ML19354D551
| ML19354D551 | |
| Person / Time | |
|---|---|
| Issue date: | 08/21/1989 |
| From: | King T NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | |
| References | |
| NUDOCS 8912150003 | |
| Download: ML19354D551 (7) | |
Text
r
,l
- tonferen:e
Title:
IAEA Conference on Small and Medium Siaed Nuclear Reactors
'Date & Place:
August 21-23, 1989 -- San Diego, Celifornia
' NRC REVIEW OF U.S. ADVANCED REACTORS Thomas L. King, Chief
' Advanced Reactors & Generic Issues Branch Division of Regulatory Applications Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Comission Abstract NRC currently has several activities underway involving the review of ADVANCED REACTORS. These activities are associated with the review of STANDARDIZED reactor designs and are ultimately in support of DESIGN CERTIFICATION of these designs. Current emphasis in the reviews is on identification and resolution of key SAFETY ISSUES.
Introduction Currently, there is renewed interest and discussion in the U.S. on the use of
~
nuclear power for the generation of electricity. This has come about, not only from concerns about global warming, acid rain, clean air, etc., but also from advances over the past several years in the development of future reactor designs with the potential for enhanced safety beyond the level ob+ained in operating reactors. This potential for enhanced safety poses opportunities for development of safer energy supplies for the U.S. as well as poses challenges to the designers, operators and regulators to ensure that this expected enhanced safety does, in fact, result.
This paper discusses the safety evaluations being done by the Nuclear Regulatory Comission (NRC) on future reactor designs.
Background
All but one of the licensed nuclear power )lants operating in the United States today are light water reactors of tie pressurized or boiling water type. Although these plants have proven to be safe, they nevertheless have experienced sufficient technical and economic problems to warrant review r
of some fundamental aspects of reactor design for the future. Accordingly, in the early 1980s various vendor, government and utility organizations began to develop future reactor designs with improved safety. These improvements result from attempts to eliminate the problems and safety concerns uncovered by years of operating experience and also to provide simplified safety systems that are more reliable and potentially less expensive.
The NRC, recognizing the industry's efforts in developing future reactor designs, has also taken steps to establish a framework for licensing future designs and to help ensure that these designs achieve enhanced levels of safety over current generation designs. These steps include:
1) issuing severe accident, afety goal and advanced reactor policy statements, which state the Comission's safety ex6ectations for future
- designs, D (d l
egoen i
munm m
2) publishing a new rule that establishes requirements for certification of standardized reactor designs (10 CFR Part 52). Once approved by the NRC, the certified reactor designs would only require solution of site specific issues to obtain a construction permit and 3) initiating reviews of future reactor designs, including early interactions with designers to provide preliminary guidancre at the conceptual design stage on the acceptability of these designs.
Although I mention these actions briefly,ltimately licensing, of advanced they are extremely important in preparing the way for NRC's review, and u reactor designs.
Discussion The Commission considers future reactor designs to fall into three general categories.
These are:
evolutionary light water reactor (LWR) designs, advanced LWR designs, and advanced non-LWR designs, The table at the end of this paper lists the reactor designs currently under i
development by these categories.
The table also indicates those designs that are presently under review at NRC.
Although the Commission, in its Advanced Reactor Policy Statement, stated that i
advanced designs should achieve a level of safety at least equivalent to current generation reactors, it expects each of these three geneations of reactor designs to achieve an enhanced level of safety over the gevious generation.
Specifically, the evolutionary LWRs should have e9 anced safety characteristics over currently operating plants and the advanced designs should achieve an enhanced level of safety over the evolutionary designs.
In order to make judgments on the acceptability of these designs, the NRC must develop licensing requirements appropriate to e&ch design. The development of these requirements is a task the Comission has now embarked upon.
It involves reviewing each design, identifying the safety issues associated with the design and deciding how to assess its safety and set appropriate licensing criteria and requirements.
This latter item can be a complex task, particularl future designs propose a tradeoff of traditional safety features (y when the such as containnent and emergency planning) for greater core damage prevention.
Before l
describing the status of our advanced reactor reviews and and some of the key safety issues associated with these designs, I will briefly describe the approach employed in the review to illustrate how the Comission is addressing i
the development of the licensing requirements for future plants.
l-As mentioned earlier, the Commission intends to require that future plants achieve a level of safety at least equivalent to that of current gtneration l
LWRs.
To achieve this, the review of future plants has its roots in the approach, requirements and objectives developed to license current generation plants.
This includes maintaining a defense-in-depth safety approach, 4
2
complying with the guidance in the Comission's Severe Accident Policy Statement (as codified by a recent rule on standard plant reviews) and providing reasonable assurance that future designs meet the objectives proposed by the staff for implementation of the Comission's Safety Goal Policy.
