The best estimate plus uncertainty approach in licensing of Atucha II

The Atucha II nuclear power plant, designed to produce 745 MW of electrical power, is equipped with a pressurized heavy-water-cooled and -moderated reactor (PHWR). The Atucha II Construction License was issued on July 14, 1981 upon performance of a previously submitted Preliminary Safety Analysis Report (PSAR) [1], basically fulfilling the requirements on Safety Analysis Reports (SARs) established by the International Atomic Energy Agency (IAEA) standard [2], although 52 Pressurized Heavy Water Reactors its format has been prepared in accordance with a widely adopted United States standard [3]. Together with the regulatory approval for the construction, the main safety-related issues were addressed by the Argentinean Authority, making reference to a formal licensing document titled “Protocol of Understanding,” issued in November 1977 [4]. According to this, it is recognizable that the licensing procedure would follow basically a probabilistic approach. Aside from the fact that the concept of the maximum credible accident was not adopted, some special subjects were also considered—for instance, the degree of redundancy for safety systems or the recognized nine classes of accidents for design analyses, including anticipated transient without scram (ATWS). After a long delay, a decision was made to resume construction, and to bring the plant into operation until 2010. Consequently, there has been a strong push for design finalization, including the issuance of a Final SAR (FSAR). To this aim, a twofold strategy is foreseen: the original safety design philosophy must be preserved, and recent advances in nuclear safety technology should be incorporated, as long as possible. Significant progress has been made in the field of nuclear safety in the past 20 years. Much of it is related to improvements in the ability to predict plant behavior during normal and accident conditions. Evolution of analytical models, supported by comprehensive experimental efforts, has made available powerful computational tools that provide support for detailed calculation of relevant phenomena for nuclear reactor dynamics. Within the framework of Atucha II project finalization, Nucleoele´ctrica Argentina (NASA) and the University of Pisa (hereafter also called UNIPI) have signed an agreement for supporting activities in the area of deterministic safety analysis methods. Measures for prevention of operational transients and accidents are taken into account in the design of nuclear power plants. Nevertheless, the occurrences of such events cannot be guaranteed. Therefore, it has to be demonstrated that they can be controlled at any time. SARs must address the design and the safety performance of structures, systems, and components and their adequacy for the prevention of accidents and for the mitigation of accident consequences if they occur. Accident analyses and the demonstration of the adequate plant safety performance must follow well established requirements and recognized practices for licensing purposes. Usually, deterministic safety analyses do consider a reduced number of limiting transients for which conservative rules for system availability and parameter values are often applied. Complementarily, a probabilistic approach emphasizes the completeness of the set of different scenarios and the use of best estimate methods. With the exclusion of the maximum credible accident from the range of the design basis spectrum for Atucha II, a break size of 10% on reactor coolant pipe (0.1 A) was reconfirmed [5] as the basis for fulfilling traditional regulatory requirements. Derived from the connected probabilistic approach, the double-ended guillotine break is considered to be a beyond-design basis scenario. Nevertheless, the demonstration of the design capability to overcome this event has still a relevant role in the safety performance evaluation. For this aim, however, the currently used conservative approach for safety analysis may not be sufficient to guarantee that safety margins still exist. Quantification of available safety margins by a best estimate approach seems to be a better strategy. The best estimate plus uncertainty approach in licensing of Atucha II 53 Among the general attributes of a methodology to perform accident analysis of a nuclear power plant for licensing purposes, the very first should be compliance with established regulatory requirements. For Atucha II, this means the requirements issued or adopted by the Autoridad Regulatoria Nuclear (ARN) of Argentina. A second important attribute deals with the adequacy and the completeness of the selected spectrum of events which, according to international recognized safety practices, should consider the combined contributions of deterministic and probabilistic methods. For consistency with the original plant design, the starting point for the developments was the methodology described in the SIEMENS Work Report KWU NA-T/ 1995/011, known as “the Bordihn’s Report” (see, e.g. [6]). Based on qualified tools and analytical procedures developed or available at the Dipartimento di Ingegneria Meccanica, Nucleare e della Produzione of the UNIPI, a modern and technically consistent approach has been built upon best estimate methods including an evaluation of the uncertainty in the calculated results (Best Estimate Plus Uncertainty or BEPU approach). To achieve the required level of quality for supporting licensing-related decisions, technically qualified and experienced staffs are engaged in the activities of development of such approach and its application for the safety analysis of Atucha II. On the basis of the above considerations, for the development of the proposed approach and to apply it for the licensing of Atucha II plant, the work followed the top-down strategy illustrated in the figure below. NA_SA – UNIPI work strategy to perform Atucha-II accident analysis ORIGINAL DESIGN SAFETY APPROACH ATUCHA II FINAL SAFETY ANALYSIS REPORT CHAPTER 15 – ACCIDENT ANALYSES MODERN TOOLS AND METHODS ARN REQUIREMENTS UNIPI QUALIFIED STAFF AND PROCEDURES WORK STRATEGY NA-SAUNIPI work strategy to perform Atucha II accident analysis. 54 Pressurized Heavy Water Reactors The sketch shows a top-down logical roadmap that is followed when performing the activities. The requirements of the ARN (the regulatory authority in the country where the reactor is installed) identify the target of the analysis and provide the mandatory steps, first block, the tip of the pyramid. Then, consideration is to be given to the safety requirements in the country (Germany) where the reactor was designed, second block: the evolution of rules during the longer than quarter-century period between the time of the design and the time when approval request is a radically prepared, receive proper attention. The unfortunate Three Mile Island (1979) and Chernobyl (1986) events occurred during the quarter-century period concerned, leading to key modifications in regulatory requirements all over the world, third block; the growth in computational capabilities and the availability of new experiments also contributed to changes in requirements. All of this is considered, with specific attention to safety or regulatory frameworks set by the IAEA and the United States Nuclear Regulatory Commission (USNRC). A research group was active at UNIPI since the 1960s to support the ambitious nuclear program in Italy, fourth block. As is well known, the entire nuclear program was interrupted following a referendum in 1987; nevertheless, the activity of the group continued, directing attention to safety issues in various countries in cooperation with international institutions such as IAEA and the Committee on the Safety of Nuclear Installation of Nuclear Energy Agency (NEA) (part of the Organisation for Economic Co-operation and Development). Because of the group’s recognized expertise in the area of nuclear reactor safety and its direct contacts with leading international specialists in nuclear safety, it attracted the attention of the Atucha II owner in Argentina. The framework of the cooperation between NA-SA and UNIPI was fixed by the content of Chapter 15 of the FSAR, fifth block, the basis of the pyramid, following the definitions and the content of the latest version of the USNRC NUREG-0800 document “Standard Review Plan.” Namely, it was clear since the beginning of the cooperation that: (1) it was necessary to analyze all transients part of Chapter 15 (of FSAR) by a BEPU approach; (2) the Instrumentation and Control of Atucha II needed to be modeled; (3) as far as possible, a unique qualified nodalization should be used for the analysis of all transients; and (4) three-dimensional neutron kinetics coupled with thermal hydraulics should be part of the analysis considering the large size of the core of Atucha II.