Congjian Wang

About Congjian Wang

I currently work at Idaho National Laboratory, where I lead the development of RAVEN—a versatile framework for probabilistic risk analysis, validation and uncertainty quantification, parameter optimization, model reduction, and data knowledge discovery. I possess a robust background in applied mathematics, statistics, reactor physics, safety, risk and reliability analysis, and high-performance computing. My expertise includes code development, model validation and calibration, system optimization, reduced-order modeling, machine learning, artificial intelligence, and risk-informed system analysis. Additionally, I am skilled in project and team management and proficient in programming languages such as Python and C++, as well as various scientific and engineering software packages, including MOOSE, SCALE, SIMULATE, RELAP5-3D, RAVEN, and SAPHIRE.

Experience

1. Senior Computational Nuclear Scientist

   Idaho National Laboratory

   Aug 2022 — Present · Idaho Falls, ID, US

   Technical Group Lead for Risk Analysis Virtual Environment (RAVEN) framework; recipient of the 2023 R&D 100 Award.• Led the development of advanced machine learning and artificial intelligence capabilities, including genetic algorithms, deep reinforcement learning, Bayesian optimization, anomaly detection, causal inference, natural language processing, semantic and similarity analysis, knowledge graphs, and foundation models.• Project Lead for the LWRS-RISA pathway \"Digital I&C Risk Assessment\" project since August 2024. Developed an integrated framework for risk assessment of safety-related digital instrumentation and control systems.• Led the development of the Digital Analytics, Causal Knowledge Acquisition and Reasoning (DACKAR) tool to analyze equipment reliability data, providing system engineers with insights into anomalous behaviors, degradation trends, and their possible causes and consequences.• Led the development of the Bayesian Model Calibration (BayCal) tool to inversely quantify uncertainties associated with simulation model parameters based on experimental data.• Led the development of the Platform of Optimal Experiment Management (POEM) tool to automatically guide the design of experiments using automated machine learning capabilities.• Developed genetic algorithms for power plant fuel reload optimization and interfaces for the PARCS and SIMULATE reactor core simulation codes for RAVEN.• Designed and optimized a modular hydrogen-based integrated energy system to maximize revenue using deep reinforcement learning techniques.• Integrated condition-based, diagnostic, prognostic, and anomaly detection into reliability models to support predictive maintenance.• Developed tools and methods to analyze plant outage schedules and assist schedulers in improving outage resilience.• Developed model-based approaches to extract knowledge from plant equipment reliability data and created a graph-based reliability approach to assess system health.

Education

• Tsinghua University

  Bachelor, Engineering Physics

  2005 — 2009

• North Carolina State University

  Ph. D, Nuclear Engineering