Developments in Cyclic Analysis and High Temperature Design

Design for cyclic loading is emerging as a key question for next generation power systems. Recent developments in techniques for cyclic stress analysis have significant implications for high temperature design. In the same way that limit load analysis is now being used to overcome the difficulties and guesswork of stress classification for steady primary loads, so shakedown and ratcheting analysis can eliminate the more difficult problems of stress classification for cyclic loads. The paper shows how reference stresses defined for shakedown and ratcheting provide rapid and conservative information for design against rupture and creep damage, deformation and strain accumulation, and ratcheting. These techniques will provide additional insights to designers and are likely to augment rather than replace, existing options. These ideas have existed in the research literature for some time, but have now become more accessible by the general industry with a new analysis technique in a commercial finite element code. Examples are given which demonstrate the methodology for nozzles having non-thermal secondary stresses, and prediction of long-term distortion in thermal shock problems.

Cyclic Thermoelastoplasticity Structural Calculation: Reliability and Sensitivity of Material Parameters

The essential cause of the ruptures of the structures under loadings is related most of times to the cyclical component of thermo-mechanical loadings. Forecasting the state of the structures is a very important task but simulating the structures behavior depends on a large data set of various parameters. In general, these parameters could be related to the geometry, loads and material characteristics or behavior law of the used phenomenological models. They are always known with a certain margin of incertitude that might have several origins such the experimental disparity in the material behavior or dependence on some physical variables like temperature. It is essential to measure degrees of influence of these parameters, thus the need for having analysis tools allowing the evaluation of the effects of parameters uncertainty on the structures behaviour. In this work, we study at first the reliability of the structural analyses making use of the Linear Kinematic Model of behavior in cyclic thermoelastoplasticity. Reliability will be applied to a first axis-symmetric structure. Then we propose a method based on the techniques of disturbances in order to determine the sensitivities of the phenomenological models with respect to the material parameters. This method will be applied through its two proposed approaches to a second structure. We made several assumptions. Deterministic calculations made it possible to validate the assuptions which relate to the choice of the absolute limit function. They also made it possible to confirm our expectations about the performance of the software package used (localization and values of the maximum plastic deformations). The assumption on the random variables laws are justified by the fact that the major part of the values taken are condensed around the mean value and that it is probably to exceed it as much than not to exceed it (normal law). The assumption on the value of the standard deviation holds all its validity owing to the fact that truncations must remain in conformity at the same time with the concept of density and physical limits of the variables. “The equations of dependences” make it possible to keep a sense which respects the relations of order that connects the random variables of the same nature as well reducing the cost of calculations.

Computational Studies for the Exhausted Flue Gas

Pulverized coal with low average heating, producing ashes with high percentage of silica, is fired inside the furnaces of a Thermal Power Plant (TPP) of Candiota, State of Rio Grande do Sul, Brazil. The produced hot flue gas heats the water of the ECONOMIZER 01 (ECO 01) placed inside the exhausted duct. Distorted velocity profile at inlet of ECO 01 and high concentration of abrasive particles of flue gas cause drastic erosion. So intensive has been the abrasive action that some well-identified tubes end up collapsing. The unpredictable fail has caused many non-scheduled stops of the TPP. A study focused on the reduction this effect, was set up years ago. The paper shows part of this study end present results, obtained from the numerical simulation analysis of the flue gas flow. Some technical solutions are suggested to reduce the erosion of tubes providing that avoiding it showed be impossible.

