Demonstration of the Available Metrics

For a complete list of metrics and their documentation, please see the API Metrics documentation.

This demonstration will rely on the results produced in the "How To" notebook and serves as an extension of the API documentation to show what the results will look like depending on what inputs are provided.

A Jupyter notebook of this tutorial can be run from examples/metrics_demonstration.ipynb locally, or through binder.

from pprint import pprint
from functools import partial

import pandas as pd
from pandas.io.formats.style import Styler

from wombat.core import Simulation, Metrics
from wombat.utilities import plot

# Clean up the aesthetics for the pandas outputs
pd.set_option("display.max_rows", 30)
pd.set_option("display.max_columns", 10)
style = partial(
  Styler,
  table_attributes='style="font-size: 14px; grid-column-count: 6"',
  precision=2,
  thousands=",",
)

Table of Contents

Below is a list of top-level sections to demonstrate how to use WOMBAT's Metrics class methods and an explanation of each individual metric.

If you don't see a metric or result computation that is core to your work, please submit an issue with details on what the metric is, and how it should be computed.

• Setup: Running a simulation to gather the results

• Common Parameters: Explanation of frequently used parameter settings

• Availability: Time-based and energy-based availability

• Capacity Factor: Gross and net capacity factor

• Task Completion Rate: Task completion metrics

• Equipment Costs: Cost breakdowns by servicing equipment

• Service Equipment Utilization Rate: Utilization of servicing equipment

• Vessel-Crew Hours at Sea: Number of crew or vessel hours spent at sea

• Number of Tows: Number of tows breakdowns

• Labor Costs: Breakdown of labor costs

• Equipment and Labor Costs: Combined servicing equipment and labor cost breakdown

• Emissions: Emissions of servicing equipment based on activity

• Component Costs: Materials costs

• Fixed Cost Impacts: Total fixed costs

• OpEx: Project OpEx

• Process Times: Timing of various stages of repair and maintenance

• Power Production: Potential and actually produced power

• Net Present Value: Project NPV calculator

Setup

The simulations from the How To notebook are going to be rerun as it is not recommended to create a Metrics class from scratch due to the large number of inputs that are required, and the initialization is provided in the simulation API's run method.

To simplify this process, a feature has been added to save the simulation outputs required to generate the Metrics inputs and a method to reload those outputs as inputs.

sim = Simulation("COREWIND", "morro_bay_in_situ.yaml")

# Both of these parameters are True by default for convenience
sim.run(create_metrics=True, save_metrics_inputs=True)

# Load the metrics data
fpath = sim.env.metrics_input_fname.parent
fname = sim.env.metrics_input_fname.name
metrics = Metrics.from_simulation_outputs(fpath, fname)

# Delete the log files now that they're loaded in
sim.env.cleanup_log_files()

# Alternatively, in this case because the simulation was run, we can use the
# following for convenience convenience only
metrics = sim.metrics

Common Parameter Explanations

Before diving into each and every metric, and how they can be customized, it is worth noting some of the most common parameters used throughout, and their meanings to reduce redundancy. The varying output forms are demonstrated in the availability section below.

frequency

project

Computed across the whole simulation, with the resulting DataFrame having an empty index.

annual

Summary of each year in the simulation, with the resulting DataFrame having "year" as the index.

monthly

Summary of each month of the year, aggregated across years, with the resulting DataFrame having "month" as the index.

month-year

computed on a month-by-year basis, producing the results for every month of the simulation, with the resulting DataFrame having "year" and "month" as the index.

by

windfarm

Aggregated across all turbines, with the resulting DataFrame having only "windfarm" as a column

turbine

Computed for each turbine, with the resulting DataFrame having a column for each turbine

Availability

There are two methods to produce availability, which have their own function calls:

• energy: actual power produced divided by potential power produced

  • wombat.core.post_processor.Metrics.production_based_availability()

• time: The ratio of all non-zero hours to all hours in the simulation, or the proportion of the simulation where turbines are operating

  • wombat.core.post_processor.Metrics.time_based_availability()

Here, we will go through the various input definitions to get time-based availability data as both methods use the same inputs, and provide outputs in the same format.

Inputs:

• frequency, as explained above options: "project", "annual", "monthly", and "month-year"

• by, as explained above options: "windfarm" and "turbine"

Below is a demonstration of the variations on frequency and by for time_based_availability.

style(metrics.time_based_availability(frequency="project", by="windfarm"))

	windfarm
0	0.91

style(metrics.production_based_availability(frequency="project", by="windfarm"))

	windfarm
0	0.91

Note that in the two above examples, that the values are equal. This is due to the fact that the example simulation does not have any operating reduction applied to failures, unless it's a catastrophic failure, so there is no expected difference.

