OpenNeoMC:an Open-source Tool for Particle Transport Optimization that Combining OpenMC with NEORL

Related tags

NetworkingOpenNeoMC
Overview

OpenNeoMC:an Open-source Tool for Particle Transport Optimization that Combining OpenMC with NEORL

OpenMC is a community-developed Monte Carlo neutron and photon transport simulation code for particle transport. OpenMC was originally developed by members of the Computational Reactor Physics Group at the Massachusetts Institute of Technology starting in 2011.

NEORL (NeuroEvolution Optimization with Reinforcement Learning) is a set of implementations of hybrid algorithms combining neural networks and evolutionary computation based on a wide range of machine learning and evolutionary intelligence architectures. NEORL aims to solve large-scale optimization problems relevant to operation & optimization research, engineering, business, and other disciplines. NEORL was established in MIT back in 2020 with feedback, validation, and usage of different colleagues.

In OpenNeoMC, we combine these two open-source tools to empower particle transport with state-of-the-art optimization techniques. We firstly provide users with easy ways to install the framework that combines NEORL with OpenMC, and a simple example is available to test the framework. Then we offer two practical engineering optimization applications in nuclear physics. More applications that involve both optimization and nuclear physics will be added in the future. We highly welcome users and researchers in the nuclear area to contribute OpenNeoMC and solve engineering problems in this framework.

Installing OpenNeoMC

Installation on Linux/Mac with conda

Install Conda

Please install conda before proceeding, it will bring you convenience to install anaconda directly, which includes conda and other necessary python packages.

Install OpenMC

conda config --add channels conda-forge
conda search openmc

Create a new virtual environment named openneomc

conda create -n openneomc openmc

Test OpenMC

Follow with the official examples to test the OpenMC

Cross Section Configuration

You may encounter the no cross_sections.xml error when running OpenMC. This is caused by the missing of nuclear data, you could solve it refer to Cross Section Configuration

Download cross section data

Various cross section data are available on the OpenMC official website, from the OpenMC team, LANL, etc. In OpenNeoMC, we use ENDF/B-VII.1 in default. But if you have specific purpose, you can use other data that you need.

After downloading the cross-section data file, configure it as an environmental variable as follows.

Add environmental variables

## Temporary methods
# in python
import os
os.environ['OPENMC_CROSS_SECTIONS'] = '/PATH/cross_sections.xml'
# in shell
export OPENMC_CROSS_SECTIONS=../cross_sections.xml

## Once for all: you can modify the ~/.bashrc to configure environmental variables
# open ~/.bashrc
vim ~/.bashrc
# add the following command in the end 
export OPENMC_CROSS_SECTIONS=/PATH/cross_sections.xml
# update 
source ~/.bashrc

Install NEORL

Install python 3.7 to make sure the stable run of tensorflow-1.14.0

conda install python=3.7 
pip install neorl==1.6

Check the version of sciki-learn, if it is 1.x, downgrade the scikit-learn version to 0.24.2

# check version
python -c 'import sklearn; print(sklearn.__version__)'

# downgrade the sklearn version if necessary
pip install scikit-learn==0.24.2

Check if you have install NEORL successfully by unit test.

neorl

If you see the 'NEORL' logo, then you have prepared the OpenNeoMC framework, congratulations!

Test OpenNeoMC

Let's test OpenNeoMC by the 'pin_cell_test.py' example.

Remember to configure environmental variables as above!

# run 
python pin_cell_test.py

If you see the 'NEORL' logo and the log information of OpenMC, then congratulations!

Installing OpenNeoMC with Docker on Linux/Mac/Windows

Installing OpenNeoMC with docker is highly recommended! In this way, you need not worry about issues like cross-section data and software compatibility, etc. All you need to do are simply pull the image and run it in your own machine with any OS.

Install Docker

Follow the official tutorial to Install docker on your machine: get docker

Install OpenNeoMC

After installing docker, your can easily install use OpenNeoMC framework within only four steps:

# Pull docker images from dock hub  
sudo docker pull 489368492/openneomc

# Check the openmc docker images
sudo docker images

# Run the openmc images to create container named `openneomc`
sudo docker run -tid --shm-size=8G --gpus all --name openneomc -v /LocalWorkingDir/:/workspace/ 489368492/openneomc

# Execute the container
sudo docker exec -it openneomc /bin/bash

Note: in docker run step, the -v flag mounts the current working directory into the container, which is very convenient for users.

Please refer to Docker CLI for docker command-line descriptions.

Other commonly used commands

# Exit the container
exit

# Stop the container
sudo docker stop openneomc

# Start the container
sudo docker start openneomc

# Delete the container
sudo docker rm openneomc

# Delete the image(remove the container first)
sudo docker image rm 489368492/openneomc

Test OpenNeoMC

Let's test OpenNeoMC by the 'pin_cell_test.py' example, which can be found at /home

# cd /home
cd /home

# run 
python pin_cell_test.py

If you see the 'NEORL' logo and the log information of OpenMC, then congratulations!

