energy-data-trans/tmp_file/大唐玉湖数据整理.py

# -*- coding: utf-8 -*-
"""
Created on Tue Jul  9 16:28:48 2024

@author: Administrator
"""
import multiprocessing
from datetime import datetime
from os import *

import chardet
import pandas as pd

pd.options.mode.copy_on_write = True


# 获取文件编码
def detect_file_encoding(filename):
    # 读取文件的前1000个字节（足够用于大多数编码检测）
    with open(filename, 'rb') as f:
        rawdata = f.read(1000)
    result = chardet.detect(rawdata)
    encoding = result['encoding']

    if encoding is None:
        encoding = 'gb18030'

    if encoding and encoding.lower() == 'gb2312' or encoding.lower().startswith("windows"):
        encoding = 'gb18030'
    return encoding


# 读取数据到df
def read_file_to_df(file_path, read_cols=list(), header=0):
    df = pd.DataFrame()
    if str(file_path).lower().endswith("csv") or str(file_path).lower().endswith("gz"):
        encoding = detect_file_encoding(file_path)
        end_with_gz = str(file_path).lower().endswith("gz")
        if read_cols:
            if end_with_gz:
                df = pd.read_csv(file_path, encoding=encoding, usecols=read_cols, compression='gzip', header=header)
            else:
                df = pd.read_csv(file_path, encoding=encoding, usecols=read_cols, header=header, on_bad_lines='warn')
        else:

            if end_with_gz:
                df = pd.read_csv(file_path, encoding=encoding, compression='gzip', header=header)
            else:
                df = pd.read_csv(file_path, encoding=encoding, header=header, on_bad_lines='warn')

    else:
        xls = pd.ExcelFile(file_path)
        # 获取所有的sheet名称
        sheet_names = xls.sheet_names
        for sheet in sheet_names:
            if read_cols:
                df = pd.concat([df, pd.read_excel(xls, sheet_name=sheet, header=header, usecols=read_cols)])
            else:
                df = pd.concat([df, pd.read_excel(xls, sheet_name=sheet, header=header)])

    return df


def __build_directory_dict(directory_dict, path, filter_types=None):
    # 遍历目录下的所有项
    for item in listdir(path):
        item_path = path.join(path, item)
        if path.isdir(item_path):
            __build_directory_dict(directory_dict, item_path, filter_types=filter_types)
        elif path.isfile(item_path):
            if path not in directory_dict:
                directory_dict[path] = []

            if filter_types is None or len(filter_types) == 0:
                directory_dict[path].append(item_path)
            elif str(item_path).split(".")[-1] in filter_types:
                if str(item_path).count("~$") == 0:
                    directory_dict[path].append(item_path)

    # 读取所有文件


# 读取路径下所有的excel文件
def read_excel_files(read_path):
    directory_dict = {}
    __build_directory_dict(directory_dict, read_path, filter_types=['xls', 'xlsx', 'csv', 'gz'])

    return [path for paths in directory_dict.values() for path in paths if path]


# 创建路径
def create_file_path(path, is_file_path=False):
    if is_file_path:
        path = path.dirname(path)

    if not path.exists(path):
        makedirs(path, exist_ok=True)


def generate_df(pv_df, col):
    if col != '时间':
        xiangbian = col.split("逆变器")[0].replace("#", "")
        nibianqi = col.split("-")[0].split('逆变器')[1]
        pv_index = col.split("-")[1].replace("PV", "")
        now_df = pv_df[['时间', col + '输入电流()', col + '输入电压()']]
        now_df.loc[:, '箱变'] = xiangbian
        now_df.loc[:, '逆变器'] = nibianqi
        now_df.loc[:, 'PV'] = pv_index
        now_df.columns = [df_col.replace(col, "").replace("()", "") for df_col in now_df.columns]
        now_df['输入电流'] = now_df['输入电流'].astype(float)
        now_df['输入电压'] = now_df['输入电压'].astype(float)

        print(xiangbian, nibianqi, pv_index, now_df.shape)
        return now_df
    return pd.DataFrame()


def read_and_save_csv(file_path, save_path):
    begin = datetime.now()
    base_name = path.basename(file_path)
    print('开始', base_name)

    df = read_file_to_df(file_path)
    df['时间'] = pd.to_datetime(df['时间'])
    # df.set_index(keys='时间', inplace=True)

    pv_df_cols = [col for col in df.columns if col.find('输入电') > -1]
    pv_df_cols.append('时间')
    pv_df = df[pv_df_cols]
    shuru_cols = set([col.split("输入电")[0] for col in pv_df.columns])

    with multiprocessing.Pool(6) as pool:
        dfs = pool.starmap(generate_df, [(pv_df, col) for col in shuru_cols])

    saved_pv_df = pd.concat(dfs)
    saved_pv_df.sort_values(by=['箱变', '逆变器', 'PV', '时间'], inplace=True)
    save_file = path.join(save_path, path.basename(file_path).split(".")[0], 'PV.csv')
    create_file_path(save_file, True)

    saved_pv_df.to_csv(save_file, encoding='utf-8', index=False)

    print('结束', base_name, '耗时:' + str(datetime.now() - begin))


if __name__ == '__main__':
    path = r'D:\trans_data\大唐玉湖性能分析离线分析\test\yuanshi'
    save_path = r'D:\trans_data\大唐玉湖性能分析离线分析\test\zhengli'
    all_files = read_excel_files(path)

    all_datas = list(all_files)
    all_datas.sort()
    print(all_datas)

    for file in all_datas:
        read_and_save_csv(file, save_path)

    # with Pool(1) as pool:
    #     pool.starmap(read_and_save_csv, [(i, save_path) for i in all_datas])