# Demonstrate the by turbine granularity
style(metrics.time_based_availability(frequency="project", by="turbine"))

	WTG_0000	WTG_0001	WTG_0002	WTG_0003	WTG_0004	WTG_0005	WTG_0006	WTG_0007	WTG_0008	WTG_0009	WTG_0100	WTG_0101	WTG_0102	WTG_0103	WTG_0104	WTG_0105	WTG_0106	WTG_0107	WTG_0108	WTG_0109	WTG_0200	WTG_0201	WTG_0202	WTG_0203	WTG_0204	WTG_0205	WTG_0206	WTG_0207	WTG_0208	WTG_0209	WTG_0300	WTG_0301	WTG_0302	WTG_0303	WTG_0304	WTG_0305	WTG_0306	WTG_0307	WTG_0308	WTG_0309	WTG_0400	WTG_0401	WTG_0402	WTG_0403	WTG_0404	WTG_0405	WTG_0406	WTG_0407	WTG_0408	WTG_0409	WTG_0500	WTG_0501	WTG_0502	WTG_0503	WTG_0504	WTG_0505	WTG_0506	WTG_0507	WTG_0508	WTG_0509	WTG_0600	WTG_0601	WTG_0602	WTG_0603	WTG_0604	WTG_0605	WTG_0606	WTG_0607	WTG_0608	WTG_0609	WTG_0700	WTG_0701	WTG_0702	WTG_0703	WTG_0704	WTG_0705	WTG_0706	WTG_0707	WTG_0708	WTG_0709
0	0.90	0.92	0.92	0.91	0.91	0.92	0.90	0.91	0.91	0.91	0.92	0.91	0.91	0.90	0.91	0.91	0.91	0.90	0.91	0.91	0.90	0.91	0.90	0.90	0.91	0.92	0.91	0.91	0.89	0.91	0.90	0.92	0.91	0.92	0.90	0.92	0.92	0.91	0.90	0.91	0.89	0.91	0.91	0.91	0.92	0.91	0.91	0.91	0.89	0.93	0.90	0.90	0.90	0.91	0.91	0.90	0.92	0.90	0.92	0.93	0.92	0.92	0.90	0.91	0.92	0.91	0.92	0.91	0.89	0.91	0.90	0.90	0.90	0.91	0.90	0.90	0.92	0.90	0.92	0.91

# Demonstrate the annualized outputs
style(metrics.time_based_availability(frequency="annual", by="windfarm"))

	windfarm
year
2002	0.96
2003	0.91
2004	0.91
2005	0.92
2006	0.91
2007	0.92
2008	0.90
2009	0.93
2010	0.90
2011	0.91
2012	0.86
2013	0.93
2014	0.88
2015	0.90
2016	0.90
2017	0.91
2018	0.91
2019	0.92
2020	0.92
2021	0.88

# Demonstrate the month aggregations
style(metrics.time_based_availability(frequency="monthly", by="windfarm"))

	windfarm
month
1	0.88
2	0.86
3	0.87
4	0.86
5	0.87
6	0.90
7	0.93
8	0.95
9	0.96
10	0.94
11	0.94
12	0.92

# Demonstrate the granular monthly reporting
style(metrics.time_based_availability(frequency="month-year", by="windfarm"))

		windfarm
year	month
2002	1	0.95
	2	0.97
	3	0.97
	4	0.94
	5	0.97
	6	0.93
	7	0.98
	8	0.98
	9	0.99
	10	0.97
	11	0.93
	12	0.92
2003	1	0.72
	2	0.90
	3	0.90
	4	0.89
	5	0.90
	6	0.90
	7	0.92
	8	0.97
	9	0.98
	10	0.94
	11	0.95
	12	0.92
2004	1	0.75
	2	0.89
	3	0.90
	4	0.90
	5	0.89
	6	0.89
	7	0.94
	8	0.98
	9	0.97
	10	0.97
	11	0.95
	12	0.92
2005	1	0.78
	2	0.93
	3	0.90
	4	0.89
	5	0.91
	6	0.91
	7	0.96
	8	0.97
	9	0.97
	10	0.94
	11	0.94
	12	0.95
2006	1	0.89
	2	0.74
	3	0.90
	4	0.89
	5	0.90
	6	0.90
	7	0.90
	8	0.92
	9	0.97
	10	0.97
	11	0.97
	12	0.93
2007	1	0.82
	2	0.93
	3	0.94
	4	0.90
	5	0.90
	6	0.90
	7	0.90
	8	0.94
	9	0.96
	10	0.96
	11	0.95
	12	0.93
2008	1	0.88
	2	0.61
	3	0.93
	4	0.93
	5	0.90
	6	0.90
	7	0.89
	8	0.92
	9	0.97
	10	0.96
	11	0.94
	12	0.94
2009	1	0.90
	2	0.88
	3	0.93
	4	0.91
	5	0.93
	6	0.93
	7	0.91
	8	0.96
	9	0.95
	10	0.95
	11	0.92
	12	0.95
2010	1	0.92
	2	0.75
	3	0.78
	4	0.89
	5	0.91
	6	0.91
	7	0.91
	8	0.94
	9	0.95
	10	0.93
	11	0.92
	12	0.93
2011	1	0.89
	2	0.89
	3	0.86
	4	0.92
	5	0.92
	6	0.92
	7	0.94
	8	0.91
	9	0.92
	10	0.92
	11	0.93
	12	0.93
2012	1	0.94
	2	0.56
	3	0.63
	4	0.79
	5	0.93
	6	0.91
	7	0.95
	8	0.94
	9	0.91
	10	0.94
	11	0.93
	12	0.90
2013	1	0.92
	2	0.94
	3	0.79
	4	0.93
	5	0.95
	6	0.92
	7	0.97
	8	0.96
	9	0.93
	10	0.93
	11	0.95
	12	0.95
2014	1	0.92
	2	0.94
	3	0.64
	4	0.67
	5	0.93
	6	0.90
	7	0.95
	8	0.97
	9	0.96
	10	0.92
	11	0.91
	12	0.90
2015	1	0.93
	2	0.93
	3	0.93
	4	0.63
	5	0.80
	6	0.97
	7	0.97
	8	0.98
	9	0.95
	10	0.94
	11	0.90
	12	0.89
2016	1	0.90
	2	0.89
	3	0.89
	4	0.77
	5	0.86
	6	0.91
	7	0.92
	8	0.97
	9	0.95
	10	0.92
	11	0.90
	12	0.90
2017	1	0.89
	2	0.90
	3	0.90
	4	0.82
	5	0.81
	6	0.90
	7	0.91
	8	0.97
	9	0.97
	10	0.94
	11	0.95
	12	0.94
2018	1	0.92
	2	0.91
	3	0.90
	4	0.88
	5	0.80
	6	0.92
	7	0.93
	8	0.96
	9	0.97
	10	0.92
	11	0.92
	12	0.89
2019	1	0.90
	2	0.90
	3	0.90
	4	0.90
	5	0.82
	6	0.91
	7	0.94
	8	0.97
	9	0.96
	10	0.94
	11	0.95
	12	0.92
2020	1	0.90
	2	0.90
	3	0.89
	4	0.92
	5	0.79
	6	0.94
	7	0.94
	8	0.96
	9	0.98
	10	0.96
	11	0.95
	12	0.90
2021	1	0.90
	2	0.89
	3	0.90
	4	0.90
	5	0.63
	6	0.73
	7	0.90
	8	0.93
	9	0.98
	10	0.94
	11	0.96
	12	0.96

