Self-teaching neural network learns difficult reactor control problem

Conference · Sun Jan 01 00:00:00 EST 1989 · Transactions of the American Nuclear Society; (USA)

OSTI ID:6769081

Jouse, W C

A self-teaching neural network used as an adaptive controller quickly learns to control an unstable reactor configuration. The network models the behavior of a human operator. It is trained by allowing it to operate the reactivity control impulsively. It is punished whenever either the power or fuel temperature stray outside technical limits. Using a simple paradigm, the network constructs an internal representation of the punishment and of the reactor system. The reactor is constrained to small power orbits.

• OSTI ID: 6769081
• Report Number(s): CONF-891103-; CODEN: TANSA; TRN: 90-023321
• Journal Information: Transactions of the American Nuclear Society; (USA), Vol. 60; Conference: Winter meeting of the American Nuclear Society (ANS) and nuclear power and technology exhibit, San Francisco, CA (USA), 26-30 Nov 1989; ISSN 0003-018X
• Country of Publication: United States
• Language: English

Related Subjects

22 GENERAL STUDIES OF NUCLEAR REACTORS
NUCLEAR POWER PLANTS
REACTOR CONTROL SYSTEMS
ARTIFICIAL INTELLIGENCE
HUMAN FACTORS
NEURAL NETWORKS
REACTOR OPERATION
REACTOR OPERATORS
TRAINING
CONTROL SYSTEMS
NUCLEAR FACILITIES
OPERATION
PERSONNEL
POWER PLANTS
THERMAL POWER PLANTS
220400* - Nuclear Reactor Technology- Control Systems