主要记录科研中遇到的通用数据处理过程

源代码下载：Deformation of Amorphous Polyethylene，不过里面的data不是以文件形式提供的，不太方便。

所以可以从我的资源里下载：deformTest.tar.gz，里面含有基本的源数据和对应的处理py脚本。

基本步骤

1. 得到数据文件（来源于各种仿真程序，通常可以导致指定格式的文件。批量读取文件的话，需要考虑到文件夹内文件的排序以及文件类型问题）
2. 得到文件中的数据（读取上述文件中的数据，需要考虑到数据的筛选清洗、以及时间步的平均调整）
3. 对数据进行可视化（哪怕excel都可以做出好看的图，只是综合来看可以完成以上步骤的同时还需要可视化，建议使用一门编程语言，例如Python）

软件推荐

IDE部分建议使用Spyder，界面和Matlab类似。

代码部分

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri Jan  8 12:37:24 2021

@author: yuanbaoqiang

This is a simple py to handle the data in multi txt files, mainly including three steps:
    1. ENSURE YOUR DATA FILES
    2. OBTAIN YOUR DATA
    3. PLOT
"""

# Import modules you need to deal with your files
# glob,os  ---> file processing
import glob,os
# pandas ----> data processing
import pandas as pd 
# matplotlib ----> plot
import matplotlib.pyplot as plt

########################################################
# 1. FIND THE PATH, AND LOAD NEEDED FILES IN THIS PATH

# To get the py file's absolute path 
path=os.getcwd()
# Add all .txt files in the path above
file=glob.glob(os.path.join("*.txt"))
# Sort all .txt files you have added
file.sort()
# You can also sort all files by using the keyword in your file name with the lamba experession below
# file.sort(key = lambda x: int(x[13:-4]))
print('You have added ' + str(len(file)) + ' files, and they are ' + str(file))

# Define a empty list
# List named "df" is to store your data in every file seperately in a dataframe element
df = []
# List named "names" is to define column names of every dataframe
names = []
names1 = ['strain', 'pxx', 'pyy', 'pzz', 'lx', 'ly', 'lz' , 'temp'];
names2 = ['strain', 'epair', 'ebond', 'eangle', 'edihed']
names.append(names1)
names.append(names2)

########################################################
# 2. OBTAIN THE DATA IN FILES YOU HAVE LOADED

# Traverse the list and obtain the needed frame in each file
i = 0
for f in file:
    # Official interpretation about "pandas.read_csv", which is the most "read" command I have used
    # https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_csv.html
    df.append(pd.read_csv(f, delim_whitespace=True, names=names[i], skiprows=[0]))
    i = i + 1

########################################################
# 3. JUST PLOT USING THE DATA YOU NEEDED
    
# Plot a simple graph, and you can see a rough prototype like the "strain-stress" curve in the paper:
# 10.1016/j.polymer.2010.10.009
plt.plot(df[0][:]['strain'], df[0][:]['pxx'])

草图输出

不含数据的处理，只是一个基本草图输出，但从趋势来看是和论文大致相同的。

10.1016/j.polymer.2010.10.009论文中的图：

福利

本群在lammps学习过程中给予了我莫大的帮助，群里很活跃，欢迎小伙伴的加入~

欢迎加入qq群：698671585

欢迎关注群主的公众号，lammps_zhushou，文章持续更新中~