Plotting Availability

As of v0.9, the ability to plot the wind farm and turbine availability has been enabled as an experimental feature. Please see the plotting API documentation for more details.

# Demonstrate the granular monthly reporting
plot.plot_farm_availability(sim=sim, which="energy", farm_95_CI=True)

Capacity Factor

The capacity factor is the ratio of actual (net) or potential (gross) energy production divided by the project's capacity. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.capacity_factor().

Inputs:

• which

  • "net": net capacity factor, actual production divided by the plant capacity

  • "gross": gross capacity factor, potential production divided by the plant capacity

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by, as explained above, options: "windfarm" and "turbine"

Example Usage:

net_cf = metrics.capacity_factor(which="net", frequency="project", by="windfarm").values[0][0]
gross_cf = metrics.capacity_factor(which="gross", frequency="project", by="windfarm").values[0][0]
print(f"  Net capacity factor: {net_cf:.2%}")
print(f"Gross capacity factor: {gross_cf:.2%}")

  Net capacity factor: 50.41%
Gross capacity factor: 55.59%

Task Completion Rate

The task completion rate is the ratio of tasks completed aggregated to the desired frequency. It is possible to have a >100% completion rate if all maintenance and failure requests submitted in a time period were completed in addition to those that went unfinished in prior time periods. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.task_completion_rate().

Inputs:

• which

  • "scheduled": scheduled maintenance only (classified as maintenance tasks in inputs)

  • "unscheduled": unscheduled maintenance only (classified as failure events in inputs)

  • "both": Combined completion rate for all tasks

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

Example Usage:

scheduled = metrics.task_completion_rate(which="scheduled", frequency="project").values[0][0]
unscheduled = metrics.task_completion_rate(which="unscheduled", frequency="project").values[0][0]
combined = metrics.task_completion_rate(which="both", frequency="project").values[0][0]
print(f"  Scheduled Task Completion Rate: {scheduled:.2%}")
print(f"Unscheduled Task Completion Rate: {unscheduled:.2%}")
print(f"    Overall Task Completion Rate: {combined:.2%}")

  Scheduled Task Completion Rate: 96.59%
Unscheduled Task Completion Rate: 96.21%
    Overall Task Completion Rate: 96.35%

Equipment Costs

Sum of the costs associated with a simulation's servicing equipment, which excludes materials, downtime, etc. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.equipment_costs().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_equipment

  • True: Aggregates all equipment into a single cost

  • False: Computes for each unit of servicing equipment

Example Usage:

# Project total at the whole wind farm level
style(metrics.equipment_costs(frequency="project", by_equipment=False))

	equipment_cost
0	2,348,371,239.26

# Project totals at servicing equipment level
style(metrics.equipment_costs(frequency="project", by_equipment=True))

	Anchor Handling Tug	Cable Laying Vessel	Crew Transfer Vessel 1	Crew Transfer Vessel 2	Crew Transfer Vessel 3	Crew Transfer Vessel 4	Crew Transfer Vessel 5	Crew Transfer Vessel 6	Crew Transfer Vessel 7	Diving Support Vessel	Heavy Lift Vessel
0	121,060,772.23	163,946,399.58	25,563,204.54	25,563,186.65	25,563,182.63	25,563,191.59	25,563,186.50	25,563,201.56	25,563,185.81	154,436,881.56	1,729,984,846.60

Service Equipment Utilization Rate

Ratio of days when the servicing equipment is in use (not delayed for a whole day due to either weather or lack of repairs to be completed) to the number of days it's present in the simulation. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.service_equipment_utilization().

