Testing of large-scale concrete containment structural elements

The tests described in this paper are part of an Electric Power Research Institute (EPRI) program (Research Project 2172-2) to provide a test-verified analytical method of estimating capacities of concrete reactor containment buildings under internal overpressurization from postulated degraded core accidents.Experimental study in Phase 2 of the investigation, on which this paper is based, includes tests of five large-scale specimens with steel liner plates representing structural elements of prestressed concrete containment buildings. Four square wall element specimens and one specimen representing the wall/basemat junction region were tested.This experimental work indicates that under internal overpressurization or other accident conditions, highly localized strains in the steel liner plate can result in liner tearing and subsequent containment leakage. These results support the theory of leak before break where liner tearing occurs in a controlled manner and leakage and depressurization occur rather than global failure.     

Results of large-scale test of discontinuity region in a prestressed concrete containment

The test described in this paper is part of an Electric Power Research Institute (EPRI) program (Research Program RP2172-2) to provide a test-verified analytical method of estimating capacities of concrete reactor containment buildings under internal overpressurization from postulated degraded core accidents.Phase 2 of the EPRI program, on which this paper is based, includes tests of five large-scale specimens with steel liner plates. The specimens represent structural elements of prestressed concrete containment buildings. Four full-scale square wall element specimens and one specimen representing the wall/basemat junction region were tested. This paper describes results of the wall/basemat junction region test.Results of this experimental work indicate that highly localized strains in the steel liner plate caused by internal overpressurization or other accident conditions can result in liner tearing and subsequent containment leakage. It appears that this liner tearing occurs in a controller manner. Extrapolating from these test results, leakage and depressurization is more likely to occur than global failure.     

Recent results on the evaluation of the overpressure response of concrete and steel containments

Analytical studies have been performed for the evaluation of the ultimate load capacity of concrete containment structures. In addition, analyses of steel containment models were carried out to validate computer codes for the analysis of steel containment structures. This paper reports on some of the results of these analyses, dealing first with the global ultimate load behavior of typical prestressed and reinforced concrete containment structures. The results of these analyses are described, with particular attention given to identifying local effects and failure mechanisms of concrete containment structures. On the basis of the global analysis results, local effects analyses were carried out which show clear evidence of large strain concentrations in the liner. The utility of the ABAQUS-EPGEN code is also demonstrated for three steel containment small-scale models tested by Sandia National Laboratory. The basic geometry of the models consisted of a thin cylindrical shell with a hemispherical dome. One of the models included ring stiffeners in the cylinder, and the other model included penetrations without ring stiffeners. The results of these calculations are presented without test data comparisons.     

Tension tests of concrete containment wall elements

Tension tests of concrete containment wall elements were conducted as part of a three-phase research program sponsored by the Electric Power Research Institute (EPRI). The objective of the EPRI experimental/analytical program is twofold. The first objective is to provide the utility industry with a test-verified analytical method for making realistic estimates of actual capacities of reinforced and prestressed concrete containments under internal over-pressurization from postulated degraded core accidents. The second objective is to determine qualitative and quantitative leak rate characteristics of typical containment cross-sections with and without penetrations. This paper covers the experimental portion the the EPRI program.The testing program for Phase 1 included eight large-scale specimens representing elements from the wall of a containment. Each specimen was 60-in (1525-mm) square, 24-in (610-mm) thick, and had full-size reinforcing bars. Six specimens were representative of prototypical reinforced concrete containment designs. The remaining two specimens represented prototypical prestressed containment designs.Various reinforcement configurations and loading arrangements resulted in data that permit comparisons of the effects of controlled variables on cracking and subsequent concrete/reinforcement/liner interaction in containment elements.Subtle differences, due to variations in reinforcement patterns and load applications among the eight specimens, are being used to benchmark the codes being developed in the analytical portion of the EPRI program.Phases 2 and 3 of the test program will examine leak rate characteristics and failure mechanisms at penetrations and structural discontinuities.     

