[en] The opportunities for improvement of reactor controls are generally in two areas. First the complexity of safety systems (which evolves from the philosophy of safety reliability through redundancy, diversity and independence) can be reduced. The reliability and diverse nature of passive shutdown, convective heat removal and hydrostatic mitigation of leaks in the primary boundary allow considerable simplification or deletion of active control and protection system, while at the same time improving reliability of the safety functions. Second, the operability of reactors may be improved by emphasizing passive response. If inherent safety features can be made to be an outer bound to a control system, then there will be more freedom to innovate and improve plant control with advanced diagnostics and control methods. We in EBR-II first identified constraints on control system design for inherent safety while doing the safety analysis for the inherent safety demonstration tests. Tests conducted in November 1987 showed several possibilities for controlling the plant with minimum control rods. Further tests are planned to investigate these alternate control system designs and identify issues and tradeoffs in inherent safety, complexity, and plant operability. The balance of this paper is thus divided into three sections, passive safety considerations and a summary of the earlier test results, the origin of focus on control and supporting tests in EBR-II, and future tests. 10 refs, 4 figs

[en] Two accidents are considered independently by USSR regulatory documents as maximum project accident (MPA) at nuclear power stations with fast reactors: pipe-line sealfailure of the first loop, which has no protective housing; contraction or blocking of the flow area in separate fuel assemblies (FA) due to swelling up, sedimentation of impurities from the coolant or due to ingress of foreign bodies, that leads to the coolant flow-rate decrease through this FA as well as to the damage, destruction or fusion of fuel elements in it with an extension of damages at one row of surrounding fuel assemblies. The considered system is intended for the monitoring of processes, which are associated with the second MPA accident. The initial stage of heat removal decline is accompanied by local sodium overheat and boiling. Possible result consists in the overgrowing of one fuel assembly local overheat and fusion into the whole core overheat and fusion. The task is to detect local overheat early and to prevent its overgrowing into the whole core overheat and fusion. For a number of years investigations have been carried out in the USSR on various methods of sodium boiling monitoring. The final aim of these investigations is working out the criteria, which determine the dangerous stage of the sodium boiling process in order to automate protective actions which exclude core fusion. The developed system of sodium boiling monitoring is intended for fast-neutron power reactors, which are being built now. 1 ref., 3 figs

[en] The Specialists' Meeting on ''Advanced Controls for Fast Reactors'' was held in Argonne, Illinois, USA, from June 20 to 22, 1989. The meeting was sponsored by the International Atomic Energy Agency on the recommendation of the IAEA International Working Group on Fast Reactors and was hosted by Argonne National Laboratory and the US Department of Energy. It was attended by 20 participants and observers from Argentina, France, Germany, Japan, India, the USSR, the United Kingdom, the United States of America, and the IAEA. The purpose of the meeting was to provide an opportunity for participating countries to review and discuss their views on design and technology for advanced control in fast reactors. During the meeting papers were presented by the participants on behalf of their countries and organizations. Presentations were followed by open discussions on the subjects covered by the papers and summaries of the discussions were drafted. After the formal sessions were completed, a final discussion session was held and summaries, general conclusions and recommendations were approved by consensus. A separate abstract was prepared for each of the 22 papers presented at this meeting. Refs, figs, tabs, diagrams and photos

[en] Power Reactor and Nuclear Fuel Development Corporation (PNC) and Japan Atomic Power Company (JAPC, supported by the Japanese ten electric utility companies) are jointly developing the self actuating shutdown system (SASS) for liquid metal fast breeder reactors. A synthetic program was made and put into practice to investigate the feasibility of temperature sensitive electromagnet (TSEM) type self actuated shutdown system to the Demonstration Fast Breeder Reactor whose construction is expected to start in late 1990s. As of the end of 1988, fundamental function and properties of the system have been obtained, which suggest that the shutdown system with the TSEM would meet the severer safety requirement of fast reactors. Successive studies are under progress to examine the performance and availability of the system. A potential demonstration test of the SASS in actual reactor condition is being studied. (author). 4 refs, 9 figs, 2 tabs

