(fatigue-extra-04)=

# Fatigue Analysis

The fatigue extra is a feature that allows the user to analyze cycle-count
metrics.

## Installation

The extra is compatible with Python versions from 3.8 to 3.10, and installs the
[py-fatigue](https://owi-lab.github.io/py_fatigue/) and
[swifter](https://github.com/jmcarpenter2/swifter) packages.

To install the extra, run the following command in your terminal:

::::{tab-set-code}

```powershell
pip install api-24sea[fatigue]
```

```bash
pip install 'api-24sea[fatigue]'
```

::::

## Usage

Suppose you have already defined a {class}`pandas.DataFrame` and authenticated
with the API. For instructions on how to do this, refer to the
{ref}`getting started tutorial <quick-start-01>`.

### Importing the fatigue extra

```python
# %%
# - Local imports
from api_24sea.datasignals import fatigue
```

### Checking the available metrics after authentication

If your *Metrics Overview* table shows metrics whose name starts with `CC_`,
then the fatigue extra will be available for use.

### Getting data

Similar to the data retrieval process described in the
{ref}`getting started tutorial <quick-start-01>`, the data retrieval process
for the fatigue extra follows the same steps. The only difference is that the
selected metrics must include cycle-counts (CC).

```python
# %%
# **Data Retrieval**
# Besides SCADA data, we will query cycle-count metrics, which are available
# by looking for "CC" (cycle-count) and ["Mtn", "Mtl"] (i.e. Normal and
# Lateral Bending moment).

sites = ["wf"]
locations = ["a01", "a02"]
metrics = ["mean WinDSpEed", "mean pitch", "mean-Yaw", "mean power",
           "cc mtn", "cc mtl"]

start_timestamp = "2020-03-01T00:00:00Z"
end_timestamp = "2020-06-01T00:00:00Z"

df.datasignals.get_data(sites, locations, metrics,
                        start_timestamp, end_timestamp, as_dict=False)
```

### Analyzing cycle-count metrics

Converting the cycle-count JSON objects to {class}`py_fatigue.CycleCount`
objects is the first step in the fatigue analysis. This is done by calling the
{func}`api_24sea.datasignals.fatigue.Fatigue.cycle_counts_to_objects` method.

```python
# %%
# **Fatigue Analysis**
# The fatigue analysis is done by calling the cycle_counts_to_objects() method
# from the fatigue accessor.
try:
    df.fatigue.cycle_counts_to_objects()
except ImportError as ie:
    print(f"\033[31;1mImportError\033[22m: {ie}")
```

At this point, you can treat your {class}`py_fatigue.CycleCount` objects as
you would normally do with
[py-fatigue](https://owi-lab.github.io/py_fatigue/).

For more information, check py-fatigue's
[beginner's guide](https://owi-lab.github.io/py_fatigue/user/01-absolute-noob.html).