• OrdinaryKriging3D插值出的结果都是均值问题查找

    一、问题:OrdinaryKriging3D插值出的结果都是均值
    代码如下:

    1. # -*- coding: utf-8 -*-
    2. """
    3. Created on Wed Nov 23 17:47:48 2022
    4. @author: junco
    5. """
    6. import numpy as np
    7. from pykrige.ok3d import OrdinaryKriging3D
    8.  
    9. #准备插值的原始数据
    10. lons=[114.82629297477898, 115.0061462015791, 114.83528563611898, 114.79931499075896, 114.844278297459, 115.25794071909925, 115.10506547631915, 115.12305079899916, 115.10506547631915, 115.64462515671948, 115.6536178180595, 114.83528563611898]
    11. lats=[41.31965410512094, 41.32864676646095, 41.35562475048096, 41.364617411820966, 41.418573379861, 41.60741926800112, 41.7333165267612, 41.7333165267612, 41.7423091881012, 41.373610073160975, 41.373610073160975, 41.832235801501255]
    12. depths=[2000, 2000, 2000, 2000, 2000, 2000, 2000, 2000, 2000, 3000, 3000, 3000] #1000m一个深度
    13. values=[51.584, 48.020004, 50.936005, 45.428, 45.428, 54.5, 51.584, 47.372, 45.104004, 49.64, 46.076004, 46.4]
    14. print ("均值：")
    15. print (sum(values)/len(values))
    16. #准备插值后的网格
    17. grid_lon = np.arange(114.7, 115.7, 0.1)
    18. grid_lat = np.arange(41.3, 41.9, 0.1)
    19. grid_depth = np.arange(2000, 3000, 1000) # 1000m一个深度
    20.  
    21. print(grid_lon)
    22. print(grid_lat)
    23. print(grid_depth)
    24.  
    25. # 普通克里金3维插值类
    26. ok3d = OrdinaryKriging3D(lons, lats, depths, values, variogram_model='gaussian')
    27. z1, ss1 = ok3d.execute('grid', grid_lon, grid_lat, grid_depth)
    28. #print ("插值后的网格数据形状是：")
    29. #print(z1.shape)  #插值后的网格数据形状是
    30. #print ("插值后的网格数据：")
    31. #print(z1.data)  #插值后的网格数据
    32. print ("插值后的数据最大值：")
    33. print(np.max(z1.data))
    34. print ("插值后的数据最小值：")
    35. print(np.min(z1.data))

    插值结果如下:

    1. 均值：
    2. 48.5060014167
    3. [ 114.7  114.8  114.9  115.   115.1  115.2  115.3  115.4  115.5  115.6
    4.   115.7]
    5. [ 41.3  41.4  41.5  41.6  41.7  41.8  41.9]
    6. [ 2000.  3000.]
    7. 插值后的数据最大值：
    8. 48.5060014167
    9. 插值后的数据最小值：
    10. 48.5060014167

    得到的结果都是48.5060014167，为原始值的均值。
    二、解决:经过排查，是插值后的网格点离输入的原始点太远。改下代码将插值后的网格点包含输入的原始点，结果显示不再全部为原始值的均值。
    修改后的代码如下:

    1. # -*- coding: utf-8 -*-
    2. """
    3. Created on Wed Nov 23 17:47:48 2022
    4. @author: junco
    5. """
    6. import numpy as np
    7. from pykrige.ok3d import OrdinaryKriging3D
    8.  
    9. #准备插值的原始数据
    10. lons=[114.82629297477898, 115.0061462015791, 114.83528563611898, 114.79931499075896, 114.844278297459, 115.25794071909925, 115.10506547631915, 115.12305079899916, 115.10506547631915, 115.64462515671948, 115.6536178180595, 114.83528563611898]
    11. lats=[41.31965410512094, 41.32864676646095, 41.35562475048096, 41.364617411820966, 41.418573379861, 41.60741926800112, 41.7333165267612, 41.7333165267612, 41.7423091881012, 41.373610073160975, 41.373610073160975, 41.832235801501255]
    12. depths=[2000, 2000, 2000, 2000, 2000, 2000, 2000, 2000, 2000, 3000, 3000, 3000] #1000m一个深度
    13. values=[51.584, 48.020004, 50.936005, 45.428, 45.428, 54.5, 51.584, 47.372, 45.104004, 49.64, 46.076004, 46.4]
    14. print ("均值：")
    15. print (sum(values)/len(values))
    16. #准备插值后的网格
    17. grid_lon = np.arange(114.79931499075896, 115.6536178180595, 0.008992661340005603)
    18. grid_lat = np.arange(41.31965410512094, 41.832235801501255, 0.008992661340005603)
    19. grid_depth = np.arange(2000, 3000, 1000) # 1000m一个深度
    20.  
    21. print(grid_lon)
    22. print(grid_lat)
    23. print(grid_depth)
    24.  
    25. # 普通克里金3维插值类
    26. ok3d = OrdinaryKriging3D(lons, lats, depths, values, variogram_model='gaussian')
    27. z1, ss1 = ok3d.execute('grid', grid_lon, grid_lat, grid_depth)
    28. #print ("插值后的网格数据形状是：")
    29. #print(z1.shape)  #插值后的网格数据形状是
    30. #print ("插值后的网格数据：")
    31. #print(z1.data)  #插值后的网格数据
    32. print ("插值后的数据最大值：")
    33. print(np.max(z1.data))
    34. print ("插值后的数据最小值：")
    35. print(np.min(z1.data))