Analytic study of the steel liner near the equipment hatch in a 1:4 scale containment model

A containment scale-model test, performed at Sandia National Laboratories, was loaded by overpressurization and the first leak was concluded to be caused by tears in the steel liner found near the equipment hatch. These tears were located in the vicinity of the vertical fold in between the general curved part and the embossment (vertical bend line). A 3D finite element analysis of the region near the equipment hatch, shows that high localized strains will develop in the vicinity of the bend line. It is shown that the liner separates from the concrete wall near the bend line when the containment expands. The tensioned liner will be in contact with the surface of the concrete wall in general, but near the vertical bend line the liner tends to be straightened out. This flexural behaviour cause high strains in the weld located in the bend line. The actual peak strain level is depending on the detailed geometry in the bend line and the failure strain level of a welded biaxial stressed zone is difficult to define. However, the analysis presented in this paper shows that the flexural behaviour in the bend line most likely contributed to the liner tears found in the scale-model test. A general conclusion from the study presented in this paper is that, the non-linear plastic behaviour of the liner is very sensitive to the detailed design and the interaction between the liner and the concrete.     

Current state of knowledge on the behavior of steel liners in concrete containments subjected to overpressurization loads

In the US, concrete containment buildings for commercial nuclear power plants have steel liners that act as the internal pressure boundary. The liner abuts the concrete, acting as the interior concrete form. The liner is attached to the concrete by either studs or by a continuous structural shape (such as a T-section or channel) that is either continuously or intermittently welded to the liner. Studs are commonly used in reinforced concrete containments, while prestressed containments utilize a structural element as the anchorage. The practice in some countries follows the US practice, while in other countries the containment does not have a steel liner. In this latter case, there is a true double containment, and the annular region between the two containments is vented.This paper will review the practice of design of the liner system prior to the consideration of severe accident loads (overpressurization loads beyond the design conditions).An overpressurization test of a 1:6 scale reinforced concrete containment at Sandia National Laboratories resulted in a failure mechanism in the liner that was not fully anticipated. Post-test analyses and experiments have been conducted to understand the failure better. This work and the activities that followed the test are reviewed. Areas in which additional research should be conducted are given.     

Concept study for a combined reinforced concrete containment

A variety of different types of steel and concrete containments have been designed and constructed in the past. Most of the concrete containments had been pre-stressed, offering the advantage of small displacements and a certain leak-tightness of the concrete itself. However, considerable stresses in concrete as well as in the tendons have to be maintained during the whole lifetime of the plant in order to guarantee the required pre-stressing. The long-time behaviour and the ductility in the case of beyond-design-load cases must be verified. Contrary to a pre-stressed containment a reinforced containment will only be significantly loaded during test conditions or when needed in case of an accident. It offers additional margins which can be used especially for dynamic loads such as impacts or for beyond-design events.The aim of this paper is to show the feasibility of a so-called combined containment which means a containment capable of resisting both severe internal accidents and external hazards, mainly the aircraft crash impact as considered in the design of nuclear power plants in Germany.The concept is based on a lined reinforced containment without pre-stressing. The mechanical resistance function is provided by the reinforced concrete and the leak-tightness function is provided by a so-called composite liner made of non-metallic materials. Some results of tests performed at Siemens laboratories and at the University of Karlsruhe which show the capability of a composite liner to bridge over cracks at the concrete surface will be presented in the paper.The study shows that the combined reinforced concrete containment with a composite liner offers a robust concept with high flexibility with respect to load requirements, beyond-design events and geometrical shaping (arrangement of openings, an integration of adjacent structures). The concept may be further optimized by partial pre-stressing at areas of high concentration of stresses such as at transition zones or at disturbances around large openings.     

Construction of a concrete containment model

This paper discusses the features and construction of a reinforced-concrete containment model that has been built at Sandia National Laboratories in Albuquerque, New Mexico. The model Light-Water-Reactor (LWR) containment building was designed and built to the American Society of Mechanical Engineers (ASME) code by United Engineers and Constructors, Inc. The containment model will be tested to failure to determine its response to static internal overpressurization. The results from testing the heavily instrumented containment will be used to assess the capability of analytical methods for predicting the performance of containments subject to severe accident loads as part of the US Nuclear Regulatory Commission's program on containment integrity.The scaled dimensions of the cylindrical wall and hemispherical dome are typical of a full-size containment. Features representative of a prototypical containment and included in the heavily reinforced model are equipment hatches, personnel airlocks, several small piping penetrations, and a thin steel liner attached to the concrete by headed studs.     

