Potential application of neural networks to nuclear power plants

Conference · Tue Jan 01 04:00:00 EST 1991 · Proceedings of the American Power Conference; (United States)

OSTI ID:5105686

Uhrig, R E ^[1]

---

1. Oak Ridge National Lab., TN (United States). Instrumentation and Controls Div.

This paper reports on application of neural networks to the operation of nuclear power plants being investigated under a U.S. Department of Energy sponsored program at the University of Tennessee. Projects include the feasibility of using neural networks for the following tasks: diagnosing specific abnormal conditions, detection of the change of mode of operation, signal validation, monitoring of check valves, plant-wide monitoring using autoassociative neural networks, modeling of the plant thermodynamics, emulation of core reload calculations, monitoring of plant parameters, and analysis of plant vibrations. Each of these projects and its status are described briefly in this article. The objective of each of these projects is to enhance the safety and performance of nuclear plants through the use of neural networks.

• DOE Contract Number: FG07-88ER12824
• OSTI ID: 5105686
• Report Number(s): CONF-9104106--
• Conference Information: Journal Name: Proceedings of the American Power Conference; (United States) Journal Volume: 53
• Country of Publication: United States
• Language: English

Related Subjects

22 GENERAL STUDIES OF NUCLEAR REACTORS
220400* -- Nuclear Reactor Technology-- Control Systems
99 GENERAL AND MISCELLANEOUS
990200 -- Mathematics & Computers
COMPUTER CALCULATIONS
CONTROL SYSTEMS
FEASIBILITY STUDIES
NATIONAL ORGANIZATIONS
NEURAL NETWORKS
NUCLEAR FACILITIES
NUCLEAR POWER PLANTS
OPERATION
POWER PLANTS
REACTOR CONTROL SYSTEMS
REACTOR MONITORING SYSTEMS
REACTOR SAFETY
SAFETY
THERMAL POWER PLANTS
US DOE
US ORGANIZATIONS