Platform of Optimal Experiment Management (POEM)

Platform of Optimal Experiment Management (POEM)

POEM is an open-source platform that revolutionizes experimental design by integrating automated machine learning. It guides experiment design, identifies optimal solutions quickly, and minimizes human intervention, making it invaluable for researchers and scientists.

Advancements in nuclear energy technology are essential for addressing global warming, enhancing remote access to electricity, and enabling space exploration. However, these advancements face challenges such as improving reactor economics, developing novel materials, and ensuring reactor safety. POEM was created to accelerate the discovery and qualification of new nuclear materials by optimizing experimental designs using machine learning.

POEM supports a range of model explorations and experiment designs:

• Random Model Explorations: Facilitates diverse experiment designs.
• Sparse Grid Model Explorations: Uses Gaussian Polynomial Chaos surrogate models to speed up experiment design.
• Time-Dependent Sensitivity and Uncertainty Analysis: Identifies critical features for experiment design.
• Model Calibrations: Integrates experiments via Bayesian inference to enhance model performance.
• Bayesian Optimization: Guides the optimal design of experiments.

By leveraging the RAVEN platform, POEM offers extensive scalability, reduced computational costs, and access to advanced sampling and machine learning capabilities.

Advantages:

• Automated Experiment Design: Reduces the need for manual intervention in designing experiments.
• Accelerated Discovery: Speeds up the identification of optimal solutions using machine learning.
• Comprehensive Analysis: Performs sensitivity and uncertainty analysis to highlight essential features.
• Scalable and Cost-Efficient: Utilizes RAVEN for large-scale applications and reduces computational costs.
• Improved Research Output: Enhances the efficiency and technological output of research activities.

Applications:

• Nuclear Energy Research: Accelerates the discovery of new materials and improves reactor designs.
• Materials Science: Optimizes experiments for developing novel materials with superior properties.
• Space Exploration: Enhances the design and qualification of materials for space missions.
• General Scientific Research: Simplifies and improves the design and analysis of experiments across various fields.

This software is open source and available at no cost. Download now by visiting the product's GitHub page.