Virtual Prototyping for Marine Crane Design and Operations

Abstract

Marine crane system design is an interdisciplinary process involving such as mechanical design, kinematics, multi-body dynamics, hydraulics, and operational control-related tasks. Marine crane operations are inherently challenging due to system stiffness, heavy loading, unstable working platform, and external disturbances, especially under harsh weather conditions in rough sea fields. This dissertation introduces virtual prototyping (VP) for marine crane system design and operations, such as produce design space exploration, analytical study, risk finding, and training. Modeling and simulation in a virtual environment provides the user comprehensive time- and cost-efficient insights to the behaviors of complex dynamic systems. The proposed VP system for modeling, simulation and visualization not only supports the product and system design process, but also adds physics and dynamics to real-time simulations of crane operations. Despite the prevalent use of various computer-based tools, model development and handling the simulation of the various dynamic models in different domains simultaneously presents non-trivial complications. Currently, modeling and simulation of complex engineering systems is carried out domain-specifically and application-dependently focusing on the most interested and critical subsystem or operation phase. This is partly due to the fact that the behaviors of these dynamic subsystems in different domains are purposefully described mathematically based on their constitutive laws. What’s more, different users’ preference for the software tools complicates the communication between them. Effective and efficient communication setup requires not only the data exchange between the interactive dynamic models using different software tools, but also the user interface for manipulation, control and views. Reusing the existing knowledge in modeling and simulation efficiently requires a common standard for model integration. Correspondingly, it is necessary to redefine the component model structure and interfaces to provide the flexibility for model modification and interaction at the system level. The main objective of research is to develop an open, flexible, and efficient heterogeneous platform for the simulation of various interactive dynamic models, particularly models in different domains handled by different software tools. The proposed VP framework is based on the application of the functional mock-up interface (FMI) standard, which defines a shared format to support model exchange and co-simulation of dynamic models. With this common standard available, model development and handling can be performed separately with different domain-specific tools. Integration of the simulation only needs to take care of the data for interaction at a proper frequency. Therefore, modeling and simulation of the possibly stiff systems may be represented in different complexity levels and can be handled at their own timesteps. To support the integration of simulation and co-simulation using the VP framework, a component-based multi-objective approach is introduced based on the object-oriented modeling (OOM) method for model development of complex dynamic systems. The component library provides generalized basic models with different complexity levels for model integration and exportation depending on the purpose of the simulation. The VP system is designed to bridge the following two gaps in the current marine crane system simulation. Firstly, the need for an open and flexible platform oriented to the overall product and system design, modeling, simulation and visualization. Secondly, the need to reinforce virtual crane operation simulators with high fidelity models of physics and dynamics. As a case study for verification, the VP system was tested on mechanical design, model development, and simulation of the knuckle boom crane (KBC) systems. The crane designer and crane operation simulator proved the effectiveness of the proposed VP system in solving the identified challenges regarding modeling and simulation of complex multi-domain systems.