Inputs:

• frequency, as explained above, options: "project" and "annual"

Example Usage:

# Project totals
style(metrics.service_equipment_utilization(frequency="project"))

	Anchor Handling Tug	Cable Laying Vessel	Crew Transfer Vessel 1	Crew Transfer Vessel 2	Crew Transfer Vessel 3	Crew Transfer Vessel 4	Crew Transfer Vessel 5	Crew Transfer Vessel 6	Crew Transfer Vessel 7	Diving Support Vessel	Heavy Lift Vessel
0	0.82	0.93	1.00	1.00	1.00	1.00	1.00	1.00	1.00	0.88	0.97

Vessel-Crew Hours at Sea

The number of vessel hours or crew hours at sea for offshore wind power plant simulations. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.vessel_crew_hours_at_sea().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_equipment

  • True: Aggregates all equipment into a single cost

  • False: Computes for each unit of servicing equipment

• vessel_crew_assumption: A dictionary of vessel names (ServiceEquipment.settings.name, but also found at Metrics.service_equipment_names) and the number of crew onboard at any given time. The application of this assumption transforms the results from vessel hours at sea to crew hours at sea.

Example Usage:

# Project total, not broken out by vessel
style(metrics.vessel_crew_hours_at_sea(frequency="project", by_equipment=False))

	Total Crew Hours at Sea
0	694,031.69

# Annual project totals, broken out by vessel
style(metrics.vessel_crew_hours_at_sea(frequency="annual", by_equipment=True))

	Total Crew Hours at Sea	Crew Transfer Vessel 1	Crew Transfer Vessel 2	Crew Transfer Vessel 3	Crew Transfer Vessel 4	Crew Transfer Vessel 5	Crew Transfer Vessel 6	Crew Transfer Vessel 7	Heavy Lift Vessel	Anchor Handling Tug	Cable Laying Vessel	Diving Support Vessel
year
2002	22,795.70	1,531.69	1,221.45	1,142.22	1,227.84	1,413.44	1,336.93	1,324.20	7,096.69	4,275.26	2,225.99	0.00
2003	31,874.37	2,618.61	2,612.89	2,552.89	2,607.44	2,545.28	2,624.83	2,552.65	7,604.90	3,187.13	2,967.76	0.00
2004	36,512.71	2,719.79	2,513.04	2,631.33	2,479.92	2,581.16	2,450.42	2,535.69	8,018.53	2,007.77	2,914.37	5,660.68
2005	35,317.55	2,238.41	2,327.96	1,980.47	2,305.55	1,963.75	2,461.90	2,260.21	7,614.77	4,504.25	5,416.18	2,244.09
2006	36,743.90	2,549.99	2,392.73	2,675.45	2,533.97	2,505.46	2,495.44	2,736.27	7,615.30	3,463.72	1,809.98	5,965.61
2007	32,020.14	2,534.35	2,459.96	2,387.57	2,470.09	2,452.93	2,484.16	2,567.19	7,209.80	2,287.74	1,761.56	3,404.79
2008	35,800.81	2,417.33	2,460.02	2,296.08	2,352.46	2,435.75	2,410.32	2,202.07	7,721.39	3,871.14	1,917.50	5,716.75
2009	33,013.69	2,512.96	2,097.45	2,266.65	1,977.63	2,159.45	2,291.56	1,997.71	8,217.46	4,036.46	2,346.68	3,109.70
2010	37,326.07	2,464.13	2,849.76	2,483.01	2,453.55	2,580.74	2,699.78	2,618.38	7,574.88	4,462.37	1,617.94	5,521.52
2011	34,792.23	2,643.07	2,747.68	2,629.31	2,657.73	2,470.69	2,720.00	2,619.47	7,837.63	4,152.67	804.87	3,509.11
2012	35,955.10	2,519.89	2,587.04	2,481.99	2,621.43	2,704.57	2,394.30	2,640.32	7,424.98	720.50	4,204.08	5,655.99
2013	34,543.49	2,275.43	2,173.69	2,146.15	2,232.57	2,269.15	2,371.72	2,465.26	7,650.51	1,837.93	3,804.06	5,317.00
2014	40,260.87	2,496.82	2,492.41	2,294.50	2,387.40	2,410.05	2,408.34	2,555.63	8,043.24	3,016.18	6,544.60	5,611.71
2015	31,169.45	2,212.07	1,973.20	2,476.91	2,143.68	1,895.20	2,174.66	2,333.59	7,946.14	318.50	2,301.80	5,393.71
2016	36,410.38	2,758.74	2,618.57	2,556.44	2,670.68	2,628.67	2,660.38	2,597.76	7,377.62	3,395.17	2,135.96	5,010.39
2017	36,977.94	2,555.77	2,589.75	2,704.37	2,650.90	2,761.98	2,621.15	2,462.29	7,959.88	2,718.48	2,228.76	5,724.62
2018	35,342.77	2,479.39	2,477.76	2,591.19	2,431.53	2,568.29	2,460.77	2,477.05	7,833.00	2,833.02	2,759.53	4,431.24
2019	38,528.54	2,368.89	2,620.20	2,582.82	2,338.66	2,876.70	2,306.96	2,462.11	8,214.70	3,911.29	3,179.40	5,666.81
2020	36,262.62	2,323.17	2,166.15	2,411.61	2,286.86	2,086.79	2,153.23	2,355.31	7,705.07	2,122.98	5,644.86	5,006.60
2021	32,383.37	2,645.37	2,616.79	2,366.27	2,570.82	2,372.97	2,563.30	2,513.36	7,662.83	1,893.50	546.71	4,631.45

