REG NASA-LLIS-0819-2000 Lessons Learned Structural Stress Analysis.pdf

1、Best Practices Entry: Best Practice Info:a71 Committee Approval Date: 2000-04-18a71 Center Point of Contact: GSFCa71 Submitted by: Wil HarkinsSubject: Structural Stress Analysis Practice: This paper describes the general methodology for performing stress analysis for structures used in space applica

2、tions.Programs that Certify Usage: This practice has been used on theHubble Space Telescope, Gamma Ray Observatory, Superfluid Helium On-Orbit Transfer and Get Away Special (GAS) programs.Center to Contact for Information: GSFCImplementation Method: This Lesson Learned is based on Reliability Practi

3、ce number PD-AP-1318, from NASA Technical Memorandum 4322A, Reliability Preferred Practices for Design and Test.Benefit:Reliability of spacecraft structural components is greatly increased, and their cost and weight reduced by the systematic and rigorous application of sound stress analysis principl

4、es as an integral part of the design process.Implementation Method:Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Objectives:Structural stress analysis is performed in order to ensure that a structure will fulfill its intended function in a given lo

5、ads environment. It is important to anticipate all the possible failure modes and design against them. For a space structure, the most common modes of failure are as follows:a. Ultimate failure, rupture, and collapse due to stresses exceeding material ultimate strength,b. Detrimental yielding that u

6、ndermines structural integrity or performance due to stresses exceeding material yield strength,c. Instability (buckling) under a combination of loads, deformations, and part geometry such that the structure faces collapse before material strength is reached,d. Fatigue of material due to crack initi

7、ation and propagation under cyclic loads and fracture due to unstable crack propagation,e. “Excessive“ elastic static or dynamic deformations causing loss of function, preload or alignment, interference, and undesirable vibrational noise,f. Other time dependent material failure modes including stres

8、s corrosion, creep, stress rupture, and thermal fatigue.A spacecraft (S/C) structure is usually classified as primary or secondary. The primary structure consists of those elements which react to the overall S/C bending, axial, shear, and torsional loads. Secondary structure comprises those elements

9、 which do not appreciably contribute to overall S/C stiffness. Non-flight components are referred to as mechanical ground support equipment (MGSE). Structural stress analysis should define and address all the loads acting on the S/C primary and secondary structures. Table 1 summarizes the most commo

10、n loads encountered in the space applications.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-refer to D descriptionD Table 1. Summary of Spacecraft Loads Structural loads are specified at the maximum expected level and referred to as the design or l

11、imit loads. Usually, two or more of these loads act simultaneously and their combined effect needs to be considered. Note that the loads environment applied to the structure during the verification testing may be more significant than the loads experienced during flight. Many structural failures hav

12、e occurred during testing in the past. Therefore, these loads must be considered very carefully in the strength and fatigue calculations. It should be noted that this practice does not address all the possible loads a structure may encounter, such as impact with orbital debris.Analysis Philosophy:Th

13、e structural analysis should guide the design of the S/C and sizing of the components and provide a high degree of confidence. The analysis should be an integral part of the design process, thus minimizing design effort and time by eliminating redesign caused by failure during structural verificatio

14、n testing. An important benefit of performing stress analyses is the ability to determine design sensitivities and to conduct trade studies. Thus, effective optimization of the structure can be achieved, enhancing reliability while reducing cost and weight.Provided by IHSNot for ResaleNo reproductio

15、n or networking permitted without license from IHS-,-,-It is essential for the analysis to be conservative, i.e., the failure load predicted should be less than the actual load the structure can withstand. This is necessary in view of the uncertainties in the analysis assumptions and the variations

16、in the applied loads and material properties within normal bounds. The concept of an overall safety factor (SF) is introduced to account for various uncertainties and the limit loads are increased in proportion to the SF (Ultimate Load = SF x Limit Load). A typical SF value used for the ultimate fai

17、lure of flight structures is 1.4. In addition, a yield SF typically equal to 1.25 is selected to prevent structural damage or detrimental yielding during structural testing or flight. Additional safety factors may be used for fittings, castings, etc. to account for related uncertainties. The SF requ

18、irements may change depending on the responsible NASA center, the sponsoring agency, and the project.In addition to applying a SF, care should be given to conduct a conservative analysis using lower bounds for estimating the structures load carrying capacity. This will lead to a more reliable design

19、; however, there will be a weight penalty. It should also be noted that the analysis effort decreases with increasing conservatism. Therefore, at the start of the analysis, factors such as weight criticality of the structure, uncertainties in data, and available time for analysis should be considere

20、d.Analysis Overview:Stress analysis activities vary depending on the function and maturity of the phase, namely: (a) the Conceptual and Preliminary Design, (b) the Detail Design, and (c) the Verification phases.For the conceptual and the preliminary design activities, the design loads and the safety

21、 factors are considered to evaluate the feasibility and adequacy of the load paths and to size the major structural elements. Most of the trade/optimization studies are conducted in this phase. In the detail design phase, the bulk of the stress analysis activities takes place. Sizing and checking of

22、 the load paths is carried out in detail and the design is finalized. In the verification phase, stress analysis is used to analytically show that the structural testing will create the required minimum response (usually 1.25 times the limit loads) and the maximum response will not cause structural

23、damage or detrimental yielding.Analysis Methods:The general method and techniques used in structural stress analysis are outlined in Table 2. A description of each of these activities is given below.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-ref

24、er to D descriptionD Table 2. Stress/Failure Analysis Outline 1. Determination of the Structural Requirements and Loads: The first step of the analysis is the establishment of the requirements concerning strength, loads, displacements, service (cyclic) life, and verification (reference 1). In additi

