Dr. Alexander Heifetz
Principal Engineer
Nuclear Science and Engineering Division
Argonne National Laboratory

Abstract

High-temperature fluid advanced reactors (ARs) under development, such as sodium fast reactors (SFR) and molten salt cooled reactors (MSCR) are expected to offer lower cost of energy compared to the aging fleet of light water reactors (LWRs). While ARs are in the development stage, it is important to integrate advanced sensing and instrumentation features into their design to minimize the operation and maintenance (O&M) costs. The objective of our work is to develop machine learning (ML) algorithms for thermal hydraulic system sensors to enable autonomous and uninterrupted operation of ARs. In one project, we have developed a deep learning long short-term memory (LSTM) autoencoder for continuous monitoring of a cold trap and insipient anomaly detection. Transient data was obtained from the Mechanisms Engineering Test Loop (METL) liquid sodium facility at Argonne National Laboratory. Loss-of-coolant type anomaly in the cold trap operation was generated by temporarily choking one of the blowers, which resulted in temperature and flow rate spikes. Results demonstrate detection of anomalies with sensor-averaged signal to noise ratio (SNR) close to unity. In another project, we have developed an ML method for reconstruction of temperature field using multimodal measurements and an LSTM autoencoder. Fiber optic distributed temperature sensor allows to obtain information about temperature field in the reactor coolant fluid. However, fiber material can degrade due to long-term exposure to elevated temperature and ionizing radiation. The performance of our method is demonstrated using data measured in Thermal Hydraulic Experimental Test Article (THETA) high temperature liquid sodium vessel with co-located multipoint thermocouple arrays and distributed fiber optic temperature sensor.

Biography

Dr. Alexander Heifetz is a Principal Engineer with the Nuclear Science and Engineering Division at Argonne National Laboratory. He received BS, MS, and PhD degrees from Northwestern University. At Argonne, he has been involved in nuclear energy enabling technology research focused on advanced sensing and nondestructive evaluation. Dr. Heifetz has authored over 50 peer-reviewed journal papers and 60 conference papers. He is a Senior Member of IEEE, member of ANS and APS. He is a co-recipient of two Best Paper Awards in 2019 and 2020, and Second Place Award in 2023 at the IEEE international Conference on Electro-Information Technology (EIT). He received the Impact Argonne Award in 2023.

Thursday, November 9. 2023
4:00 pm seminar

zoom link upon request