Accordingly, for the future designs, the staff utilizes current rules and guidelines where practical; however, in many areas, additional guidance needs to be developed. Maintaining defense in depth is of primary importance in developing this guidance.
Defense-in-depth has been and remains a key component of the Commission's safety philosophy.
It ensures that reactors are designed with appropriate attention to accident prevention, accident termination, accident mitigation and emergency planning provisions. While defense in depth has been codified for i
current generation LWRs through NRC's rules and regulatory guidance, its i
application to future designs requires some new thinking as well as interpretation. The staff review approach allows flexibility in the j
application of defense-in-depth to future designs; however, it does not allow i
elimination of any one of its major components. For example, several advanced designs heve proposed the elimination of offsite emergency planning based upon a projected low probability of a release of radioactive material. This would be in conflict with the defense-in-depth concept, and I think the elimination of offsite emergency planning would not be acceptable, however remote the need for it might be. Therefore, we are evaluating the appropriateness of changes in current emergency planning requirements to give credit to reactor designs with long transient response times and a reduced likelihood of release, while still maintaining equivalent protection to the public.
With respect to the Comission's Severe Accident and Safety Goal Policies, the staff has proposed that the requirements for future plants shog/ reactor year ld be directed toward ensuring a core damage frequency of no greate than 10-and a large release frequency of no greater than 10 g/ reactor year.
To meet such objectives, improvements over current designs will likely be necessary.
However, the staff will not judge the acceptability of future designs by means l
of a single or limited set of top level gosis such as comparison of PRA results to LWR PRA results or safety goals.
Rather, the staff will identify those accidentsequencesandphenomenawhichthe{uturedegignsneedtoaddressto l
have a reasonable chance of meeting the 10- and 10- objectives.
With this brief overview of the NRC safet) evaluation approach, let me now turn to some of the key safety aspects and isstes associated with the advanced designs.
The advanced designs (both LWR and non-LWR) have certain common objectives and characteristics directed toward enhanced safety. These include:
Certification of a standardized design Simplification of design and higher levels of safety system l
reliability through the use of passive safety systems l
Greater emphasis on prevention of accidents that could lead to significant core damage Reduced depenoance on and vulnerability to operator actions l
Smaller size (i.e. modular designs and fabrication techniques) 3
t 4
c'l For the advanced LWR designs, the key new features being proposed include a passive decay heat removal and emergency core cooling capability, in addition, in the event of a core melt accident, passive core debris and containment heat removal capability are also provided. These designs are to have the capability, without operator action or AC power, to respond to various loss of power and loss of coolant type accidents in a highly reliable fashion.
Sufficient water inventory is to be provided to enable several days of passive cooling without additional water supplies being added.
For the advanced non-LWR designs, passive decay heat and containment heat removal capability are also being provided. However, these designs propose to use the natural convection of air, instead of water to. air, to remove decay heat. The use of air cooling essentially allows indefinite heat removal without operator action or electric power.
For such designs, the designers propose that core melt accidents can be essentially eliminated f rom consideration.
In addition, they have the capability to demonstrate by test the performance of the plant under certain severe chc11enges. However, as a result of this potential improved core melt prevention capability, these designs are proposing changes to some traditional features such as containment and emergency planning.
t Although the Comission has not made a final determination on all the safety issues associated with these designs, it is, nevertheless, useful to at least see what the issues are. Therefore, a brief sumary of the major safety issues associated with the advanced designs under review follows. Even though the thrust of this paper is advanced reactors, I will begin by sumarizing the major safety issues with the evolutionary LWR designs because of their potential applicability to the advanced LWRs.
The major safety issues associated with the evolutionary reactor designs are those involving the treatment of severe accidents. As I mentioned earlier, the staff in setting requ rements for future designs, is att to ensure that the objectives of 10 j/R-Y core damage frequency and 10 gmptinyarge release
/R-Y f requency are met. Accordingly, in the review of the evolutionary designs, certain severe accident issues and phenomena were identified which, based on existing risk studies, the staff believes must be prevented or mitigated in order for the evolutionary LWR designs to meet the core damage and large release frequency objectives. These issues and phenomena are:
providing sufficient redundancy, diversity and independence in the reactor shutdown, decay heat removal and coolant inventory control functions to essentially be able to withstand the failure to two active components (in addition to the initiating fault) and still perform the function, accommodate the generation and control of H 2
prevent or mitigate high pressure melt ejection accomodate and minimize core-concrete interaction provide containment heat removal capability under severe accident conditions 4
1 j
e
)
~
prevent the more likely containment bypass scenarios prevent the direct attack of core debris on the containment wall address severe external events It is ex >ected that the advanced LWR designs will also need to address these issues;. iowever, it is recognized the solutions may be different than for the evolutionary LWRs. Our review of the first of the advanced LWR designs i
(AP-600) is just beginning and the treatment of severe accident issues is an item that will receive much attention.
for the advanced non-LWR designs the major safety issues are broader than for the LWRs. They encompass:
the range of events / accidents to be considered in the design l
determination of appropriate siting source terms l
determination of appropriate defense in depth (particularly with l
respect to containment) l determination of appropriate emergency planning requirements numerous detailed technical issues such as control room design, primary coolant system integrity, etc.
