Cylindrical Containment Vessel Research Articles

Unsteady Natural Convection in a Cylindrical Containment Vessel (CIGMA) With External Wall Cooling: Numerical CFD Simulation

In the case of a severe accident, natural convection plays an important role in the atmosphere mixing of nuclear reactor containments. In this case, the natural convection might not in the steady-state condition. Hence, instead of steady-state simulation, the transient simulation should be performed to understand natural convection in the accident scenario within a nuclear reactor containment. The present study, therefore, was aimed at the transient 3-D numerical simulations of natural convection of air around a cylindrical containment with unsteady thermal boundary conditions (BCs) at the vessel wall. For this purpose, the experiment series was done in the CIGMA facility at Japan Atomic Energy Agency (JAEA). The upper vessel or both the upper vessel and the middle jacket was cooled by subcooled water, while the lower vessel was thermally insulated. A 3-D model was simulated with OpenFOAM®, applying the unsteady Reynolds-averaged Navier–Stokes equations (URANS) model. Different turbulence models were studied, such as the standard k-ε, standard k-ω, k-ω shear stress transport (SST), and low-Reynolds-k-ε Launder–Sharma. The results of the four turbulence models were compared versus the results of experimental data. The k-ω SST showed a better prediction compared to other turbulence models. Additionally, the accuracy of the predicted temperature and pressure were improved when the heat conduction on the internal structure, i.e., flat bar, was considered in the simulation. Otherwise, the predictions on both temperature and pressure were underestimated compared with the experimental results. Hence, the conjugate heat transfer in the internal structure inside the containment vessel must be modeled accurately.

Numerical simulation of the effect of flange radial length on strain growth of cylindrical containment vessels

In this paper, the effect of radial length of flange on the strain growth in the elastic range of the cylindrical shell is studied by numerical simulation using LS-DYNA. It is found that the influence of the flange length on the first strain peak is small. As the radial length of the flange increases, the bending disturbance of the various frequencies of the cylindrical shell is excited which makes the linear modal coupling response is enhanced, so that the strain growth factor is increased. When more high-frequency parts are introduced into the strain response, the strain growth time will be correspondingly shortened. Therefore, it is recommended to use a flange as small as possible when designing the explosion containment vessel.

Strain Growth in a Finite-Length Cylindrical Shell Under Internal Pressure Pulse

Strain growth is a phenomenon observed in the elastic response of containment vessels subjected to internal blast loading. The local dynamic response of a containment vessel may become larger in a later stage than its response in the earlier stage. In order to understand the possible mechanisms of the strain growth phenomenon in a cylindrical vessel, dynamic elastic responses of a finite-length cylindrical shell with different boundary conditions subjected to internal pressure pulse are studied by finite-element simulation using LS-DYNA. It is found that the strain growth in a finite-length cylindrical shell with sliding–sliding boundary conditions is caused by nonlinear modal coupling. Strain growth in a finite-length cylindrical shell with free–free or simply supported boundary conditions is primarily caused by the linear modal superposition, possibly enhanced by the nonlinear modal coupling. The understanding of these strain growth mechanisms can guide the design of cylindrical containment vessels.

Fracture Mode Transition for Explosively Loaded GB/JB 20 Steel Containment Vessels

A rupture experiment was conducted on cylindrical explosion containment vessels (ECVs), where the fracture mode transition was observed. Microstructure examinations indicate the material GB/JB20 (AISI 1020) experienced a fibrous-to-cleavage fracture mechanism transition with increment of loading rate. Different from fracture mechanics method, a rate-dependent failure criterion is proposed to account for the dynamic fracture behavior, which is compatible with experimental observation that the material fails at low effective plastic strain when at high strain rates. A finite element analysis of a cylindrical containment vessel with different sizes of initial cracks was performed, where the overpressure caused by detonation was calculated, and the dynamic crack propagation and fracture mode transition were reproduced. In addition, a failure assessment including the estimation of limiting crack sizes corresponding to impulsive loading was conducted. It was found that a small variation of initial crack size has minor influence on the final fracture mode and profile, which is mainly dependent upon the intensity of impulsive load as well as the loading rate. The results also indicate that the crack propagates with strongly nonlinear speeding, most cracking length developed during the first structural vibration cycle.

