# 读取数据 import numpy as np import pandas as pd import preprocess.preprocess import importlib import ana.parameters import datetime import os import matplotlib import openpyxl file_path = r'D:\Work\Interior\Cloud_BMS\data_ana\7250.csv' sn_list = pd.read_csv(file_path,encoding='gbk') sn_list = sn_list.values.tolist() # print(sn_list) sn_list = [['PK504B10100004301']] df_res = pd.DataFrame(columns=['sn', 'date', 'odo', 'capacity', 'energy', 'energy_rate', 'acc_time_drive', 'acc_time_stand', 'acc_time_charge', 'avr_stand_tem', 'avr_stand_tem_rate', 'max_pwr', 'regen_flag', 'regen_rate', 'avr_speed', 'max_speed', 'avr_drive_tem', 'avr_drive_tem_rate', 'center_long', 'center_lat', 'rad']) df_soh_res = pd.DataFrame(columns=['sn', 'time', 'accum_ah', 'soh_cell01', 'soh_cell02', 'soh_cell03', 'soh_cell04', 'soh_cell05', 'soh_cell06', 'soh_cell07', 'soh_cell08', 'soh_cell09', 'soh_cell10', 'soh_cell11', 'soh_cell12', 'soh_cell13', 'soh_cell14', 'soh_cell15', 'soh_cell16', 'soh_cell17', 'soh_cell18', 'soh_cell19', 'soh_cell20']) df_cell_res = pd.DataFrame(columns=['sn', 'time', 'status', 'volt_cell01', 'volt_cell02', 'volt_cell03', 'volt_cell04', 'volt_cell05', 'volt_cell06', 'volt_cell07', 'volt_cell08', 'volt_cell09', 'volt_cell10', 'volt_cell11', 'volt_cell12', 'volt_cell13', 'volt_cell14', 'volt_cell15', 'volt_cell16', 'volt_cell17', 'volt_cell18', 'volt_cell19', 'volt_cell20', 'volt_max', 'volt_min', 'volt_avr', 'soc_diff', 'soc_diff_div', 'over_discharge']) for sn in sn_list: # 数据读取 base_path = r'D:\Work\Interior\Cloud_BMS\data_ana\Data_Files' time = '_from_2020-04-27_to_2021-04-27.csv' bms_file = base_path + r'\\' + sn[0] + '\BMS_' + sn[0] + time gps_file = base_path + r'\\' + sn[0] + '\GPS_' + sn[0] + time # print(bms_file) if not os.path.exists(bms_file): print('{} bms file does not exist!!!!!!'.format(sn[0])) continue if not os.path.exists(gps_file): print('{} gps file does not exist!!!!!!'.format(sn[0])) continue df_bms = pd.read_csv(bms_file, encoding='GB2312') df_gps = pd.read_csv(gps_file, encoding='GB2312') df_bms = df_bms.drop_duplicates(['时间戳']) df_gps = df_gps.drop_duplicates(['时间戳']) df_bms = df_bms.reset_index(drop=True) df_gps = df_gps.reset_index(drop=True) df_bms['时间戳'] = pd.to_datetime(df_bms['时间戳']) df_gps['时间戳'] = pd.to_datetime(df_gps['时间戳']) print('{} data read Well Done!'.format(sn[0])) df_bms.to_csv('result2.csv', index=False, encoding='GB2312') # 数据预处理 # 1、数据分段 # 2、GPS数据可靠与否的判断(根据GPS计算出的平均车速<2认为可靠) importlib.reload(preprocess.preprocess) df_bms = preprocess.preprocess.data_split(df_bms) df_bms.to_csv('result.csv',encoding='GB2312') df_bms, df_gps, prepro_record = preprocess.preprocess.data_gps_judge(df_bms, df_gps, 2) print('{} data preprocess Well Done!'.format(sn[0])) df_bms.to_csv('result.csv', index=False, encoding='GB2312') df_gps.to_csv('result1.csv', index=False, encoding='GB2312') importlib.reload(ana.parameters) # test = ana.parameters.sta_one_drive_cycle(df_bms, df_gps, prepro_record, time_window=3600*24, step=3600*24, start_time="00:00:00") # print(test) # 计算Daily数据 start_time = df_bms.loc[0, '时间戳'] # print(start_time) cur_time = start_time index = 0 while cur_time < df_bms.loc[len(df_bms) - 1, '时间戳']: if cur_time.day != start_time.day: break else: index = index + 1 cur_time = df_bms.loc[index, '时间戳'] start = '00:00:00' start_time = datetime.datetime.strptime(str(df_bms.loc[index, '时间戳'])[0:10] + ' ' + start, '%Y-%m-%d %H:%M:%S') # print(index, start_time) timeDelta = datetime.timedelta(days=1) end_time = start_time + timeDelta # print(start_time, end_time) while end_time < df_bms.loc[len(df_bms) - 1, '时间戳']: # while end_time < df_bms.loc[8500, '时间戳']: df_sel_bms = df_bms[(df_bms['时间戳'] >= start_time) & (df_bms['时间戳'] <= end_time)] df_sel_bms = df_sel_bms.reset_index(drop=True) df_sel_gps = df_gps[(df_gps['时间戳'] >= start_time) & (df_gps['时间戳'] <= end_time)] df_sel_gps = df_sel_gps.reset_index(drop=True) if len(df_sel_bms) == 0: continue date = ana.parameters.get_date(df_sel_bms) odo_total, odo_sum, odo_plus = ana.parameters.get_daily_odo(df_sel_bms, df_sel_gps) print('total={},odo={},plus={}'.format(odo_total, odo_sum, odo_plus)) # 计算单日里程，单日平均车速，单日最高车速 # print(start_time) odo, avr_speed, max_speed = ana.parameters.get_daily_odo_and_speed(df_sel_bms, df_sel_gps) # 计算单日使用容量，单日使用能量 capacity, energy = ana.parameters.get_daily_capacity_and_energy(df_sel_bms, 55) # 计算单日平均电耗：kwh/100km if odo > 10: energy_rate = energy / odo * 100 else: energy_rate = 0 # 计算单日累积时间（行驶、静置、充电） acc_time_drive = ana.parameters.get_daily_accum_time(df_sel_bms, 'drive') acc_time_stand = ana.parameters.get_daily_accum_time(df_sel_bms, 'stand') acc_time_charge = ana.parameters.get_daily_accum_time(df_sel_bms, 'charge') # 计算过程平均温度，平均温升（静置、行驶） avr_stand_tem, avr_stand_tem_rate = ana.parameters.get_daily_stand_temp(df_sel_bms, 'stand') avr_drive_tem, avr_drive_tem_rate = ana.parameters.get_daily_stand_temp(df_sel_bms, 'drive') # 计算车辆功率 max_pwr = ana.parameters.get_daily_max_pwr(df_sel_bms, 'drive') # 计算regen数据，是否有regen以及regen比例 regen_flag, regen_rate = ana.parameters.get_daily_regen(df_sel_bms, 'drive') # 计算活动范围，圆心和半径 center_long, center_lat, rad = ana.parameters.get_working_scope(df_sel_bms, df_sel_gps, 'drive') df_res = df_res.append({'sn': sn[0], 'date': date, 'odo': odo, 'capacity': capacity, 'energy': energy, 'energy_rate': energy_rate, 'acc_time_drive': acc_time_drive, 'acc_time_stand': acc_time_stand, 'acc_time_charge': acc_time_charge, 'avr_stand_tem': avr_stand_tem, 'avr_stand_tem_rate': avr_stand_tem_rate, 'max_pwr': max_pwr, 'regen_flag': regen_flag, 'regen_rate': regen_rate, 'avr_speed': avr_speed, 'max_speed': max_speed, 'avr_drive_tem': avr_drive_tem, 'avr_drive_tem_rate': avr_drive_tem_rate, 'center_long': center_long, 'center_lat': center_lat, 'rad': rad}, ignore_index=True) start_time = end_time end_time = start_time + timeDelta # print(end_time,df_bms.loc[len(df_bms) - 1, '时间戳']) print('{} daily data process Well Done!'.format(sn[0])) # 计算电池周期性数据 df_soh_res_single, df_cell_res_single = ana.parameters.get_soh(df_bms, sn) df_soh_res = pd.concat([df_soh_res, df_soh_res_single], ignore_index=True) df_cell_res = pd.concat([df_cell_res, df_cell_res_single], ignore_index=True) print('{} period data process Well Done!'.format(sn[0])) path = r'D:\Work\Interior\Cloud_BMS\data_ana\report\daily_data_analysis.csv' df_res.to_csv(path, index=False, encoding='GB2312') path = r'D:\Work\Interior\Cloud_BMS\data_ana\report\period_data_analysis.xlsx' writer = pd.ExcelWriter(path) df_soh_res.to_excel(writer, "SOH数据") df_cell_res.to_excel(writer, "一致性数据") writer.save()