Merge pull request #74 from LDAR-Sim/73-feat-initialize-number-of-leaks

73 feat initialize number of leaks

LDAR_Sim/src/default_parameters/p_default.yml

@@ -24,6 +24,8 @@ emissions:
   subtype_leak_dist_file: "_placeholder_str_"
   units: ["kilogram", "hour"]
 NRd: 365
+n_init_leaks: "_placeholder_int_"
+n_init_days: "_placeholder_int_"
 economics:
   carbon_price_tonnesCO2e: 40.0
   cost_CCUS: 20.0

LDAR_Sim/src/initialization/leaks.py

@@ -125,14 +125,23 @@ def generate_initial_leaks(program, site):
         NRd = program['NRd']
         LPR = program['emissions']['LPR']
 
-    n_leaks = random.binomial(NRd, LPR)
+    if program['n_init_leaks'] is not None:
+        n_leaks = program['n_init_leaks']
+    else:
+        n_leaks = random.binomial(NRd, LPR)
+
+    if program['n_init_days'] is not None:
+        init_max_days_active = program['n_init_days']
+    else:
+        init_max_days_active = NRd
+
     prog_start_date = datetime(*program['start_date'])
     initial_site_leaks = []
     site.update({'initial_leaks': n_leaks, 'cum_leaks': n_leaks})
     leak_count = 0
     for leak in range(n_leaks):
         leak_count += 1
-        days_active = random.randint(0, high=NRd)
+        days_active = random.randint(0, high=init_max_days_active)
         leak_start_date = prog_start_date - timedelta(days=days_active)
         initial_site_leaks.append(
             generate_leak(program, site, leak_start_date, leak_count, days_active))

LDAR_Sim/src/out_processing/prog_table.py

@@ -87,21 +87,30 @@ def generate(sim_results, baseline_program, programs):
     leaks['mit_vol_kg'] = leaks['nat_volume_kg'] - leaks['volume_kg']
     # Get responsible companies and attribute leak mitigation to them.
     leaks['init_detect_by'] = leaks['init_detect_by'].replace([None], 'active')
-
-    sim_leaks_by_NRD = agg_flatten(leaks[leaks['init_detect_by'] == 'natural'],
-                                   ps_idx,
-                                   ['volume_kg'],
-                                   ['sum', 'count'],
-                                   prefix="nat_vol",
-                                   include_col_name=False, include_agg_name=True)
+    if not leaks[leaks['init_detect_by'] == 'natural'].empty:
+        sim_leaks_by_NRD = agg_flatten(leaks[leaks['init_detect_by'] == 'natural'],
+                                       ps_idx,
+                                       ['volume_kg'],
+                                       ['sum', 'count'],
+                                       prefix="nat_vol",
+                                       include_col_name=False, include_agg_name=True)
+
+        sim_progs = merge(sim_stats_daily, sim_leaks_by_NRD,
+                          how='left', on=["program_name", 'sim'])
+    else:
+        # these columns are created by the agg_flatten function call above.
+        # future work should initialize these columns all in one location instead of changing
+        # the data strcuture mid way through the code.
+        sim_progs = sim_stats_daily
+        sim_progs['nat_vol_sum'] = float('nan')
+        sim_progs['nat_vol_count'] = float('nan')
+        sim_progs['emis_nat_perc'] = float('nan')
 
     sim_emis_mit = agg_flatten(leaks, ps_idx,
                                ['mit_vol_kg', 'volume_kg'],
                                agg_types=['sum'],
                                include_col_name=True, include_agg_name=True)
 
-    sim_progs = merge(sim_stats_daily, sim_leaks_by_NRD,
-                      how='left', on=["program_name", 'sim'])
     sim_progs = merge(sim_progs, sim_emis_mit, how='left',
                       on=["program_name", 'sim'])
     sim_progs = merge(sim_progs, sim_sites_count,
@@ -119,14 +128,14 @@ def generate(sim_results, baseline_program, programs):
     sim_progs['emis_nat_perc'] = sim_progs['nat_vol_sum'] / \
         sim_progs['volume_kg_sum']
     sim_progs.replace([inf, -inf], NaN, inplace=True)
-
     sim_progs = sim_progs.rename(columns={
         'cost_day_sum': 'cost', 'nat_vol_count': 'nat_leak_repair_count'})
     out_table = sim_progs[[
         'program_name', 'sim', 'emis_kg_day_site_med', 'act_leaks_day_site_med',
         'mit_vol_tco2e', 'mit_vol_perc', 'cost', 'cost_mit_vol_tco2e',
         'nat_leak_repair_count', 'emis_nat_perc'
     ]]
+
     out_table = agg_flatten(out_table, ['program_name'], agg_types=['mean'],
                             include_col_name=True, include_agg_name=False)
     out_table = out_table.drop(columns=['sim'])

LDAR_Sim/testing/unit_testing/test_initialization/test_leaks/leak_testing_fixtures.py

