| 12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061 |
- import pandas as pd
- import numpy as np
- import logging
- class DataClean:
- def __init__(self, df_algo_pack_param) -> None:
- self.df_algo_pack_param = df_algo_pack_param
- def datacleaning(self, df_in):
- if not df_in.empty:
- # df['Time'] = pd.to_datetime(list(df['Time']), utc=True, unit='ms').tz_convert('Asia/Shanghai')
- df_in=df_in.replace('[]', np.nan)
- df_in.dropna(axis=0,subset = ["time", "sn", "cell_voltage", "cell_temp", "pack_crnt"], inplace=True)
- df_in['time'] = pd.to_datetime(df_in['time'], format='%Y-%m-%d %H:%M:%S')
- df_in.drop(df_in.index[(df_in['pack_volt'] < 0.001) | (df_in['pack_volt'] > 1000) | (df_in['pack_soc'] > 100) | (df_in['pack_soc'] < 0) | (df_in['pack_crnt'] > 1000) | (df_in['pack_crnt'] < -1000)], inplace=True)
- if not df_in.empty:
- df_in = df_in.groupby('sn',group_keys=False).apply(lambda x:x.sort_values('time'))
- df_in.reset_index(drop=True, inplace=True)
- #电压、温度分列
- CellVoltNums=int(self.df_algo_pack_param['CellVoltTotalCount'])
- CellTempNums = int(self.df_algo_pack_param['CellTempTotalCount'])
- cellvolt_name=['cell_voltage'+str(x) for x in range(1, CellVoltNums+1)]
- celltemp_name=['cell_temp'+str(x) for x in range(1, CellTempNums+1)]
- df_volt = df_in['cell_voltage'].apply(lambda x : pd.Series(list(x)[:CellVoltNums]))
- df_volt.columns = cellvolt_name
- df_volt=df_volt.astype('float')
- cellvoltmax = df_volt.max(axis=1)
- cellvoltmin = df_volt.min(axis=1)
- df_volt[['cell_volt_max','cell_volt_min']] = pd.concat([cellvoltmax,cellvoltmin], axis=1)
- df_temp = df_in['cell_temp'].apply(lambda x : pd.Series(list(x)[:CellTempNums]))
- df_temp.columns = celltemp_name
- df_temp=df_temp.astype('float')
- celltempmax = df_temp.max(axis=1)
- celltempmin = df_temp.min(axis=1)
- df_temp[['cell_temp_max','cell_temp_min']] = pd.concat([celltempmax,celltempmin], axis=1)
- #其他温度分列
- if len(df_in['other_temp_value'].loc[0]):
- df_otherTemp_name=['mos_temp', 'env_temp', 'fastcharg_connector_temp',
- 'onc_connector_temp', 'heat_plate1_temp', 'heat_plate2_temp', 'connector_1_temp','connector_2_temp', 'pcb_temp', 'bat_inner_temp']
- df_otherTemp=pd.DataFrame([list(x[0]) for x in np.array(df_in[['other_temp_value']])]).iloc[:,list(range(len(df_otherTemp_name)))]
- df_otherTemp.columns=df_otherTemp_name
- df_otherTemp=df_otherTemp.astype('float')
- df_out = pd.concat([df_in, df_volt, df_temp, df_otherTemp],axis=1)
- else:
- df_out = pd.concat([df_in, df_volt, df_temp],axis=1)
-
- df_out.dropna(axis=0, inplace=True)
- # df_out.dropna(axis=0,subset = cellvolt_name+celltemp_name, inplace=True)
- df_out.reset_index(inplace=True, drop=True)
- df_table = df_out.drop_duplicates(subset=['sn'], keep='first', ignore_index=True)
- df_table = df_table.set_index('sn')
- else:
- df_out = pd.DataFrame()
- df_table = pd.DataFrame()
- cellvolt_name = []
- celltemp_name = []
- return df_out, df_table, cellvolt_name, celltemp_name
- else:
- return pd.DataFrame(), pd.DataFrame(), [], []
|