The program runs around 3 minutes(may vary depending on your CPU), and the results are like:

------------------------ JAYA Summary --------------------------
Best fitness (y) found: 0.0015497217274231812
Best individual (x) found: [2.01355604]
--------------------------------------------------------------
---JAYA Results---
x: [2.01355604]
y: 0.0015497217274231812
JAYA History:
 [0.018311916874464318, 0.0017114252626817539, 0.0017114252626817539, 0.0017114252626817539, 0.0015497217274231812]
running time:
 155.2281835079193

Reference

OpenMC: https://docs.openmc.org/en/stable

OpenMC image: https://hub.docker.com/r/openmc/openmc

NEORL: https://neorl.readthedocs.io/en/latest/

OpenNeoMC image: https://hub.docker.com/r/489368492/openneomc

Contact

If you have any suggestions or issues, please feel free to contact Xubo Gu([email protected])

NetworkX is a Python package for the creation, manipulation, and study of the structure, dynamics, and functions of complex networks.

NetworkX is a Python package for the creation, manipulation, and study of the structure, dynamics, and functions of complex networks.

NetworkX 12k Jan 02, 2023
Visualize the electric field of a point charge network.

ElectriPy ⚡ Visualize the electric field of a point charges network. 🔌 Installation Install ElectriPy package: $ pip install electripy You are all d

Dylan Tintenfich 29 Aug 29, 2022
This is a Client-Server-System which can share the screen from the server to client and in the other direction.

Screenshare-Streaming-Python This is a Client-Server-System which can share the screen from the server to client and in the other direction. You have

VFX / Videoeffects Creator 1 Nov 19, 2021
Tool that creates a complete copy of your server

Discord-Server-Cloner Tool that creates a complete copy of your server Setup: Open run.bat If the file closes, open cmd And write: pip install -r requ

DEEM 3 Dec 13, 2021
Python implementation of the Session open group server

API Documentation CLI Reference Want to build from source? See BUILDING.md. Want to deploy using Docker? See DOCKER.md. Installation Instructions Vide

Oxen 36 Jan 02, 2023
ServerStatus with node management and monitor

ServerStatus with node management and monitor

lidalao 162 Jan 01, 2023
A simple python script to send cute messages to my boyfriend.

Morning Messages A simple python script to send cute messages to my boyfriend. It gives him the weather and news currently. Installation git clone htt

Sabrina Medwinter 3 Oct 12, 2022
Tool to get the top 100 of the fastest nodes in the Tor network. Based on Kirzahk tool.

Tor Network Top 100 IPs Tool to get the top 100 of the fastest nodes in the Tor network. Based on Kirzahk tool. Just execute top100ipstor.py to get th

Juan Manuel 0 Jan 23, 2022
This is simple script that changes the config register of a cisco router over serial so that you can reset the password

Cisco-router-config-bypass-tool- This is simple script that changes the config register of a cisco router over serial so that you can bypass the confi

James 1 Jan 02, 2022
Multi-vendor library to simplify CLI connections to network devices

Netmiko Multi-vendor library to simplify CLI connections to network devices Why Netmiko? Network automation to screen-scraping devices is primarily co

Kirk Byers 3k Jan 01, 2023
Extended refactoring capabilities for Python LSP Server using Rope.

pylsp-rope Extended refactoring capabilities for Python LSP Server using Rope. This is a plugin for Python LSP Server, so you also need to have it ins

36 Dec 24, 2022
CSP-style concurrency for Python

aiochan Aiochan is a library written to bring the wonderful idiom of CSP-style concurrency to python. The implementation is based on the battle-tested

Ziyang Hu 127 Dec 23, 2022
ARTEMIS: Real-Time Detection and Automatic Mitigation for BGP Prefix Hijacking.

ARTEMIS: Real-Time Detection and Automatic Mitigation for BGP Prefix Hijacking. This is the main ARTEMIS repository that composes artemis-frontend, artemis-backend, artemis-monitor and other needed c

INSPIRE Group @FORTH-ICS 273 Jan 01, 2023
AdaFruit Funhouse publishing Temperature, Humidity and Pressure to MQTT / Apache Pulsar

pulsar-adafruit-funhouse AdaFruit Funhouse publishing Temperature, Humidity and Pressure to MQTT / Apache Pulsar Device Get your own from adafruit Ada

Timothy Spann 1 Dec 30, 2021
The module that allows the collection of data sampling, which is transmitted with WebSocket via WIFI or serial port for CSV file.

The module that allows the collection of data sampling, which is transmitted with WebSocket via WIFI or serial port for CSV file.

Nelson Wenner 2 Apr 01, 2022
Quickly fetch your WiFi password and if needed, generate a QR code of your WiFi to allow phones to easily connect

wifi-password Quickly fetch your WiFi password and if needed, generate a QR code of your WiFi to allow phones to easily connect. Works on macOS and Li

Siddharth Dushantha 2.6k Jan 05, 2023
Ultimate transformation library that supports validation, contexts and aiohttp.

Trafaret Ultimate transformation library that supports validation, contexts and aiohttp. Trafaret is rigid and powerful lib to work with foreign data,

Mikhail Krivushin 174 Nov 27, 2022
Port Traffic/Bandwidth Monitor Script

python-switch-port-traffic-alarm Port Traffic/Bandwidth Monitor Script That's an Switch Port Traffic monitor program is checking the switch uplink por

goksinenki 4 Sep 02, 2021
Readable, simple and fast asynchronous non-blocking network apps

Fast and readable async non-blocking network apps Netius is a Python network library that can be used for the rapid creation of asynchronous non-block

Hive Solutions 120 Nov 20, 2022
This is a zeep based SOAP client wrapper for simple communication with the Bricknode SOAP API.

This is a zeep based SOAP client wrapper for simple communication with the Bricknode SOAP API.

Nord Fondkommission AB 2 Dec 15, 2021