A Component-Centric Approach to Structural Analysis of Mechanisms

摘要

Structural analyses of mechanisms with components that move relative to each other provide a unique problem to the analyst building and running structural models. In these situations, the analyst usually has to either simplify the problem to a point where the results are unusable or maintain multiple models, which will take more effort to maintain and more resources to run the models. If a mechanism is simplified down to just analyzing one component at a time without regard for the other components in the system, the results will not be accurate because the loading effects of the other components will not be accounted for. In cases where all the components are included in the model the loading effects from the other components will be accounted for, however, a separate model will be required for each position. This paper presents a method of breaking down the complex mechanism into a component level model for each part of the assembly, while still accounting for all loading effects of the other components; in the Pivot Method the component under analysis stays stationary and the loading moves around the component to represent the different positions that it can take. In order to accomplish this task, a simplified model is used to generate loads at each of the joints. Once the pivot loads are known, a spreadsheet can be used to transform the loads to a coordinate system in which the individual component is being modeled. With the pivot loads known and all the loads transformed into the proper coordinate system the structural analysis of the component under investigation can continue. The intention of this paper is to introduce the Pivot Method and to demonstrate that it provides a good trade off between both the complexity of methods that model the assembly as a system, and those that focus on the component under question alone. To accomplish this, the analysis results of the Pivot Method models will be compared to results obtained from other methods, with the intention of showing that the Pivot Method will provide the same results while requiring less effort to model and less resources to run.