Investigation of Shock Analysis Techniques for Naval Applications
Open Access
- Author:
- Sammut, Jason
- Graduate Program:
- Mechanical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 20, 2023
- Committee Members:
- Robert Francis Kunz, Professor in Charge/Director of Graduate Studies
Matt Lear, Outside Field Member & Dissertation Advisor
Allison Beese, Major Field Member
Robert Campbell, Chair of Committee
Dan Russell, Outside Unit Member
Brian Olson, Special Member
Ron Couch, Special Member - Keywords:
- Mechanical Shock
Naval Design
Shock Spectrum
Finite Element Analysis - Abstract:
- Mechanical shock requirements are often a major limiting factor in naval designs. To ensure equipment is able to withstand the severe dynamic loading environments caused by near-miss shock events, components must be designed with sufficient shock durability without hindering system performance. Therefore, analyzing the shock response of a component to potential in-service or qualification test environments is a critical step in the design process. However, due to the complexity of these shock events and a system’s unique dynamic behavior, substantial inconsistencies exist between test environments, simulation methods, and analysis techniques. This research contributes to the advancement of naval shock analysis by addressing the usage of shock equivalency metrics, shock response analysis methods for qualification tests, and techniques for simulating shock events through existing test data. Although multiple types of shock spectra may be used to compare shock events across a range of frequencies, shock spectra alone cannot be used to determine equivalency between shocks for a given application. This research derives a relationship between shock spectra, modal parameters, and maximum structural response to form a singular metric for shock severity referred to as the Modal-Weighted Shock Spectrum Summation (Λsum). Λsum calculated from different types of shock spectra were shown to be directly correlated to the estimated maximum response parameters of a given structure. Therefore, this metric may be used to quantitatively compare the severity of shock events as applied to a particular structure of interest to better inform the development of shock specifications. Additionally, these derivations guide the appropriate usage of spectrum calculations by identifying which types of shock spectra are directly correlated with specific response parameters. Next, this research proposes the simulation of shock qualification test environments through explicit dynamic analysis to address the limitations presented by existing shock response analysis methods. Utilizing modern modeling techniques and computing power, explicit dynamic analyses are able to capture the response dynamics of potential designs to qualification test environments within reasonable computation times, leading to more relevant and accurate shock analyses. To demonstrate the feasibility of explicit dynamic modeling of qualification test setups, an initial model of the Medium-Weight Shock (MWS) test was created and validated with two sets of experimental data. The shock environment that was produced from this initial model closely matched the existing test data, but it was shown that further characterization of each test setup is necessary to ensure detailed system dynamics are accurately captured with final model validation. MWS simulations of sample test articles were then performed with this initial model, and results were compared to corresponding analyses from existing naval shock analysis methods. This comparison illustrated that, unlike explicit dynamic simulations, the existing analysis methods suppress high-frequency test article response which significantly reduces estimated shock severity compared to qualification test events. This further demonstrated the need to move qualification test analyses toward explicit dynamic simulations for future designs. Finally, a common technique for performing shock test simulations from experimental data was investigated. A comparison of analytical beam dynamics indicated that applying acceleration profiles from internal measurement locations to drive a model of a tested system effectively changes the boundary conditions and excitation profile of that system. Consequently, an acceleration-point-driven model cannot properly replicate the loading environment experienced by a test structure, leading to highly inaccurate dynamic response results. This inequality was derived by comparing the modal responses of an acceleration-point-driven beam model to those of a base-excited beam model, and it was shown that the total dynamic response throughout both beam models cannot be matched for any arbitrary shock input. Three case studies were then performed to demonstrate the disparity between each beam model for different types of sample shock environments. The case studies showed that acceleration-driven models tend to greatly overestimate structural shock response and perform particularly poorly with high-displacement shock events. Therefore, it was concluded that the alternate methods of transient transfer functions, empirical mode decomposition, or shock reconstruction should be further developed to leverage experimental data in future analyses.