Number of Tows

The number of tows performed during the simulation. If tow-to-port was not used in the simulation, a DataFrame with a single value of 0 will be returned. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.number_of_tows().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_tug

  • True: Computed for each tugboat (towing vessel)

  • False: Aggregates all the tugboats

• by_direction

  • True: Computed for each direction a tow was performed (to port or to site)

  • False: Aggregates to the total number of tows

Example Usage:

# Project Total
# NOTE: This example has no towing, so it will return 0
style(metrics.number_of_tows(frequency="project"))

	total_tows
0	0

Labor Costs

Sum of all labor costs associated with servicing equipment, excluding the labor defined in the fixed costs, which can be broken out by type. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.labor_costs().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_type

  • True: Computed for each labor type (salary and hourly)

  • False: Aggregates all the labor costs

Example Usage:

# Project total at the whole wind farm level
total = metrics.labor_costs(frequency="project", by_type=False)
print(f"Project total: ${total.values[0][0] / metrics.project_capacity:,.2f}/MW")

Project total: $42,184.63/MW

# Project totals for each type of labor
style(metrics.labor_costs(frequency="project", by_type=True))

	hourly_labor_cost	salary_labor_cost	total_labor_cost
0	0.00	50,621,557.60	50,621,557.60

Equipment and Labor Costs

Sum of all labor and servicing equipment costs, excluding the labor defined in the fixed costs, which can be broken out by each category. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.equipment_labor_cost_breakdown().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_category

  • True: Computed for each unit servicing equipment and labor category

  • False: Aggregated to the sum of all costs

reason definitions:

• Maintenance: routine maintenance, or events defined as a wombat.core.data_classes.Maintenance

• Repair: unscheduled maintenance, ranging from inspections to replacements, or events defined as a wombat.core.data_classes.Failure

• Mobilization: Cost of mobilizing servicing equipment

• Crew Transfer: Costs incurred while crew are transferring between a turbine or substation and the servicing equipment

• Site Travel: Costs incurred while transiting to/from the site and while at the site

• Weather Delay: Any delays caused by unsafe weather conditions

• No Requests: Equipment and labor is active, but there are no repairs or maintenance tasks to be completed

• Not in Shift: Any time outside the operating hours of the wind farm (or the servicing equipment's specific operating hours)

Example Usage:

# Project totals
style(metrics.equipment_labor_cost_breakdowns(frequency="project", by_category=False))

	total_cost	total_hours
reason
Maintenance	22,887,326.00	71,936.88
Repair	591,071,912.16	161,450.00
Crew Transfer	9,261,317.16	11,046.00
Site Travel	95,382,896.80	123,023.55
Mobilization	135,215,000.00	153,216.00
Weather Delay	1,416,717,578.98	606,373.44
No Requests	101,031,872.90	404,080.00
Not in Shift	27,424,892.86	105,133.24

# Project totals by each category
style(metrics.equipment_labor_cost_breakdowns(frequency="project", by_category=True))

	hourly_labor_cost	salary_labor_cost	total_labor_cost	equipment_cost	total_cost	total_hours
reason
Maintenance	0	2,189,906.24	2,189,906.24	20,697,419.76	22,887,326.00	71,936.88
Repair	0	7,611,176.19	7,611,176.19	583,460,735.97	591,071,912.16	161,450.00
Crew Transfer	0	345,509.87	345,509.87	8,915,807.29	9,261,317.16	11,046.00
Site Travel	0	3,900,115.38	3,900,115.38	91,482,781.41	95,382,896.80	123,023.55
Mobilization	0	0.00	0.00	135,215,000.00	135,215,000.00	153,216.00
Weather Delay	0	21,980,035.02	21,980,035.02	1,394,737,543.96	1,416,717,578.98	606,373.44
No Requests	0	11,576,175.06	11,576,175.06	89,455,697.84	101,031,872.90	404,080.00
Not in Shift	0	3,018,639.83	3,018,639.83	24,406,253.03	27,424,892.86	105,133.24

Emissions

Emissions (tons or other provided units) of all servicing equipment activity, except overnight waiting periods between shifts. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.emissions().

Inputs:

• emissions_factors: Dictionary of servicing equipment names and the emissions per hour of the following activities: transit, maneuvering, idle at site, and idle at port, where port is stand-in for wherever the servicing equipment might be based when not at site.

• maneuvering_factor: The proportion of transit time that can generally be associated with positioning servicing, by default 10%.

• port_engine_on_factor: The proportion of the idling at port time when the engine is running and producing emissions, by default 25%.

# Create the emissions factors, in tons per hour
emissions_factors = {
    "Crew Transfer Vessel 1": {
        "transit": 4,
        "maneuvering": 3,
        "idle at site": 0.5,
        "idle at port": 0.25,
    },
    "Field Support Vessel": {
        "transit": 6,
        "maneuvering": 4,
        "idle at site": 1,
        "idle at port": 0.5,
    },
    "Heavy Lift Vessel": {
        "transit": 12,
        "maneuvering": 7,
        "idle at site": 1,
        "idle at port": 0.5,
    },
    "Diving Support Vessel": {
        "transit": 4,
        "maneuvering": 7,
        "idle at site": 0.2,
        "idle at port": 0.2,
    },
    "Anchor Handling Vessel": {
        "transit": 4,
        "maneuvering": 3,
        "idle at site": 1,
        "idle at port": 0.25,
    },
}