Piping Fluid Transients Made Simple: Part I — Theory of the SSFFC Methodology

Piping systems transporting fluid between plant components are subjected to a variety of anticipated and/or postulated flow changes that disturb their steady state operations. These changes cause the fluid flow to accelerate and/or decelerate. However, consideration of fluid elasticity transforms these disturbances into weak and/or strong propagating sound waves, depending upon the abruptness level of the fluid state change. This generates dynamic forces on the pipe segments of the piping system. A simple concept for understanding the piping fluid transient phenomenon from its physical perspective is presented. The piping system consists of several pipe segments, each segment having a constant cross-sectional flow area. The pipe segment is further divided into a consecutive series of zones. Each zone comprises two or three sub-zones of quasi steady state flow. The sub-zones are separated by interface fronts at which the jump in fluid pressure and velocity occurs across them. These fronts propagate and clash with each other to create the next temporal set of sub-zones quasi steady state flow. This method is denoted in this paper as steady state flow fronts clashing ‘SSFFC’. Clashing between the incident, transmitted and/or reflected wave fronts within the zone is introduced. As a precursor to the second part of a two-part publication, the SSFFC is physically illustrated and mathematically formulated to establish the temporal fluid steady state contained within each sub-zone constituting the pipe segment. The developed formulations are comparable to those instituted by the conventional method of characteristics. The pipe segment generalized fluid flow transient forces based on SSFFC methodology are also formulated. In the concurrent publication that forms part two of this presentation [8], sample applications of SSFFC methodology are illustrated.

The Structural Safety Test Research for 300 MW PWR Control Rod Driver Machine Pressured Shell

The safety analysis and test research are done to inspect the safety of 300MW control rod drive machine (CRDM) pressured shell structure in a systematic way. The test result agrees accords with FEA result. In actual operating mode (300 °C) inner-pressure fatigue simulative test, the axial and hoop pre-cracks are made, and the fatigue crack growth is observed. The load-bearing capability and deformation are tested in shell-burst test. Based on test research, this paper analyses theoretical load-bearing capability and safety margin in accordance with ASME CODE Sec. III design is calculated. Elastic-plastic fracture mechanics theory and GEGB.R6 method is used to analyze structure safety. The test results show pressure shell’s safety allowance is large enough.

Buckling Considerations for U-Shaped Bellows Utilized in Flexible Metal Hoses

This paper describes some of the considerations for evaluating the structural adequacy of flexible metal hoses utilized in a petro-chemical or process type environment. Specifically, the issues associated with the instability of the metal U-shaped bellows, from which the hose derives its overall flexibility and name, are reviewed and discussed in detail. In an effort to provide a comprehensive examination of the flexible hose’s use in the petro-chemical industry, a discussion of the applied mechanics associated with both column buckling of the bellows (also known as “squirm”) and in-plane buckling is presented. Results from a non-linear column buckling finite element analysis (FEA) of the U-shaped bellows are described and compared against previously published theoretical works on the instability of shells of revolution and most specifically, toroids. The applied loads in the finite element analyses include both internal pressure and transverse displacements (i.e., translations perpendicular to the longitudinal axis of the hose/bellows assembly). In addition, the guidance provided by the rules of the Expansion Joint Manufacturers Association Standards (EJMA) with regard to squirm are also reviewed and discussed. Finally, the results of both the theoretical and analytical investigations into the squirm phenomenon are utilized to identify some very practical solutions and recommendations to avoid the possibility of catastrophic failure of U-shaped bellows from column type instability.

Fatigue and Corrosion Fatigue Failures

A metallurgical failure analysis was performed for a hanger rod and a waterwall tube sample. The hanger is a rigid type and supports a long vertical run of piping. The fracture is in one of the threaded ends and the fracture surface consists of three regions. The outermost portion adjacent to the thread root has ratchet marks that are an indication of fatigue crack initiation. The center portion has concentric, oval shaped beachmarks. The oval shape is consistent with an applied loading due to two bending moments. The inner portion is the final fracture and is approximately 1/4 of the thread root area indicating relatively low remote stresses. The failure mechanism is fatigue based on the beachmarks on the fracture surface and the transgranular cracking. The lower slope waterwall tube failure had a window opening fracture appearance. The axial fractures forming the window are located at the edge of the membrane welds on the cold or backside. There are shallow toe cracks at the membrane weld on the tube outside surface. The fracture surface had multiple, thumbnail-shaped fatigue cracks connected to the inside surface. These fatigue cracks are due to the corrosion fatigue mechanism based on two factors: (1) the stress responsible for their growth is related to the unit thermal cycling and the welded panel geometry near the corner of the boiler, and (2) they are oxidized indicating a corrosion contribution.

