使用说明更新

本库的基本用法

目前主要包含以下几个功能

  • 计算原子温度
  • 计算原子结构熵
  • 计算中心对称参数
  • 建立简单晶体模型
  • 建立原子近邻列表

导入必要的package

1
2
3
4
import mdapy as mp # 无特殊说明本库简写为 mp
import numpy as np
import taichi as ti
ti.init(arch=ti.cpu) #arch=ti.cpu 使用CPU. # arch=ti.gpu, device_memory_GB=5. 使用GPU
[Taichi] version 1.1.2, llvm 10.0.0, commit f25cf4a2, win, python 3.7.3
[Taichi] Starting on arch=x64

查看版本号

1
mp.__version__
'0.4.0'

读取一个系统

可在此处下载example.dump文件

1
2
%%time
system = mp.System('example.dump') 
Wall time: 2.96 s

查看系统数据

1
system.data
id type x y z vx vy vz c_1 c_2
0 192295 3 -27.9096 42.0121 -22.2491 -2.35674 0.481940 -0.257686 5.06650 -3.71151
1 255932 5 -30.0338 43.2841 -22.1938 1.55510 -1.125920 -1.983760 5.16323 -3.69935
2 319374 2 -29.4179 42.0487 -20.1799 -1.56154 -0.105773 -1.052900 6.02419 -3.51732
3 383068 3 -27.3053 43.2945 -20.1681 1.75865 0.962938 -2.273980 5.15436 -3.90756
4 700644 2 -27.9671 44.5056 -22.1799 2.11749 -0.684720 -0.795077 6.15314 -3.85788
... ... ... ... ... ... ... ... ... ... ...
2287143 1715363 1 329.5120 257.9240 323.0480 1.93039 1.215740 -1.869160 5.89311 -3.92515
2287144 1905765 5 327.2950 256.6070 318.9410 2.16960 -4.590730 1.395140 4.94041 -4.08289
2287145 1966370 3 329.5330 257.9350 318.9310 2.65692 -3.539740 -1.125020 5.70667 -3.83374
2287146 2030121 1 328.9320 256.6390 321.0620 1.25297 0.868006 1.104770 5.42057 -3.39951
2287147 2093740 5 326.7530 257.7980 320.8960 2.40437 -0.944472 3.358390 6.04798 -2.90699

2287148 rows × 10 columns

生成邻域列表

1
2
%%time
system.build_neighbor(rc=5., max_neigh=60)
Wall time: 3.82 s

计算中心对称参数

1
2
%%time
system.cal_centro_symmetry_parameter(N=12)
Wall time: 8.7 s

计算原子构型熵

1
2
%%time
system.cal_atomic_entropy()
Wall time: 4.31 s

计算平均原子温度

1
2
%%time
system.cal_atomic_temperature(amass=np.array([58.933, 58.693, 55.847, 26.982, 63.546]))
Wall time: 1.21 s

查看计算结果

1
system.data
id type x y z vx vy vz c_1 c_2 csp atomic_entropy atomic_temp
0 192295 3 -27.9096 42.0121 -22.2491 -2.35674 0.481940 -0.257686 5.06650 -3.71151 5.066506 -3.704394 271.311096
1 255932 5 -30.0338 43.2841 -22.1938 1.55510 -1.125920 -1.983760 5.16323 -3.69935 5.163225 -3.693706 244.521713
2 319374 2 -29.4179 42.0487 -20.1799 -1.56154 -0.105773 -1.052900 6.02419 -3.51732 6.024189 -3.511253 240.937302
3 383068 3 -27.3053 43.2945 -20.1681 1.75865 0.962938 -2.273980 5.15436 -3.90756 5.154361 -3.897303 230.167953
4 700644 2 -27.9671 44.5056 -22.1799 2.11749 -0.684720 -0.795077 6.15314 -3.85788 6.153219 -3.849362 243.981369
... ... ... ... ... ... ... ... ... ... ... ... ... ...
2287143 1715363 1 329.5120 257.9240 323.0480 1.93039 1.215740 -1.869160 5.89311 -3.92515 5.893107 -3.916707 260.688812
2287144 1905765 5 327.2950 256.6070 318.9410 2.16960 -4.590730 1.395140 4.94041 -4.08289 4.940413 -4.073879 320.674103
2287145 1966370 3 329.5330 257.9350 318.9310 2.65692 -3.539740 -1.125020 5.70667 -3.83374 5.706692 -3.825464 344.000671
2287146 2030121 1 328.9320 256.6390 321.0620 1.25297 0.868006 1.104770 5.42057 -3.39951 5.420572 -3.393086 241.169174
2287147 2093740 5 326.7530 257.7980 320.8960 2.40437 -0.944472 3.358390 6.04798 -2.90699 6.048002 -2.902135 245.842743

2287148 rows × 13 columns

查看平均温度

1
system.data['atomic_temp'].mean() # K
293.87168742301066

与LAMMPS计算结果进行对比

1
2
3
# csp 对比
error = abs(system.data['c_1'] - system.data['csp'])
error[error>0.1], error.max()
(Series([], dtype: float64), 0.0049196543884280075)

1
2
3
# 构型熵对比, 这里由于中间数值积分的地方我不清楚lammps的具体方式,结果数值上会有一些区别。
error = abs(system.data['c_2'] - system.data['atomic_entropy'])
error[error>0.1], error.max()
(Series([], dtype: float64), 0.08606090263366717)

保存计算结果到dump文件中

1
2
%%time
system.write_dump() # 生成一个新的dump文件
Wall time: 30.1 s

建立简单晶体结构

1
2
3
%%time
FCC = mp.LatticeMaker(4.05, 'FCC', 50, 50, 50)
FCC.compute()
Wall time: 506 ms

保存为data格式

1
2
%%time
FCC.write_data()
Wall time: 3.23 s