Using ACCERT (With Python)

Using ACCERT!

To begin using ACCERT, there are terms and applications to familiarize yourself with.

• Code of Account (COA)

  • Every component of a nuclear reactor has its own unique COA.

    • The codes have different varying levels, even some up to 5.

    • Most codes are between levels 0-3.

    • The levels just break down each cost element by different subtasks within them.

  • The COA has a cost aggregation that consists of three sub-categories at its most basic level.

    • Factory equipment, which is usually described in dollars, or could be automatically converted.

    • Labor costs.

    • Material costs, which have specific quantities. (square feet, tons, cubic yards, linear feet, etc.)

• Workbench

  • ACCERT uses the workbench interface to display a table of costs.

• Databases

  • ACCERT, as mentioned before, comprises of three major components, one of them being the relational database.

  • The relational database compiles all the different costs of various lower-level costs, and displays them based on their level.

  • The databases are based off of cost reports of two model reactors as of June 2023:

    • The Pressurized Water Reactor (PWR12-BE)

    • The Advanced Burner reactor (ABR-1000)

  • ACCERT bases all of the algorithms off of previously estimated reactors, and uses something called an elevation element to account for inflation throughout the years.

Testing Python

• To begin, open up a command terminal, and type the following into it.

$ cd ACCERT
$ python -m pip install --upgrade pip

• If python is installed, it will uninstall pip and reinstall it.

• Install pip requirements

$ pip install -r requirements.txt

• A lot of code pops up, and at the end it should say a lot of different things were successfully installed.

• Next in the command terminal, type the following:

$ cd test
$ pytest

• This part might take a minute or two, and at the end there will be a few different percentages, hopefully with the code passing the tests.

Using Python to Run ACCERT

• Begin by opening your Python text editor of choice. (Visual Studio Code is a good option.)

  • Open the file of the reference reactor you would like to compare your cost estimate to.

  • There are two required parameters that need to be changed based on the theoretical reactor you are creating.

    • The thermal and electrical power produced by the theoretical reactor should be put into the parameters listed in the first few lines of code.

    • There are also several variables or parameters that the user can change to fit their theoretical reactor design.

    • For example, one of the radii of the theoretical design can be written in the code as var(Cont_rad_out_m){value = xx unit = m}.

    • The value would be the variable that the user can change, and would alter every cost estimate that the COA has ties to.

• After making all the adjustments to the desired variables, move into the tutorial folder in the command terminal and run the python code.

$ cd ..
$ cd tutorial
$ python ../src/Main.py -i myinput.son
$ code output.out

NOTICE!

Replace ../src/Main.py with the path to your Main.py file, and the myinput.son file with the desired reactor .son file.

• This is used to collect various inputs from any given database for a nuclear reactor.

  • For example:

$ python ACCERT/src/Main.py -i PWR12-BE.son

• After running the code, open the output.out file in a python coding application.

  • This will display a list of variables that the user altered, and the cost elements associated with it.

  • With this, the cost aggregation algorithm will start to work on summing up the costs with respect to the changed parameters.

  • ACCERT will then sum up all the costs and display it into one easy to read graph that displays the COA, account description, costs, unit, level, and the review status.

    • The review status is just whether or not the COA or any of it’s sub-levels have been changed.

NOTICE!

An output of ACCERT can be found in the “Output of ACCERT” tab.