Specific safety issues have been identified for the MHTGR in a draft Safety EvaluationReport(NUREG-1338). For the PRISM, a draft safety Evaluation Report has been prepared but not yet released.
Although the advanced designs are incorporating features that have the potential to result in safer plants, in many cases their performance remains to be demonstrated and their acceptance by the NRC is still under review.
Complicating their acceptance is the fact that, in some cases, the advanced designs are requesting trade-offs in other traditional safety features because of their increased core melt prevention capability.
Specifically, the MHTGR and PRISM designs are proposing significant changes in traditional containment l
and emergency planning requirements as well as other design features (such as control room design).
In addition, it is our understanding that the advanced LWRs are also considering proposing changes in the emergency planning area as well as other design features (such as elimination of the auxiliary feedwater system on the advanced PWR). This makes the assessment of their overall safety difficult, particularly since these designs and many of their new features are still in the development stuge. Comission review and action on those issues with policy implications will also be required.
i Accordingly, the questions we face go to the core of the philosophy we have been operating under for the past 30 years, as well as to important policies we have developed in more recent years. The potential magnitude of the changes l
embodied in our decisior,s and the profound implications they are likely to have on the future of nuclear power in the United States are forcing a particularly careful and thorough review. While our pace may seem slow, I 5
'[
l 1
i believe soundly based, thoroughly researched decisions will ultimately best serve the public and the industry.
J Perhaps the most visible decision we face is the issue of containment.
Clearly, this issue embodies a number of questions, some technical, some nontechnical. While the NRC decision-making process focuses on the technical question of whether a proposed design or procedure ensures adequate safety, we cannot be unmindful of either the status of the knowledge base on which we make our decisions, or the ultimate acceptability of new technologies to the public.
It is the lessons of ex>erience and understanding developed to date that NRC must consider. While tie passive features of the advanced designs do indeed appear to offer very significant safety benefits for a wide range of postulated accident scenarios, there are still some scenarios, particularly those involving primary system and fuel performance and certain low probability severe events, where the presence of a containment could improve safety by preventing the release of radioactive material to the environment.
These are, of course, very low probability events, so the difference in risk levels is quite small. Nevertheless we cannot ignore the full ran e of accidents, so thatwecanaccesstradeoffsbetweenaconventionalcontanmentandother alternatives.
Perhaps more important than these low probability identified events are the unknowns.
This history of technological development, in the nuclear industry as well as in other industries, amply demonstrates that we have seldom, if ever, enbarked on the development of a new technology with a thorough understanding of all the problems well in hand. We have always discovered, over the course of time, technological problems we hadn't envisioned, interactions we hadn't expected, and societal impacts we couldn't have imagined.
I hope we have become more skilled in predicting performance.
I believe we know more technically and we have a better perception of some of the systems interactions that are so important.
However, the NRC must take into account, in its decisions on advanced reactors, the possibility that there may be some new and as yet unanticipated vulnerabilities, and must require systems that are likely to be robust to such developments. Is a conventional containment structure the answer to this concern? That is one of the questions we are facing.
Conclusion i
We have embarked on task that holds the promise of safer nuclear power.
However, there are many hurdles yet to overcome. The Consnission encourages the work and early licensing interactions on the advanced designs continue. Based upon our reviews to date we believe that these designs ht e the potential to achieve a level of safety beyond that achieved by currently operation LWRs, provided satisf actory resolution can be reached on the many issues remaining to be settled, including the acceptability of some fundamental changes in how to achieve enhanced reactor safety.
6
f i-a.*
b Future Reactor Designs Evolutionary LWRs Designer Size Advanced Boiling Water Reactor (ASWR)*
General Electric 1356 Mwe System 80**
Combustion Engineering 1270 Pwe t
StandardPlant-90(SP/90)*
Westinghouse 1350 Mwe Advanced LWRs Advanced Passive-600 ( AP-600)*
Westinghouse 600 Mwe Simplified Boiling Water Reactor (SBWR)
General Electric 600 Mwe Safe Integral Reactor (SIR)
Combustion Engineering 320 Mwe Process inherent Ultimate Safety (PlUS)
ASEA Brown Boveri 600 Mwe Advanced Non-LWRs-Modular High Temperature Gas-Cooled General Atomic 137 Mwe Reactor (MHTGR)*
per module Power Reactor inherently Safe Module General Electric 138 Mwe L
(PRISit)*
per module CANDU-3 Atomic Energy of 450 Mwe Canada Limited
- Currently under NRC review l
- - - -