Ultimate analysis of BWR Mark III reinforced concrete containment subjected to internal pressure

Numerical analyses are carried out by using the ABAQUS finite element program to predict the ultimate pressure capacity and the failure mode of the BWR Mark III reinforced concrete containment at Kuosheng Nuclear Power Plant, Taiwan, R.O.C. Material nonlinearity such as concrete cracking, tension stiffening, shear retention, concrete plasticity, yielding of reinforcing steel, yielding of liner plate and degradation of material properties as a result of high temperature effects are all simulated with proper constitutive models. Geometric nonlinearity as a result of finite deformation has also been considered. The results of the analysis show that when the reinforced concrete containment fails, extensive cracks take place at the apex of the dome, the intersection of the dome and the cylinder and the lower part of cylinder where there is a discontinuity in the thickness of the containment. In addition, the ultimate pressure capacity of the containment is 23.9 psi and is about 59% higher than the design pressure 15 psi.     

Study of leaktightness integrity of containment wall without liner in high performance concrete under accidental conditions—I. Experimentation

When a partially saturated concrete wall is subjected to accidental conditions (high temperature and steam water pressure, as a LOCA or more severe conditions), water vapour penetrates the containment wall until saturation level of the containment atmosphere is achieved. The rate of penetration of water vapour through concrete is progressively reduced, leading to improvement of the leaktightness integrity of the concrete wall. In this paper, experimental studies involving the measurement of temperature, moisture propagation and pore pressures in a concrete containment wall are presented. The tests have been carried out on cylindrical specimens, made of high performance concrete (HPC) and having 1.3 m thickness (same thickness as a containment wall of a nuclear power plant). A finite element analysis is used to study the heat and mass transfer through the concrete wall. The results of this numerical modelling technique are presented in the second part of this study.     

Design and behaviour of French containments

In this report, the point is made that the French nuclear installations have two types of containments:

• - The first consisting of a pre-stressed concrete inner containment with a leakproof liner.

• - The second consisting of a pre-stressed concrete inner containment without a leaktight liner and an outer containment of reinforced concrete concentric with the former. The space between the two containments is maintained at a negative pressure, to intercept any leaks from the internal containment, which are filtered and discharged outside in the event of an accident.

After covering the mechanical design of these two types of containments, this report examines the existing safety margins for aircraft crashes and explosions resulting from the industrial environment.The report then considers in greater detail the leaktightness results of the double containments obtained during acceptance tests, as well as the leaktightness conditions while the reactor is operating.Finally, the report describes, for the case of containments with leakproof liners, the conditions of aging of the concrete and the associated pre-stressing.     

Research needs for design of concrete containment structures

Potential failure modes of reinforced concrete containment shells are outlined, especially those associated with pressure-induced cracking and seismic forces. A summary is given of experimental and analytical research needed to evaluate tangential shear capacity and stiffness, the interaction between liner and cracked concrete, peripheral (punching) shear capacity, radial shear behavior, and nonlinear dynamic analysis approaches.     

Ultimate capacity of a mark II primary containment using new probabilistic failure criteria

In the past few years, new failure criteria to determine the ultimate capacity of nuclear primary containments associated with exceeding probability have been developed. In this paper, a study concerning the Laguna Verde Mark II reinforced concrete containment is reported. This study was accomplished using an advanced non-linear constitutive and finite element model. Analyses were performed for beyond-original-design pressure and temperature assumptions. The paper describes the non-linear analysis methodology, the various failure criteria, and the application of the results in a probabilistic framework. The probabilistic approach addresses criteria for predicting liner tear, penetration failure, and through-wall shear failure. It attempts to assign reasonable estimates of leak area to different failure mechanisms and it allows the evaluation of conditional probabilities for the postulated severe accidents selected.     

Refined cracked concrete analysis of concrete containment structures subject to operational and environmental loadings

Recent commercial nuclear power plant containment concepts involve the use of large reinforced concrete structures to form pressure boundaries. Where these structures are not provided with an integral steel liner, excessive cracking of the concrete under loads could result in the loss of the pressure boundary integrity with the risk of over-pressurization of other structures. Cracking of concrete is a local phenomenon and considerable detail must be included in any analytical model to obtain sufficiently refined results for the prediction of crack size and propagation. This imposes severe limitations on the overall size of structures or structural components for which detailed cracking analysis can be considered directly. To overcome this restriction, a two step procedure was developed in which linear analyses were performed to obtain the gross response, and nonlinear cracking analyses were performed for selected portions of the structure to evaluate local cracking in detail. Through iteration, compatibility of behavior between the linear and nonlinear analyses was achieved with the gross response being used to extrapolate the local cracking results to predict cracking over the entire structure. This paper discusses the analysis procedures for the detailed evaluation of cracking in large reinforced concrete structures and components. Analyses performed for an actual unlined reinforced concrete containment structure using these procedures are discussed and results are presented.     