    修改后的代码插值结果如下:

    1. 均值：
    2. 48.5060014167
    3. [ 114.7  114.8  114.9  115.   115.1  115.2  115.3  115.4  115.5  115.6
    4.   115.7]
    5. [ 41.3  41.4  41.5  41.6  41.7  41.8  41.9]
    6. [ 2000.  3000.]
    7. 插值后的数据最大值：
    8. 48.5060014167
    9. 插值后的数据最小值：
    10. 48.5060014167
    11.  
    12. runfile('C:/Users/admin/Desktop/nc/untitled0.py', wdir='C:/Users/admin/Desktop/nc')
    13. 均值：
    14. 48.5060014167
    15. [ 114.79931499  114.80830765  114.81730031  114.82629297  114.83528564
    16.   114.8442783   114.85327096  114.86226362  114.87125628  114.88024894
    17.   114.8892416   114.89823427  114.90722693  114.91621959  114.92521225
    18.   114.93420491  114.94319757  114.95219023  114.96118289  114.97017556
    19.   114.97916822  114.98816088  114.99715354  115.0061462   115.01513886
    20.   115.02413152  115.03312419  115.04211685  115.05110951  115.06010217
    21.   115.06909483  115.07808749  115.08708015  115.09607281  115.10506548
    22.   115.11405814  115.1230508   115.13204346  115.14103612  115.15002878
    23.   115.15902144  115.16801411  115.17700677  115.18599943  115.19499209
    24.   115.20398475  115.21297741  115.22197007  115.23096274  115.2399554
    25.   115.24894806  115.25794072  115.26693338  115.27592604  115.2849187
    26.   115.29391136  115.30290403  115.31189669  115.32088935  115.32988201
    27.   115.33887467  115.34786733  115.35685999  115.36585266  115.37484532
    28.   115.38383798  115.39283064  115.4018233   115.41081596  115.41980862
    29.   115.42880128  115.43779395  115.44678661  115.45577927  115.46477193
    30.   115.47376459  115.48275725  115.49174991  115.50074258  115.50973524
    31.   115.5187279   115.52772056  115.53671322  115.54570588  115.55469854
    32.   115.5636912   115.57268387  115.58167653  115.59066919  115.59966185
    33.   115.60865451  115.61764717  115.62663983  115.6356325   115.64462516
    34.   115.65361782]
    35. [ 41.31965411  41.32864677  41.33763943  41.34663209  41.35562475
    36.   41.36461741  41.37361007  41.38260273  41.3915954   41.40058806
    37.   41.40958072  41.41857338  41.42756604  41.4365587   41.44555136
    38.   41.45454403  41.46353669  41.47252935  41.48152201  41.49051467
    39.   41.49950733  41.50849999  41.51749265  41.52648532  41.53547798
    40.   41.54447064  41.5534633   41.56245596  41.57144862  41.58044128
    41.   41.58943395  41.59842661  41.60741927  41.61641193  41.62540459
    42.   41.63439725  41.64338991  41.65238257  41.66137524  41.6703679
    43.   41.67936056  41.68835322  41.69734588  41.70633854  41.7153312
    44.   41.72432387  41.73331653  41.74230919  41.75130185  41.76029451
    45.   41.76928717  41.77827983  41.78727249  41.79626516  41.80525782
    46.   41.81425048  41.82324314]
    47. [2000]
    48. 插值后的数据最大值：
    49. 54.5
    50. 插值后的数据最小值：
    51. 45.104004

  • 相关阅读:
    C#:实现字符串转整形算法(附完整源码)
    基于SqlSugar的开发框架循序渐进介绍（17）-- 基于CSRedis实现缓存的处理
    如何使用Python快速实现文件共享并通过内网穿透技术公网访问？
    [极致用户体验] 你的 Link Button 能让用户选择新页面打开吗？
    Dijkstra、A*算法python实现及对比分析
    微信小程序登录后端
    『外卖好评（通用50字）』
    下一个行业风口：NFT 数字藏品，是机遇还是泡沫？
    笔记本、台式机、平板二合一？Mac、Win、Linux？
    计算机操作系统：实验2 【银行家算法】
  • 原文地址：https://blog.csdn.net/cwr888/article/details/128112051