Controlling the power output of a nuclear reactor with fuzzy logic

Controlling the power output of a nuclear reactor with fuzzy logic

Ruan, D.
van der Wal, A.J.
TNO Fysisch en Elektronisch Laboratorium

1998

The application of fuzzy logic control (FLC) in the domain of nuclear industry presents a tremendous challenge. The main reason for this is the public awareness of the risks of nuclear reactors and the very strict safety regulations in force for nuclear power plants. The very same regulations prevent a researcher from quickly introducing novel control methods into this field. On the other hand, the application of FLC has, despite the ominous sound of the word "fuzzy" to nuclear engineers, a number of very desirable advantages over classical control, e.g. its robustness and the capability to include human experience into the controller. In the present paper we describe an FLC for controlling the power level of a nuclear reactor. The study is intended to assess the applicability of FLC in this domain. The final goal is to develop an optimised and intrinsically safe controller. After reviewing the available literature on FLC in nuclear reactors, an FLC is proposed and first tested by comparing it with the classical controller of BR1 (Belgium's first research reactor). In the next step the BR1 at the Belgian Nuclear Research Centre (SCK · CEN) was used as a test bed to implement a PLC-based hardware controller. The BR1 reactor is internationally regarded as a nuclear calibration reference. It therefore provides an excellent environment for this type of experiments, because over the years considerable knowledge of the static and dynamic properties of the reactor has been accumulated. The progress made in these experiments will be presented and discussed

Fuzzy logic control
Rule-base control
BR1 reactor
Fuzzy sets
Knowledge based systems
Nuclear power plants
Nuclear reactors
Power control
Programmable logic controllers
Risk management

http://resolver.tudelft.nl/uuid:acad2778-5b1c-4cfc-b56c-d00dc6e52fed

95115

Information Sciences, 110, 151-177

article