# Add in CTVs 2 through 7
for i in range(2, 8):
    emissions_factors[f"Crew Transfer Vessel {i}"] = emissions_factors[f"Crew Transfer Vessel 1"]

style(metrics.emissions(emissions_factors=emissions_factors, maneuvering_factor=0.075, port_engine_on_factor=0.20))

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
Cell In[22], line 39
     36 for i in range(2, 8):
     37     emissions_factors[f"Crew Transfer Vessel {i}"] = emissions_factors[f"Crew Transfer Vessel 1"]
---> 39 style(metrics.emissions(emissions_factors=emissions_factors, maneuvering_factor=0.075, port_engine_on_factor=0.20))

File /opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/wombat/core/post_processor.py:1623, in Metrics.emissions(self, emissions_factors, maneuvering_factor, port_engine_on_factor)
   1588 """Calculates the emissions, typically in tons, per hour of operations for
   1589 transiting, maneuvering (calculated as a % of transiting), idling at the site
   1590 (repairs, crew transfer, weather delays), and idling at port (weather delays),
   (...)   1618     equipment definition in ``emissions_factors``.
   1619 """
   1620 if missing := set(self.service_equipment_names).difference(
   1621     [*emissions_factors]
   1622 ):
-> 1623     raise KeyError(
   1624         f"`emissions_factors` is missing the following keys: {missing}"
   1625     )
   1627 valid_categories = ("transit", "maneuvering", "idle at port", "idle at site")
   1628 emissions_categories = list(
   1629     chain(*[[*val] for val in emissions_factors.values()])
   1630 )

KeyError: "`emissions_factors` is missing the following keys: {'Cable Laying Vessel', 'Anchor Handling Tug'}"

Component Costs

All the costs associated with maintenance and failure events during the simulation, including delays incurred during the repair process, but excluding costs not directly tied to a repair. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.component_costs().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_category

  • True: Computed across each cost category

  • False: Aggregated to the sum of all categories

• by_action

  • True: Computed by each of "repair", "maintenance", and "delay", and is included in the MultiIndex

  • False: Aggregated as the sum of all actions

action definitions:

• maintenance: routine maintenance

• repair: unscheduled maintenance, ranging from inspections to replacements

• delay: Any delays caused by unsafe weather conditions or not being able to finish a process within a single shift

Example Usage:

# Project totals by component
style(metrics.component_costs(frequency="project", by_category=False, by_action=False))

	total_cost
component
cable	170,245,604.16
drive_train	158,057,345.37
electrical_system	1,228,522,201.85
generator	296,024,693.77
hydraulic_system	35,387,589.33
rotor_blades	64,487,044.77
supporting_structure	227,424,053.67
transformer	1,320,764.04
yaw_system	30,550,398.92

# Project totals by each category and action type
style(metrics.component_costs(frequency="project", by_category=True, by_action=True))

		materials_cost	total_labor_cost	equipment_cost	total_cost
component	action
cable	delay	0.00	2,894,339.66	99,039,818.76	101,934,158.42
	maintenance	0.00	15,342.60	525,000.00	540,342.60
	repair	0.00	1,616,355.01	55,309,164.06	56,925,519.07
drive_train	delay	0.00	806,757.43	115,243,091.38	116,049,848.81
	maintenance	0.00	0.00	0.00	0.00
	repair	0.00	217,754.42	33,113,383.33	33,331,137.75
electrical_system	delay	0.00	4,949,641.76	830,288,669.92	835,238,311.67
	maintenance	0.00	0.00	0.00	0.00
	repair	0.00	1,669,708.10	334,492,032.00	336,161,740.11
generator	delay	0.00	4,806,285.90	192,583,070.87	197,389,356.78
	maintenance	0.00	940,282.20	5,005,000.00	5,945,282.20
	repair	0.00	560,989.78	71,242,537.38	71,803,527.16
hydraulic_system	delay	0.00	3,508,928.64	19,518,231.04	23,027,159.69
	maintenance	0.00	0.00	0.00	0.00
	repair	0.00	890,117.38	5,626,000.00	6,516,117.38
rotor_blades	delay	0.00	1,370,494.43	38,723,811.03	40,094,305.46
	maintenance	0.00	0.00	0.00	0.00
	repair	0.00	366,907.32	16,658,625.00	17,025,532.32
supporting_structure	delay	0.00	6,197,427.62	100,695,063.84	106,892,491.46
	maintenance	0.00	1,209,294.92	15,034,419.76	16,243,714.68
	repair	0.00	2,200,804.60	60,071,077.52	62,271,882.12
transformer	delay	0.00	148,369.83	789,753.32	938,123.15
	maintenance	0.00	24,986.52	133,000.00	157,986.52
	repair	0.00	1,643.85	8,750.00	10,393.85
yaw_system	delay	0.00	314,750.58	22,253,349.75	22,568,100.34
	maintenance	0.00	0.00	0.00	0.00
	repair	0.00	86,895.74	6,939,166.67	7,026,062.40

Fixed Cost Impacts

Computes the total costs of the fixed costs categories. For further documentation, see the definition docs, here: wombat.core.data_classes.FixedCosts, or the API docs here: wombat.core.post_processor.Metrics.fixed_costs().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• resolution (also, demonstrated below)

  • "high": Computed across the most granular cost levels

  • "medium": Computed for each general cost category

  • "low": Aggregated to a single sum of costs

pprint(metrics.fixed_costs.hierarchy)