Thermal Fatigue Evaluation Method Based on Power Spectrum Density Functions Against Fluid Temperature Fluctuation

Fluid temperature fluctuates at an incomplete mixing area of high and low temperature fluids in nuclear components. It induces random variations of local temperature gradients in structural walls, which lead to cyclic thermal stresses. When thermal stresses and cycle numbers are large, there are possibilities of fatigue crack initiations and propagations. It is recognized that there are attenuation factors depending on fluctuation frequency in the transfer process from fluid temperature to thermal stresses. If a frequency of fluctuation is very low, whole temperature of the wall can respond to fluid temperature, because thermal diffusivity homogenizes structural temperature. Therefore, low frequency fluctuations do not induce large thermal stress due to temperature gradients in structures. On the other hand, a wall surface cannot respond to very high frequency fluctuation, since a structure has a time constant of thermal response. High frequency fluctuations do not lead to large thermal stress. Paying attention to its attenuation mechanism, Japan Nuclear Cycle Development Institute (JNC) has proposed a fatigue evaluation method related to frequencies. The first step of this method is an evaluation of Power Spectrum Density (PSD) on fluid, from design specifications such as flow rates, diameters of pipes and materials. In the next step, the PSD of fluid is converted to PSD of thermal stress by the frequency transfer function. Finally, the PSD of thermal stress is transformed to time history of stress under an assumption of random phase. Fatigue damage factors can be evaluated from stress ranges and cycles obtained by the rain flow wave count method. Proposed method was applied to evaluate fatigue damage of piping junction model tests conducted at Oarai Engineering Center. Through comparison with direct evaluation from measurements and predictions by conventional methods, the accuracy of the proposed method was validated.

Analytic Prediction of Plastic Collapse Failure Pressure of Line Pipes

Accurate prediction of burst pressure of line pipes is essential for the safety design and integrity assessment of transmission pipelines. Different analytical and empirical formulae for determining the limit load of defect-free pipes have been proposed, but none is widely accepted and broadly validated. The commonly used method to determine the burst pressure by code in pipeline industry is based on the hoop stress of a pipe when it reaches a critical stress, such as flow stress or ultimate tensile stress of the pipeline steel. Recent experiments have shown that the criterion by code may be too conservative for modern high strength pipeline steels. Based on the plastic instability theory and the von Mises or Tresca yield criterion, theoretical solutions to predict the burst pressure of defect-free pipes have been proposed for years. However, it can be shown that experimental data for various pipeline steels lie between the two theoretical solutions of burst pressure of pressurized defect-free pipes, and fit the average result of the two solutions. In general, the von Mises prediction is the upper bound, and the Tresca prediction is the lower bound of burst pressure. Because traditional criteria like Mises and Tresca tend to bounded experimental data, a new failure criterion referred to as Average Shear Stress criterion is proposed. A plastic collapse analysis solution is developed as the corresponding theoretical solution of pipe limit pressure at plastic collapse. The new solution is formulated as a function of pipe geometry, strain hardening exponent and ultimate tensile stress of materials. Finite element results and experimental data are then introduced to validate the proposed solution. Comparisons indicate that the present solution matches the numerical results and the average experimental data of burst pressure of defect-free pipes for various pipeline steel grades.

Lateral Loading of Internally Pressurized Steel Pipes

The paper examines the denting response of tubular members and pipes subjected to lateral (transverse) quasi-static loading, in the presence of internal pressure. Tubes are modeled with nonlinear shell finite elements, and the numerical results are in good agreement with available experimental data. Using the numerical tools, a parametric study is conducted to examine the effects of pressure level, as well as those of denting device size and pipe end conditions. It is mainly concluded that for a given denting displacement, the presence of internal pressure increases significantly the corresponding denting force. A simplified two-dimensional heuristic model is also adopted, which yields closed-form expressions for the denting force. The model equations are in fairly good agreement with the test results and illustrate pipe denting response in an elegant manner.