25、on to strength, the design and sizing is sometimes dictated by maximum displacement requirements. The service life requirements may also dictate design and are to be clearly defined in every structural design and stress analysis activity.A S/C structure is subjected to a dynamic loads environment du

26、e to time varying accelerations, pressures, temperatures, and structurally or acoustically transmitted vibratory disturbances. The time history of loads seen by a specific component will be determined by its relative location as well as its stiffness and thermal paths to the rest of the S/C. This is

27、 determined by means of a dynamic structural analysis of the overall S/C referred to as “Coupled Loads Analysis.“ This is usually performed by the “Loads Group“ and is out of the scope of this paper. The “Loads Group“ provides the stress analyst with given equivalent static loads which envelope the

28、dynamic loads. It is important to make sure that the component does not see higher mechanical forces and this can Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-usually be accomplished by means of checking the resonant frequencies of the structure a

29、s discussed under activity 4 below. Coupled loads transient analysis is repeated and refined as the design progresses to provide more realistic and less conservative load levels. Dynamic or time-phased stresses can also be calculated for a structure to determine the actual stress history and peaks.

30、This requires calculation of stresses in conjunction with the coupled loads transient analysis.2. Material Characterization: Selection of proper materials for a given structure is based on various considerations such as strength to weight (specific strength) and stiffness to weight (specific stiffne

31、ss) ratios, ductility, resistance to corrosion (reference 2), thermal characteristics, cost, and ease of manufacturability. These and other structurally important material parameters are summarized in Table 3.refer to D descriptionD Table 3. Structural Goals Versus Material Parameters The stress ana

32、lyst must understand the pros and cons of stock type, material temper, and fabrication processes, since these may significantly affect material characteristics. (reference 3). Certain types of materials, for example, graphite bonded joints, require special consideration and development testing may b

33、e necessary for each specific application.For structural model development and stress analysis, the selected material can be classified as follows:Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-a. Homogeneity - Characterizes the dependence of struct

34、ural properties on location within the material.b. Isotropy - A measure of directional dependence of properties. Conventional metals can be classified as homogeneous, isotropic. A composite lamina is homogeneous (macroscopically), transversely isotropic; whereas a laminate is in general nonhomogeneo

35、us and anisotropic.c. Ductility - A ductile material can undergo a significant amount of plastic deformation before ultimate failure as opposed to a brittle material, which fails without any appreciable yield or warning. A ductile material is less sensitive to cracks and flaws since it can yield loc

36、ally and redistribute the excessive stresses. Reasonable fracture criterion will quantitatively screen out many non-ductile material applications.The classification of materials determines the type and the number of structural properties required in modeling the structure, which is discussed next.3.

37、 Structural Modeling: A mathematical model of the structure is developed in order to predict deformations, internal forces, and stresses. It is based on an idealization of the actual structure using simplifying assumptions on geometry, loads, and boundary conditions. There are basically two differen

38、t kinds of structural modeling the stress analyst can resort to:a. Computer model based on a numerical solution of the elasticity equations and boundary conditions that govern structural response. The part is represented using a finite number of degrees of freedom, by approximating the geometry usin

39、g discretization. The most common numerical method used in structural analysis is the Finite Element (FE) Method. (Reference 4). There are several commercially available FE analysis computer programs. The one most widely used in the industry is NASTRAN. (Reference 5). The structure is represented by

40、 a collection of “elements“ connected at “nodal points,“ or nodes, to each other. The governing equilibrium and compatibility equations are satisfied at each of the modal points and solved numerically. Results in the form of modal displacements, element internal forces, and stresses are output. Diff

41、erent kinds of structural analyses (e.g., stress, normal modes, forced dynamic response) that need to be performed should be identified so that the model can be built with necessary and sufficient detail.b. Analytical or hand calculations based on closed form solutions or empirical data given in var

42、ious sources for different geometries and loading conditions (References 6 and 7). The concept of a “freebody diagram“ is used to isolate and identify the internal forces or reactions acting on the part. For a “statically determinate“ case these reactions are calculated based on the equations of sta

43、tic equilibrium. For “statically indeterminate“ reactions, additional simplifying assumptions and analyses need to be made regarding structure deformations and load paths. Structure stresses and deformations are determined using the applied loads and the calculated reactions, and based on the soluti

44、ons or data available in the literature. The analyst can also utilize specialized and proven computer programs such as for the analysis of composites, pressure vessels, and truss structures.It is recommended that both approaches to structural modeling be used. The FE model should Provided by IHSNot

45、for ResaleNo reproduction or networking permitted without license from IHS-,-,-contain sufficient detail to represent the overall geometry and the important load paths.However, including “too much“ detail such as fillets, joints, and fasteners may increase the modeling (pre-processing), computing (p

46、rocessing), and results (post-processing) times significantly and sometimes without any appreciable advantage. Including too much detail also compounds the difficulty of the model and the results assessment. Therefore, it is recommended that these structural details be analyzed using the internal fo

47、rces obtained from the “coarse“ FE model and hand calculations. Hand calculations should also be used for the overall structure to approximately verify FE analysis results.4. Determination of Structural Response: The structural model(s) developed, material properties, and loading conditions are used

48、 to calculate the structural response, which consists of displacement, internal force, and stress distributions.An important consideration in the determination of structures response is whether or not it is linear. For a linear system the response is proportional to applied loads, and the principle

49、of superposition applies, that is, the response due to the application of many loads is equal to the sum of individual responses to each one of the loads. This is not the case for a nonlinear system, e.g., a structure undergoing “large“ strain (material nonlinearity). The determination of response for a nonlinear system is much more involved and time consuming than that of a