Ductile and brittle failure assessment of containment vessels subjected to internal blast loading

Several major international design methods of explosion containment vessels (ECVs) refer to the items of pressure vessel codes and standards, where the fracture mechanics analysis of pressurized components should be performed to prevent the occurrence of brittle fracture. However a ductile damage mode in the form of adiabatic shear band (ASB) is frequently found as a failure mode for the containment vessels subjected to internal blast loading. A rate-dependent failure criterion was proposed to account for ASB propagation, and a finite element analysis of a cylindrical containment vessel with different size of cracks is performed, where the overpressure caused by detonation was calculated and the propagation of cracks and the final fracture profile are obtained, which shows a good agreement with experimental result. The Failure assessments based on ASB mode and failure assessment diagram (FAD) method were conducted, respectively. It was found that the final fracture mode primarily depends upon the intensity of explosive load as well as loading rate in ASB mode. The assessment result based on FAD method resembles that of ASB at low or intermediate loading rate. However with increment of loading rate, the difference of assessment results based on two methods became obvious, the comparison indicates that for the structures under high strain-rate loading the ASB assessment provides better estimation of crack growth than FAD method does.

Freezing Controlled by Natural Convection

Experiments were performed for freezing under conditions where the liquid phase is either above or at the fusion temperature (i.e., superheated or nonsuperheated liquid). The liquid was housed in a cylindrical containment vessel whose surface was maintained at a uniform, time-invariant temperature during a data run, and the freezing occurred on a cooled vertical tube positioned along the axis of the vessel. The phase change medium was n-eicosane, a paraffin which freezes at about 36°C (97°F). In the presence of liquid superheating, the freezing process is drastically slowed and ultimately terminated by the natural convection in the liquid. The terminal size of the frozen layer and the time at which freezing terminates can be controlled by setting the temperature parameters which govern the intensity of the natural convection. The stronger the natural convection, the thinner the frozen layer and the shorter the freezing time. In the absence of liquid superheating, a cylindrical frozen layer grows continuously as predicted by theory, but the growth rate is higher than the predictions because of the presence of whisker-like dendrites on the freezing surface.

Evaluation of computational techniques for LMFBR safety analyses

Analysis of Primary Containment Transients (APRICOT) is an ERDA sponsored project in which a variety of reactor safety analysis groups around the world have been invited to participate by performing calculations to verify capabilities of large computer codes used to analyze postulated core disputive accidents of liquid metal fast breeder reactors. Nine groups have performed calculations of the first three problems which were set, using ten computer codes. Two problems were simple test problems for which analytical solutions exist, namely an ideal gas shock tube, and a suddenly pressurized spherical cavity in an infinite elastic medium. The third problem concerns an explosion in a partially water-filled overstrong cylindrical containment vessel for which experimental data exist. A critique of the results of these calculations is given in this paper.

Aseismic design and dynamic analysis of nuclear power plants: Part I. Swaying motion

Some problems of the aseismic design for nuclear power plants are surveyed, and a proposed method of solution by dynamic analysis is illustrated with examples of the first Japanese power reactor JPDR (Japan Power Demonstration Reactor). Two major buildings of the JPDR represent different types of vibration mode. The square turbine building on the surface sand and concrete pile foundation is analysed as a swaying motion type structure, and the cylindrical containment vessel supported on the sandy shale is treated to be a rocking motion type. Gauges were mounted in the containment vessel to measure the vibration behavior caused by natural earthquakes for comparison with this calculation. Data so far obtained show fairly good agreement between the analysis and measurements. This paper presentst he case of the swaying motion type. The case of the rocking motion type will be presented in part II, to be published in the next issue and compared with the observed data.