],cellLays[lay]]cellTrans cellXYZ - litoMeancellNorm scaler.transform([cellTrans])litoMatrix[lay,row,col] clf.predict(cellNorm)if i%300000:print(Processing %s cells%i)print(cellTrans)print(cellNorm)print(litoMatrix[lay,row,col])i1
Processing 0 cells
[-8553.96427073 8028.26104284 356.7050941 ]
[[-1.41791371 2.42904321 1.11476509]]
3.0
Processing 30000 cells
[-8553.96427073 8028.26104284 296.7050941 ]
[[-1.41791371 2.42904321 0.92725472]]
3.0
Processing 60000 cells
[-8553.96427073 8028.26104284 236.7050941 ]
[[-1.41791371 2.42904321 0.73974434]]
3.0
Processing 90000 cells
[-8553.96427073 8028.26104284 176.7050941 ]
[[-1.41791371 2.42904321 0.55223397]]
2.0
Processing 120000 cells
[-8553.96427073 8028.26104284 116.7050941 ]
[[-1.41791371 2.42904321 0.3647236 ]]
2.0
plt.imshow(litoMatrix[0])
matplotlib.image.AxesImage at 0x14fb8688860plt.imshow(litoMatrix[:,60])matplotlib.image.AxesImage at 0x14fb871d390np.save(../outputData/litoMatrix,litoMatrix)#matrix modification for Vtk representation
litoMatrixMod litoMatrix[:,:,::-1]
np.save(../outputData/litoMatrixMod,litoMatrixMod)
plt.imshow(litoMatrixMod[0])matplotlib.image.AxesImage at 0x14fb87825f87、规则网格VTK的生成
import pyvista
import vtk# Create empty grid
grid pyvista.RectilinearGrid()# Initialize from a vtk.vtkRectilinearGrid object
vtkgrid vtk.vtkRectilinearGrid()
grid pyvista.RectilinearGrid(vtkgrid)
grid pyvista.RectilinearGrid(vertexCols,vertexRows,vertexLays)litoFlat list(litoMatrixMod.flatten(orderK))[::-1]
grid.cell_arrays[hydrogeoCode] np.array(litoFlat)
grid.save(../outputData/hydrogeologicalUnit.vtk)8、输入数据
你可以从这个链接下载本教程的输入数据。
9、数据源
Bartolino, J.R.2019爱达荷州和俄勒冈州宝藏谷及周边地区的水文地质框架美国地质调查局科学调查报告 2019-5138第 31 页。 链接 。 Bartolino, J.R.2020爱达荷州和俄勒冈州宝藏谷及周边地区的水文地质框架美国地质调查局数据发布。链接。 原文链接3D地质神经网络模型 — BimAnt
