qalx-orcaflex-examples

A set of tutorial-style examples to get started with qalx-orcaflex

Overview

This repository contains examples of a number of different approaches to solving a very simple batch of load cases in OrcaFlex.

All the cases are based on the Orcina provided example A05 Lazy wave with FPSO.

For the purpose of this simplified example we will consider the waves and current coming from 8 directions and two offset positions; near and far.

Each example shows the different features of qalx-OrcaFlex, and the different approaches that can be taken to building batches.

Prerequisites

These examples assume you already have a basic understanding of object-oriented Python and are familiar with using pip , Python packages and virtual environments.

In addition, in order to run through the examples, you will need to have already:

• Installed python version >=3.8, <4.0.
• Obtained a qalx token from AgileTek Engineering.
• Installed and configured qalx as per the instructions here.

Installing package dependencies

Create a virtual environment with:

python -m venv venv

This will keep the dependencies for this repository isolated from your global Python installed packages.

Activate the venv and then install the requirements for this package using:

pip install -r requirements.txt

Starting the bots

The examples require you to start the bots by building the qalx factory defined in ofx_factory.plan.

• In a terminal/cmd window navigate to the root of this repository.
• Start the factory build process (see here for more information) with the following command. Use stage-1 for examples 1 and 2; stage-2 for example 3:

qalx factory-build --plan ofx_factory.plan --stage [stage-1, stage-2]

• At the end of the build process, if successful, you should see something like:

[2022-10-24 11:34:43] stage-1 - [local_sector]:    stack creation completed 
[2022-10-24 11:34:48] pyqalx:    Factory Build Successful 

• Additionally, you can verify that a bot has been deployed correctly with:

qalx bot-info [BatchBot, SimulationBot, RiserReportBot]

Example 1

Description

This example builds and executes an OrcaFlex batch from a set of pre-built .dat files, and extracts specified results from the simulation output.

Running

Start the factory as described above. Once the factory is running, the script is run by navigating to the example_1 directory and supplying the following arguments at the command line:

• Batch name
• Directory containing .dat files

For example, with a batch name of test_batch and the .dat files in the data directory as they are now:

python example.py test_batch data

This will build the batch and submit all load cases to the queue. Note that if you supply a relative path for the directory of .dat files, the script assumes that it is relative to the parent example_1 directory, however you can also provide an absolute path.

A GUI will open to allow you to keep track of the progress by either clicking the Get Update button or setting the auto-update period.

Once the batch is complete, a summary of results will be printed to the command line output.

The effective tension and curvature are pre-specified as required results.

Example 2

Description

This example dynamically generates a batch of load cases from one base .dat file by varying the direction and offset, and extracts a set of model views from the simulation output.

Running

Start the factory as described above. Once the factory is running, the script is run by navigating to the example_2 directory and supplying the following arguments at the command line:

• Batch name
• Path of base .dat file

For example, with a batch name of test_batch and the A05 Lazy wave with FPSO.dat file in the example_2 directory:

python example.py test_batch "A05 Lazy wave with FPSO.dat"

This will build the batch and submit all load cases to the queue. Note that if you supply a relative path for the base .dat file, the script assumes that it is relative to the parent example_2 directory, however you can also provide an absolute path.

A GUI will open to allow you to keep track of the progress by either clicking the Get Update button or setting the auto-update period.

Once the batch is complete, a folder containing the extracted model view images will open - note that it can take up to 1 min to summarise the results and generate the model view images after the load cases have appeared to finish processing in the GUI.

The effective tension and curvature are pre-specified as required results.

Example 3

Description

This example extends example 2, by deploying another custom bot that uses the simulation output to automatically generate a .docx report.

Running

Start the factory as described above - remember to use stage-2. Once the factory is running, the script is run by navigating to the example_3 directory and supplying the following arguments at the command line:

• Batch name
• Path of base .dat file

For example, with a batch name of test_batch and the A05 Lazy wave with FPSO.dat file in the example_3 directory:

python example.py test_batch "A05 Lazy wave with FPSO.dat"

This will build the batch and submit all load cases to the queue. Note that if you supply a relative path for the base .dat file, the script assumes that it is relative to the parent example_3 directory, however you can also provide an absolute path.

A GUI will open to allow you to keep track of the progress by either clicking the Get Update button or setting the auto-update period.

Once the batch is complete, a folder containing the extracted model view images will open - note that it can take up to 1 min to summarise the results and generate the model view images after the load cases have appeared to finish processing in the GUI.

The effective tension and curvature are pre-specified as required results.

Within a couple of minutes of the batch simulation completing, you should receive an email notification with a link to download the report generated by the RiserReportBot.