{'operations': {'annual_leases_fees': ['submerge_land_lease_costs',
                                       'transmission_charges_rights'],
                'environmental_health_safety_monitoring': [],
                'insurance': ['brokers_fee',
                              'operations_all_risk',
                              'business_interruption',
                              'third_party_liability',
                              'storm_coverage'],
                'labor': [],
                'onshore_electrical_maintenance': [],
                'operating_facilities': [],
                'operations_management_administration': ['project_management_administration',
                                                         'marine_management',
                                                         'weather_forecasting',
                                                         'condition_monitoring']}}

Example Usage:

# Project totals at the highest level
# NOTE: there were no fixed costs defined in this example, so all values will be 0, so
#       this will just be demonstrating the output format
style(metrics.project_fixed_costs(frequency="project", resolution="low"))

	operations
0	0.00

# Project totals at the medium level
style(metrics.project_fixed_costs(frequency="project", resolution="medium"))

	operations_management_administration	insurance	annual_leases_fees	operating_facilities	environmental_health_safety_monitoring	onshore_electrical_maintenance	labor
0	0.00	0.00	0.00	0.00	0.00	0.00	0.00

# Project totals at the lowest level
style(metrics.project_fixed_costs(frequency="project", resolution="high"))

	project_management_administration	marine_management	weather_forecasting	condition_monitoring	brokers_fee	operations_all_risk	business_interruption	third_party_liability	storm_coverage	submerge_land_lease_costs	transmission_charges_rights	operating_facilities	environmental_health_safety_monitoring	onshore_electrical_maintenance	labor
0	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00

OpEx

Computes the total cost of all operating expenditures for the duration of the simulation, including fixed costs. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.opex().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by_category

  • True shows the port fees, fixed costs, labor costs, equipment costs, and materials costs in addition the total OpEx

  • False shows only the total OpEx

Example Usage:

style(metrics.opex("annual"))

/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/wombat/core/post_processor.py:2054: FutureWarning: Downcasting object dtype arrays on .fillna, .ffill, .bfill is deprecated and will change in a future version. Call result.infer_objects(copy=False) instead. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`
  port_fees = port_fees.fillna(0)

	OpEx
year
2002	107,917,667.44
2003	123,544,176.00
2004	124,180,897.63
2005	124,950,100.77
2006	121,036,037.24
2007	116,565,084.55
2008	134,333,703.47
2009	115,935,808.65
2010	120,769,421.49
2011	124,812,779.71
2012	125,104,263.58
2013	121,438,670.09
2014	127,326,108.21
2015	110,266,932.82
2016	121,510,157.13
2017	122,312,718.50
2018	123,772,359.50
2019	133,753,199.51
2020	130,666,488.89
2021	118,080,609.68

style(metrics.opex("annual", by_category=True))

/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/wombat/core/post_processor.py:2054: FutureWarning: Downcasting object dtype arrays on .fillna, .ffill, .bfill is deprecated and will change in a future version. Call result.infer_objects(copy=False) instead. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`
  port_fees = port_fees.fillna(0)

	operations	port_fees	equipment_cost	total_labor_cost	materials_cost	OpEx
year
2002	0.00	0	103,562,314.55	2,373,832.89	1,981,520	107,917,667.44
2003	0.00	0	117,981,837.88	2,441,438.13	3,120,900	123,544,176.00
2004	0.00	0	119,516,309.20	2,528,488.43	2,136,100	124,180,897.63
2005	0.00	0	119,288,473.47	2,628,287.30	3,033,340	124,950,100.77
2006	0.00	0	115,728,718.59	2,542,438.64	2,764,880	121,036,037.24
2007	0.00	0	112,792,695.39	2,342,809.16	1,429,580	116,565,084.55
2008	0.00	0	128,197,105.26	2,647,498.21	3,489,100	134,333,703.47
2009	0.00	0	111,110,067.49	2,452,969.16	2,372,772	115,935,808.65
2010	0.00	0	116,617,914.10	2,601,267.39	1,550,240	120,769,421.49
2011	0.00	0	118,875,732.78	2,426,914.93	3,510,132	124,812,779.71
2012	0.00	0	120,959,427.90	2,599,735.67	1,545,100	125,104,263.58
2013	0.00	0	116,189,951.01	2,574,335.08	2,674,384	121,438,670.09
2014	0.00	0	121,820,327.80	2,737,976.42	2,767,804	127,326,108.21
2015	0.00	0	106,022,343.35	2,359,137.48	1,885,452	110,266,932.82
2016	0.00	0	116,729,503.80	2,532,553.33	2,248,100	121,510,157.13
2017	0.00	0	117,515,904.18	2,546,234.32	2,250,580	122,312,718.50
2018	0.00	0	118,495,007.28	2,534,332.21	2,743,020	123,772,359.50
2019	0.00	0	128,520,495.17	2,632,804.34	2,599,900	133,753,199.51
2020	0.00	0	125,058,109.77	2,774,275.12	2,834,104	130,666,488.89
2021	0.00	0	113,389,000.30	2,344,229.38	2,347,380	118,080,609.68

Process Times

Computes the total number of hours spent from repair request submission to completion, performing repairs, and the number of request for each repair category. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.process_times().

