Skip to content

Latest commit

 

History

History
138 lines (114 loc) · 4.95 KB

transform.md

File metadata and controls

138 lines (114 loc) · 4.95 KB

<< 返回

数据预处理

备注：这里介绍的各种预处理都分单独版和批量版，批量实现的函数名后面多了一个 s。以下举例都是按照批量转换，既传入 measure 列表的形式。演示数据如下：

meas = [
    ('station', 'TN1', 100),
    ('ts', 'TN1', 'TN2', 30.15319, 15.37591, 1000)
]

度分秒转弧度

针对 ts，angle，dir 类型的观测，将小数表示的度分秒转成弧度，以供后续计算

meas = mfit.dms2rs(meas)
print(meas)

[('station', 'TN1', 100), ('ts', 'TN1', 'TN2', 0.5281167542925584, 0.2728487764053169, 1000)]

气象距离改正数计算

根据温度，湿度，气压，计算边长改正比，单位ppm

ppm = mfit.count_ppm(35, 0.5, 1000)
print(ppm)

计算气象ppm: -24.625733318414735

加乘常数改正，气象 ppm 改正

用上述计算的ppm，以及加乘常熟，对边长测量进行改正

meas = mfit.ts_adjusts(meas, b=1, k=1, ppm=ppm)
print(meas)

加乘常数,ppm改正: [('station', 'TN1', 100), ('ts', 'TN1', 'TN2', 0.5281167542925584, 0.2728487764053169, 999.9763742666815)]

斜距改高差，平距，并进行球气差改正

将竖直角，斜距拆分成高差，平距，并进行地球曲率改正，大气折光改正

meas = mfit.ts2ahds(meas, db=1, dk=1, da=1, K=0.1)
print(meas)

水平角，竖直角，斜距转水平角，高差，平距，并进行曲率改正，折光改正 [('station', 'TN1', 100), ('axhd', 'TN1', 'TN2', 0.5281167542925584, 269.535070255318, 962.9660421210108, 1, 4.684207947929276, 1.8871853600136306)]

水准面投影改正

将平距投影到统一高程面上

meas = mfit.distprojs(meas, 500)
print(meas)

平距投影到500m海拔 [('station', 'TN1', 100), ('axhd', 'TN1', 'TN2', 0.5281167542925584, 269.535070255318, 963.0061441553868, 1, 4.684207947929276, 1.8871853600136306)]

拆分成基础观测

meas = mfit.splitmeas(meas)
print(meas)

拆分成基础观测 [('station', 'TN1', 100), ('dir', 'TN1', 'TN2', 0.5281167542925584, 1), ('dist', 'TN1', 'TN2', 963.0061441553868, 1.8871853600136306), ('lev', 'TN1', 'TN2', 269.535070255318, 4.684207947929276)]

一个从改正到解算的完整例子

# 后方交汇
points = [
    ('TN2', 1000, 1000, 101, True),
    ('TN8', 1367.77286, 1117.43005, 102, True),
    ('TN9', 1294.76794, 1789.72043, 103, True),
    ('TN1', 893, 1603, 100, False)]

measures = [
    ('station', 'TN1', 100),
    ('ts', 'TN1', 'TN2', 0, 0.00163, 612.41862),
    ('ts', 'TN1', 'TN8', 0.598815, 0.00294, 678.73168),
    ('ts', 'TN1', 'TN9', 1.8317007, 0.00678, 442.39598)
]

# 预处理
ppm = mfit.count_ppm(35, 0.5, 1000)
meas = mfit.ts_adjusts(measures, b=1, k=1, ppm=ppm)
meas = mfit.ts2ahds(meas, db=1, dk=1, da=1, K=0.1)
meas = mfit.distprojs(meas, 500)

# n高效算法到这里即可
rst, sta = mfit.backward_x(points, meas)
print(mfit.report([rst]))

点位信息及精度

name x y h type dxx dxy dyy dhh
TN1 893.6412 1603.1319 99.9805 False 0.8691 0.1680 0.8190 2.3629

以上是高效算法，如果用通用模型，需要进一步分解成基础观测

# 经典算法要继续拆分成基础观测
meas = mfit.splitmeas(meas)
pts, info, sta = mfit.solve(points, meas, db=1, dk=1, da=1)
print(mfit.report(pts, info, sta))

点位信息及精度

name x y h type dxx dxy dyy dhh
TN2 1000.0000 1000.0000 101.0000 True nan nan nan nan
TN8 1367.7729 1117.4300 102.0000 True nan nan nan nan
TN9 1294.7679 1789.7204 103.0000 True nan nan nan nan
TN1 893.6412 1603.1319 99.9805 False 0.8692 0.1681 0.8190 2.3629

方向观测及精度

measure from to mea(°) value(°) dl(″)
direction TN1 TN2 0.0000 280.0010 0.3852
direction TN1 TN8 34.3096 314.3094 0.3417
direction TN1 TN9 104.9487 24.9460 0.4289

边长观测及精度

measure from to mea(m) value(m) dl(mm)
distance TN1 TN2 612.4421 612.4380 0.7630
distance TN1 TN8 678.7555 678.7541 0.6755
distance TN1 TN9 442.4035 442.4002 0.9888

水准观测及精度

measure from to mea(m) value(m) dl(mm)
level TN1 TN2 1.0247 1.0195 2.3629
level TN1 TN8 2.0280 2.0195 2.3629
level TN1 TN9 3.0132 3.0195 2.3629

后验单位中误差

后验单位中误差: 5.1347