| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128 |
- # import datetime
- from collections import defaultdict
- import pandas as pd
- class ProccessingPrismaCl:
- def __init__(self, n_data):
- self.n_data = n_data
- def period_processing_for_report(self, start_date, end_date):
- worktime_dict = defaultdict(list)
- n_vs_zero_tr_dict = defaultdict(list)
- event_counter_fr_4 = defaultdict(list)
- breaks_dict = defaultdict(list)
- count_rate_amp_5_fr_2 = defaultdict(list)
- count_rate_amp_10_fr_1 = defaultdict(list)
- for single_date in pd.date_range(start_date, end_date):
- worktime_dict['Date'].append(single_date)
- n_vs_zero_tr_dict['Date'].append(single_date)
- count_rate_amp_5_fr_2['Date'].append(single_date)
- count_rate_amp_10_fr_1['Date'].append(single_date)
- event_counter_fr_4['Date'].append(single_date)
- single_n_data = self.n_data[self.n_data['Date'] == single_date].reset_index(drop=True)
- if len(single_n_data) == 0:
- worktime_dict['Worktime'].append(0.00)
- for i in range(16):
- n_vs_zero_tr_dict[f'n{i + 1}'].append(0.00)
- count_rate_amp_5_fr_2[f'amp{i + 1}'].append(0.00)
- count_rate_amp_10_fr_1[f'amp{i + 1}'].append(0.00)
- event_counter_fr_4['Events'].append(0.00)
- continue
- break_time_dict, worktime_item = self._counting_break_time(single_n_data, delta_time_crit=600)
- worktime_dict['Worktime'].append(worktime_item)
- # print(break_time_dict)
- if break_time_dict:
- breaks_dict['Date'].extend([single_date.date()] * len(break_time_dict['StartSeconds']))
- breaks_dict['StartSeconds'].extend(break_time_dict['StartSeconds'])
- breaks_dict['EndSeconds'].extend(break_time_dict['EndSeconds'])
- neutron_to_zero_trigger = self._neutron_to_zero_trigger(single_n_data)
- for i in range(16):
- n_vs_zero_tr_dict[f'n{i + 1}'].append(neutron_to_zero_trigger[i])
- count_rate_amp_5_fr_2[f'amp{i + 1}'].append(
- self._set_event_counter(single_n_data, a_crit=6, freq=2)['count_rate'][i + 1])
- count_rate_amp_10_fr_1[f'amp{i + 1}'].append(
- self._set_event_counter(single_n_data, a_crit=11, freq=1)['count_rate'][i + 1])
- event_counter_fr_4['Events'].append(
- self._set_event_counter(single_n_data, a_crit=6, freq=4)['sum_events'])
- amp_5_fr_2_frame = self.set_amp_df(a_crit=6, freq=2)
- amp_10_fr_1_frame = self.set_amp_df(a_crit=11, freq=1)
- worktime_frame = pd.DataFrame(worktime_dict)
- n_vs_zero_tr_frame = pd.DataFrame(n_vs_zero_tr_dict)
- breaks_frame = pd.DataFrame(breaks_dict)
- event_counter_fr_4 = pd.DataFrame(event_counter_fr_4)
- count_rate_amp_5_fr_2 = pd.DataFrame(count_rate_amp_5_fr_2)
- count_rate_amp_10_fr_1 = pd.DataFrame(count_rate_amp_10_fr_1)
- for column in [f'amp{i}' for i in range(1, 17)]:
- count_rate_amp_5_fr_2[column] = count_rate_amp_5_fr_2[column] / worktime_frame['Worktime']
- count_rate_amp_10_fr_1[column] = count_rate_amp_10_fr_1[column] / worktime_frame['Worktime']
- return worktime_frame, breaks_frame, n_vs_zero_tr_frame, event_counter_fr_4, count_rate_amp_5_fr_2, count_rate_amp_10_fr_1, amp_5_fr_2_frame, amp_10_fr_1_frame
- @staticmethod
- def _counting_break_time(n_file, delta_time_crit):
- """Метод, выявляющий в n-file 5-минутки, когда кластер не работал, возвращает начальное время остановки и
- конечное время остановки"""
- time_difference = n_file['time'].diff()
- daily_breaks_dict = defaultdict(list)
- worktime_item = 24.00
- # print(f'{time_difference=}')
- for i in range(1, len(time_difference)):
- if time_difference[i] > delta_time_crit:
- daily_breaks_dict['StartSeconds'].append(n_file['time'][i - 1])
- daily_breaks_dict['EndSeconds'].append(n_file['time'][i])
- if n_file['time'][0] > delta_time_crit:
- daily_breaks_dict['StartSeconds'].append(0)
- daily_breaks_dict['EndSeconds'].append(n_file['time'][0])
- if n_file['time'][len(n_file['time']) - 1] < 86400 - delta_time_crit:
- daily_breaks_dict['StartSeconds'].append(n_file['time'][len(n_file['time']) - 1])
- daily_breaks_dict['EndSeconds'].append(86399)
- if daily_breaks_dict:
- worktime_item = worktime_item - round(sum(
- [daily_breaks_dict['EndSeconds'][index] - daily_breaks_dict['StartSeconds'][index] for index in
- range(len(daily_breaks_dict['StartSeconds']))]) / 3600, 2)
- return daily_breaks_dict, worktime_item
- else:
- return None, worktime_item
- @staticmethod
- def _neutron_to_zero_trigger(n_file):
- """Метод, обрабатывающий данные n-файлов ПРИЗМА-32, дающий на выходе нормированное число в событии,
- отобранных как нейтрон, при самозапуске"""
- counter_zero_tr = len(n_file[n_file['trigger'] == 0].index)
- zero_tr_frame = n_file[n_file['trigger'] == 0]
- return [round(zero_tr_frame[f'n{i}'].sum() / counter_zero_tr, 3) for i in range(1, 17)]
- # except ZeroDivisionError: Нужно дописать, если допустим нет нулевых триггеров
- @staticmethod
- def _set_event_counter(n_file, a_crit, freq):
- """Метод, обрабатывающий данные n-файлов ПРИЗМА-32, на вход подаются определенная амплитуда и количество
- детекторов, на которых амплитуда превышает заданную, на выходе метод возвращает словарь с параметрами:
- 1) суммарное число событий на кластере, подходящих под заданные условия;
- 2) датафрейм с амплитудами детекторов для каждого события, подходящего под заданные условия;
- 3) количество превышений заданной амплитуды у детектора в событиях, подходящих под заданные условия; """
- cluster_count_rate = {}
- amp_frame = n_file[[f'amp{i}' for i in range(1, 17)]]
- amp_frame['fr_sum'] = amp_frame.isin(range(a_crit, 520)).sum(axis=1, skipna=True) # noqa
- amp_frame = amp_frame[amp_frame['fr_sum'] >= freq].reset_index(drop=True)
- for i in range(1, 17):
- cluster_count_rate[i] = len(amp_frame[amp_frame[f'amp{i}'] >= a_crit])
- return {'sum_events': len(amp_frame.index),
- 'count_rate': cluster_count_rate}
- def set_amp_df(self, a_crit, freq):
- amp_frame = self.n_data[[f'amp{i}' for i in range(1, 17)]]
- amp_frame['fr_sum'] = amp_frame.isin(range(a_crit, 520)).sum(axis=1, skipna=True) # noqa
- amp_frame = amp_frame[amp_frame['fr_sum'] >= freq].reset_index(drop=True)
- return amp_frame[[f'amp{i}' for i in range(1, 17)]]
|