[en] Indian nuclear power programme envisages setting up a series of commercial fast breeder power plants in the early half of next century. A 40 MW thermal (13.2 MWe) Fast Breeder Test Reactor has been set up and a 500 MW(e) Prototype Fast Breeder Reactor is under design. So far all the nuclear power plants in India have been using only conventional hard wired instrumentation and control systems. Computers have been mainly used as data loggers. A beginning has been made in the use of computers in India's first Fast Breeder Test Reactor, commissioned recently, for on-line core supervision, in addition to data logging function. With the great advancement in digital distributed control systems world over, it is planned to use microprocessors, programmable controllers and central computers for the control systems of PFBR. In this presentation the approach to be followed in the architecture of such control systems in PFBR is outlined. Use of these systems for safety functions is approached with caution. Critical parameters will be segregated from less important ones for treatment. Arrangement of plant main control room and emergency control room in enhancing efficiency of operator functioning is given due importance. Graphic displays for alarm annunciation, trend display and incident analysis will be extensively used. Multiplexing information between local control stations and control room will be resorted to for non-safety systems. Redundancy in system configuration, self diagnostic features, fault tolerant design and signal validation techniques are planned for increasing system reliability and hence plant availability. (author). 1 fig., 2 tabs

[en] In the nuclear research center at Karlsruhe, a diagnostic expert system is developed to supervise a fast breeder process (KNKII). The problem is to detect critical phases in the beginning state before fault propagation. The expert system itself is integrated in a computer network (realized by a local area network), where different computers are involved as special detection systems (for example acoustic noise, temperature noise, covergas monitoring and so on), which produce partial diagnoses, based on intelligent signal processing techniques like pattern recognition. Additional to the detection systems a process computer is integrated as well as a test computer, which simulates hypothetical and real fault data. On the logical top level the expert system manages the partial diagnoses of the detection systems with the operating data of the process computer and to produce a final diagnosis including the explanation part for operator support. The knowledge base is developed by typical Artificial Intelligence tools. Both fact based and rule based knowledge representations are stored in form of flavors and predications. The inference engine operates on a rule based approach. Specific detail knowledge, based on experience about any years, is available to influence the decision process by increasing or decreasing of the generated hypotheses. In a meta knowledge base, a rule master triggers the special domain experts and contributes the tasks to the specific rule complexes. Such a system management guarantees a problem solving strategy, which operates event triggered and situation specific in a local inference domain. (author). 3 refs, 6 figs, 2 tabs

[en] Under normal circumstances, operating decisions are made in accordance with the operators understanding of written procedures and of the plant which they are controlling. To ensure plant safety, an operator must have sufficient knowledge to respond correctly when both routine and unusual events place demands on his knowledge and understanding of the plant and procedures. In order that plant operating procedures or the methods used in qualifying operators may be improved a quantitative measure of understanding is required. This paper proposes a methodology which could be developed to yield a quantitative assessment of an individual's understanding. An extension is outlined where the knowledge relevant at each stage of a progressing task is assessed. This is intended to allow the identification of those parts of a task which are only weakly supported by an operator's background knowledge and where for example additional training and/or improved documentation could be introduced to greatest effect. (author). 12 refs, 4 figs

[en] Data from boiling noise experiments in the BOR60 reactor in USSR has been supplied by the IAEA to enable analysis techniques to be compared. The signals have been analysed at RNL using two basic techniques. High Frequency RMS analysis Pulse Counting analysis together with two more sophisticated methods, Pattern Recognition and Pulse Timing analysis. All methods indicated boiling successfully, pulse counting proved more sensitive than RMS for the detection of the onset of boiling. Pattern Recognition shows promise of a very reliable detector provided the background can be defined. Data from an ionisation chamber was also supplied and there was good correlation between neutronic and acoustic signals. (author). 3 refs, 25 figs, 3 tabs

[en] The self-welding phenomenon of the magnetic materials used for the core and the armature of the temperature sensitive electromagnet (TSEM) was taken up as the major problem to disturb the actuation of the TSEM, as the design temperature for the fast reactor was getting to be higher. Self-welding test was carried out in high temperature sodium, whose results strongly suggested that self-welding might occur under the operation condition expected in the demonstration fast reactor. A countermeasure against the self-welding between the core and the armature was proposed and the improved TSEM was put into examination. The results are going to be presented in this paper. (author). 1 ref., 8 figs, 2 tabs, 2 photos

[en] The European reactor SUPER-PHENIX (SPX 1) is at the present time the largest Fast Breeder Reactor in operation in the world. Its instrumentation and control systems have been designed more than a decade ago, but they are still at the leading edge of technology, while an extensive use of computerized systems is made, even in the reactor protection domain. The most important challenge for the utility is now to operate this power plant industrially and to prove the effectiveness of the technological solutions that have been chosen. In this context, advanced control systems and operator aids are not a major concern, at least for the present time. For longer-termed applications, developments concerning advanced controls are not reactor specific and this paper briefly presents an overview of French research programs, essentially in the light water reactors domain that could find some openings on the Fast Breeders. (author). 10 refs, 7 figs