Example Usage:

style(metrics.process_times())

	time_to_completion	process_time	downtime	time_to_start	N
category
anchor replacement	11,026.75	5,817.30	5,815.87	5,334.25	12
annual turbine inspection	1,305,436.90	169,069.67	167,668.92	1,147,176.33	1,433
array cable major repair	40,359.28	21,784.77	0.00	19,799.29	39
array cable replacement	38,796.27	27,364.11	0.00	12,985.55	26
blades major repair	200,057.12	714.60	709.88	199,380.87	13
blades major replacement	4,805.69	3,256.28	3,256.28	1,637.44	1
blades minor repair	127,606.80	66,678.13	65,953.87	70,494.58	629
buoyancy module replacement	17,754.30	12,947.03	12,917.11	5,861.69	54
direct drive generator major repair	577,909.14	5,716.19	5,702.78	571,585.90	52
direct drive generator major replacement	35,636.99	14,808.78	14,805.23	21,143.54	15
direct drive generator minor repair	148,952.38	62,107.58	61,318.09	101,916.10	836
export cable major repair	796.04	60.96	0.00	735.79	1
export cable subsea inspection	7,237.95	1,760.95	0.00	6,082.43	19
main shaft major repair	186,088.68	2,624.51	2,619.40	183,521.78	30
main shaft minor repair	52,962.06	20,233.51	19,916.48	39,574.76	345
main shaft replacement	21,585.01	9,824.60	9,820.63	12,178.72	13
major anchor repair	9,853.23	3,978.52	3,977.96	5,964.48	12
major pitch system repair	72,373.70	55,133.30	54,952.91	22,638.25	272
major pitch system replacement	4,940.98	146.50	146.50	4,794.73	1
marine growth removal	49,976.33	35,562.66	35,438.25	17,191.65	175
minor ballast pump repair	2,240.71	756.26	739.87	1,914.85	13
minor pitch system repair	251,752.39	131,478.30	130,335.95	139,471.35	1,219
mooring line major repair	14,507.98	5,960.58	5,959.69	8,554.99	18
mooring line replacement	29,749.57	16,178.48	16,176.79	13,854.67	30
oss annual inspection	13,511.73	7,051.27	7,022.64	7,500.34	38
oss minor repair	748.24	405.62	403.12	544.56	5
power converter major repair	5,187,244.39	31,973.82	31,856.12	4,476,849.45	525
power converter minor repair	133,551.48	64,355.89	63,653.85	83,818.70	728
power converter replacement	180,767.32	63,549.83	63,491.20	120,348.76	105
power electrical system major repair	298,909.30	2,143.38	2,136.87	249,564.69	21
power electrical system major replacement	3,312.53	1,474.90	1,471.28	1,904.33	8
power electrical system minor repair	83,399.04	32,250.25	31,860.80	58,930.64	498
structural annual inspection	1,265,463.44	160,504.37	159,112.42	1,109,056.01	1,395
structural subsea inspection	579,720.57	25,959.72	25,447.66	557,239.01	663
yaw system major repair	132,142.11	1,373.90	1,370.08	130,870.61	14
yaw system major replacement	1,443.59	950.18	950.18	538.34	1
yaw system minor repair	32,232.22	12,735.12	12,478.97	23,514.27	226

Power Production

Computes the total power production for the wind farm. For further documentation, see the API docs here: wombat.core.post_processor.Metrics.power_production().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• by, as explained above options: "windfarm" and "turbine"

• units

  • "kwh": kilowatt-hours (kWh)

  • "mwh": megawatt-hours (MWh)

  • "gwh": gigawatt-hours (GWh)

Example Usage:

# Project totals, in kWh, at the wind farm level
style(metrics.power_production(frequency="project", by="windfarm", units="kwh"))

	windfarm
Project Energy Production (kWh)	106,056,118,088.50

# Project totals, in MWh, at the wind farm level
style(metrics.power_production(frequency="project", units="mwh"))

	windfarm
Project Energy Production (MWh)	106,056,118.09

# Project totals, in GWh, at the wind farm level
style(metrics.power_production(frequency="project"))

	windfarm
Project Energy Production (GWh)	106,056.12

Net Present Value

Calcualtes the net present value (NPV) for the project, as \(NPV = (Power * OfftakePrice - OpEx) / (1 + DiscountRate)\).

For further documentation, see the API docs here: wombat.core.post_processor.Metrics.npv().

Inputs:

• frequency, as explained above, options: "project", "annual", "monthly", and "month-year"

• discount_rate: The rate of return that could be earned on alternative investments, by default 0.025.

• offtake_price: Price of energy, per MWh, by default 80.

style(metrics.opex("annual"))

/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/wombat/core/post_processor.py:2054: FutureWarning: Downcasting object dtype arrays on .fillna, .ffill, .bfill is deprecated and will change in a future version. Call result.infer_objects(copy=False) instead. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`
  port_fees = port_fees.fillna(0)

	OpEx
year
2002	107,917,667.44
2003	123,544,176.00
2004	124,180,897.63
2005	124,950,100.77
2006	121,036,037.24
2007	116,565,084.55
2008	134,333,703.47
2009	115,935,808.65
2010	120,769,421.49
2011	124,812,779.71
2012	125,104,263.58
2013	121,438,670.09
2014	127,326,108.21
2015	110,266,932.82
2016	121,510,157.13
2017	122,312,718.50
2018	123,772,359.50
2019	133,753,199.51
2020	130,666,488.89
2021	118,080,609.68