@@ -0,0 +1,83 @@
+""" Fixtures for testing leaks"""
+
+import pytest
+from datetime import datetime
+from typing import Any, Dict
+
+
+@pytest.fixture(name="mock_program")
+def mock_program_fix():
+    return {
+        'subtype_file': None,
+        'NRd': 10,
+        'n_init_leaks': None,
+        'n_init_days': None,
+        'start_date': (2022, 1, 1),
+        'emissions': {
+            'LPR': 0.5,
+            'leak_file': None,
+        },
+        'leak_rate_source': 'sample',
+        'empirical_leak_rates': [1],
+
+    }
+
+
+@pytest.fixture(name="mock_program_2")
+def mock_program_2_fix():
+    return {
+        'subtype_file': None,
+        'NRd': 10,
+        'n_init_leaks': 1,
+        'n_init_days': 10,
+        'start_date': (2022, 1, 1),
+        'emissions': {
+            'LPR': 0.5,
+            'leak_file': None,
+        },
+        'leak_rate_source': 'sample',
+        'empirical_leak_rates': [1],
+    }
+
+
+@pytest.fixture(name="mock_site_for_leak_test")
+def mock_site_for_leak_test_fix() -> Dict[str, Any]:
+    return {
+        'facility_ID': 'test',
+        'equipment_groups': 1,
+        'empirical_leak_rates': [1],
+        'lat': 52.0,
+        'lon': -114.0,
+        'leak_rate_source': 'sample',
+    }
+
+
+@pytest.fixture(name="mock_site_return_test")
+def mock_site_return_test_fix() -> list[Dict[str, Any]]:
+    return [{
+        'leak_ID': 'test_0000000001',
+        'facility_ID': 'test',
+        'equipment_group': 0,
+        'rate': 1,
+        'lat': 52.0,
+        'lon': -114.0,
+        'status': 'active',
+        'days_active': 0,
+        'volume': None,
+        'estimated_volume': None,
+        'estimated_volume_b': None,
+        'measured_rate': None,
+        'tagged': False,
+        'component': 'unknown',
+        'date_began': datetime(2022, 1, 1, 0, 0),
+        'day_ts_began': 0,
+        'estimated_date_began': None,
+        'date_tagged': None,
+        'tagged_by_company': None,
+        'tagged_by_crew': None,
+        'init_detect_by': None,
+        'init_detect_date': None,
+        'requires_shutdown': False,
+        'date_repaired': None,
+        'repair_delay': None,
+    }]

LDAR_Sim/testing/unit_testing/test_initialization/test_leaks/test_generate_init_leaks.py

@@ -0,0 +1,33 @@
+"""
+Module to test generate_initial_leaks
+"""
+
+from src.initialization.leaks import generate_initial_leaks
+from testing.unit_testing.test_initialization.test_leaks.leak_testing_fixtures import (
+    mock_site_for_leak_test_fix,
+    mock_site_return_test_fix,
+    mock_program_fix,
+    mock_program_2_fix,
+)
+
+
+def test_042_initialize_leaks(mock_site_for_leak_test, mock_program, mock_site_return_test, mocker):
+    mocker.patch('src.initialization.leaks.random.binomial', return_value=1)
+    mocker.patch('src.initialization.leaks.random.randint', return_value=0)
+
+    # Call the function under test
+    result = generate_initial_leaks(mock_program, mock_site_for_leak_test)
+
+    # Assert the expected result
+    assert result == mock_site_return_test
+
+
+def test_042_initialize_leaks_2(mock_site_for_leak_test, mock_program_2, mock_site_return_test, mocker):
+    mocker.patch('src.initialization.leaks.random.binomial', return_value=1)
+    mocker.patch('src.initialization.leaks.random.randint', return_value=0)
+
+    # Call the function under test
+    result = generate_initial_leaks(mock_program_2, mock_site_for_leak_test)
+
+    # Assert the expected result
+    assert result == mock_site_return_test

USER_MANUAL.md

@@ -935,6 +935,30 @@ method_labels:
 
 **Notes of caution:** N/A
 
+### n_init_leaks
+
+**Data type:** Integer
+
+**Default input:** N/A
+
+**Description:** The number of leaks present per site in the system at the time of the simulation start date
+
+**Notes on acquisition:** Estimate from empirical data or use previously published value. This value can be calculated by multiplying the n_init_days by LPR.
+
+**Notes of caution:** This will overwrite the original functionality of utilizing NRd and LPR to calculate the number of initial leaks present in the simulation. Additionally, this value is will be set for all sites, and cannot be specified for individual subtypes at this given time.
+
+### n_init_days
+
+**Data type:** Integer
+
+**Default input:** N/A (will be set to NRd if left blank)
+
+**Description:** The number of days the initial leaks could potentially been active for before the simulation started. Recommended to always set this value between 0 and NRd.
+
+**Notes on acquisition:** Known assumptions based on operations.
+
+**Notes of caution:** If value is larger than NRd, the number of leaks present in the initial system could potentially be less than the indicated value.
+
 ### NRD
 
 **Data type:** Integer
@@ -951,6 +975,8 @@ Can alternatively be provided in the ```subtype_file```, which will overwrite wh
 
 The NRd value should be the same for **all** programs
 
+Additionally, NRd is recommended to be always be equal or greater than n_init_days.
+
 ### program_name
 
 **Data type:** String

pytest.ini

@@ -1,3 +1,4 @@
 [pytest]
 python_files = test_*.py
 pythonpath = LDAR_Sim/src LDAR_Sim/external_sensors
+