In a complex system, there are several sources of physical uncertainty, as well as there are many unintended interactions between the sub-systems. To predict the physical behavior of such a system, there is a need to treat the uncertainties in the physical quantities as well as there is a strong need for scaleable and reusable models that are easy to adapt and to modify. The essence of the MOSAIC approach to behavior modeling of complex systems is to support behavior analysis and simulation in an iterative and thus complex engineering process, by using encapsulated sub-models of components and of their interfaces. Various modeling and simulation aspects of an Atlas Copco Robot Boomer are presented in this report. The rig, which is an Atlas Copco product that is used for tunneling applications, is a complex system with long and slender mechanical parts, hydraulic actuators, sensors and a control system. The system is highly stressed during normal operations, and there are increasing customer demands for performance, with precision and speed as crucial parameters. Due to the modular structure of the product, it is advantageous to configure also the numerical models of the physical behavior of the design system in a modular way. An existing Ansys based FE model of a boomer has been reorganized, modularized, and adapted to allow for a wide range of simulations and analyses of the physical behavior. Methods and techniques to configure and to connect FE submodels, and techniques to utilize optimization techniques available in the FE software Ansys are elaborated on. Movements of the boomer to a target position is accomplished by changing the state of the actuator elements. Positioning errors due to dead weight and preloading and the sensitivity to some selected feature properties are also studied. It is shown that the Ansys optimization routines can be used to find a specified target position by minimizing the positioning error for the boomer. Modal extraction is used as a tool to study the stability limits for the hydraulic cylinders. A component within the system is studied with an ESO based Ansys macro routine that has been developed as a general tool for topological optimization.
Stockholm: KTH , 1999.