Thermohydraulics in a high temperature gas-cooled reactor prestressed concrete reactor vessel during unrestricted core heatup accidents

The long term core and primary loop heatup of an HTGR subsequent to loss of all forced circulation has been analyzed using a modified version of the CORCON code. The results indicate that if the liner cooling system is operating, or can be restarted within about 60 h, safe cooldown can be achieved, but significant core damage will occur. Without functioning liner cooling system the core heatup will lead to PCRV concrete degradation and the resulting concrete gas releases will ultimately cause containment building failure after 6 to 10 days.     

Model tests of turbine missile impact on reinforced concrete

The results of 25 impact tests on 1/11-scale models of reinforced concrete nuclear plant walls are presented. These tests determined experimentally the maximum velocity at which postulated turbine missiles are contained by typical reinforced concrete walls. The parameters varied were missile weight, velocity, orientation, and impact angle, as well as target design and thickness. The results showed that the NDRC perforation formula used extensively in current practice is overly conservative, whereas a newer empirical formula (CEA-EDF) gave reasonably conservative predictions of the test results. All but the most energetic postulated missiles are stopped by containment wall models, and steel liners on these walls are effective in suppressing backface concrete scabbing.     

Announcement

A computer code utilizing an appropriate finite element, material and constitutive model has been under development as a part of a comprehensive effort by the Electric Power Research Institute (EPRI) to develop and validate a realistic methodology for the ultimate load analysis of concrete containment structures. A preliminary evaluation of the reinforced and prestressed concrete modeling capabilities recently implemented in the ABAQUS-EPGEN code has been completed. This effort focuses on using a state-of-the-art calculational model to predict the behavior of large-scale reinforced concrete slabs tested under uniaxial and biaxial tension to simulate the wall of a typical concrete containment structure under internal pressure. This paper gives comparisons between calculations and experimental measurements for a uniaxially-loaded specimen. The calculated strains compare well with the measued strains in the reinforcing steel; however, the calculations gave diffused cracking patterns that do not agree with the discrete cracking observed in the experiments. Recommendations for improvement of the calculational models are given.     

Seismic proof test of a reinforced concrete containment vessel (RCCV): Part 2: Results of shaking table tests

A 1/8-scale model was constructed of a reinforced concrete containment vessel (RCCV) used in the latest advanced boiling water reactors (ABWR). Shaking table tests were conducted on it with input motions corresponding to or exceeding a design earthquake assumed for a real Nuclear Power Plant.The objectives of the tests were to verify the structural integrity and the leak-proof functional soundness of the RCCV subjected to design earthquakes, and to determine the ultimate strength and seismic margin by an excitation that led to the model's collapse. The model, the test sequence and the pressure and leak test results were addressed in Part 1. The shaking table test method, the input motions and the test results, including the transition of the model's stiffness, natural frequencies and damping factors and the effects of vertical input motions and internal pressure on the model's characteristics and behavior, the load-deformation, the ultimate strength, the failure mode of the reinforced concrete portion and the liner plate are described here. The seismic safety margin that was evaluated by the energy input during the failure test to a design basis earthquake will be described in Part 3. The analytical results of simulation using the multi-lumped mass model will be described in Part 4.     

Evaluation of calculational and material models for concrete containment structures

A computer code utilizing an appropriate finite element, material and constitutive model has been under development as a part of a comprehensive effort by the Electric Power Research Institute (EPRI) to develop and validate a realistic methodology for the ultimate load analysis of concrete containment structures. A preliminary evaluation of the reinforced and prestressed concrete modeling capabilities recently implemented in the ABAQUS-EPGEN code has been completed. This effort focuses on using a state-of-the-art calculational model to predict the behavior of large-scale reinforced concrete slabs tested under uniaxial and biaxial tension to simulate the wall of a typical concrete containment structure under internal pressure. This paper gives comparisons between calculations and experimental measurements for a uniaxially-loaded specimen. The calculated strains compare well with the measued strains in the reinforcing steel; however, the calculations gave diffused cracking patterns that do not agree with the discrete cracking observed in the experiments. Recommendations for improvement of the calculational models are given.