Midsurface Abstraction From 3D Solid Models for Engineering Analysis

An integral part of implementing parallel product and process designs is simulation through numerical analysis. This simulation-driven design requires discretization of the 3D part in an appropriate manner. If the part is thin or has thin sections (e.g., plastic parts), then an analysis model with reduced dimensionality may be more accurate and economical than a standard 3D model. In addition, substantial simplification of some details in the design geometry may be beneficial and desirable in the analysis model. Unfortunately, the majority of CAD systems do not provide the means for abstraction of appropriate analysis models. In this paper we present a new approach, based on midsurface abstraction, which holds significant promise in simplifying simulation-driven design. The method is user-friendly because very little interaction is required to guide the software in its automatic creation of the desired analysis model. It is also robust because it handles typical parts with complex and interacting features. Application of the method for feature recognition and abstraction is also briefly discussed.

On Simultaneous Triangularization of the Coefficients of Linear Systems

The coefficients of a linear nonconservative system are arbitrary matrices lacking the usual properties of symmetry and definiteness. An efficient way for the analysis of a nonconservative system is to reduce its coefficient matrices simultaneously to upper triangular forms. The purpose of this paper is to present some criteria for simultaneous triangularization and, when applicable, to expound a constructive procedure for triangularization.

Creation and Manipulation of Complex Displacement Features During the Conceptual Phase of Design

Surface feature design is not well supported by contemporary free form surface modelers. For one type of surface feature, the displacement feature, it is shown that intuitive controls can be defined for its design. A method is described that, given a surface model, allows a designer to create and manipulate displacement features. The method uses numerically stable calculations, and feedback can be obtained within tenths of a second, allowing the designer to employ the different controls with unprecedented flexibility. The algorithm does not use refinement techniques, that generally lead to data explosion. The transition geometry, connecting a base surface to a displaced region, is found explicitly. Cross-boundary smoothness is dealt with automatically, leaving the designer to concentrate on the design, instead of having to deal with mathematical boundary conditions. Early test results indicate that interactive support is possible, thus making this a useful tool for conceptual shape design.

Dynamic Response of a Generic Vehicle Floor Model to Coupled Transient Loads

A dynamic elastic large displacement response analysis of the bottom floor of a generic vehicle hull model subjected to empirically obtained coupled blast and impact loads has been conducted using three-dimensional (3-D) shell elements in the ADINA nonlinear dynamic finite element analysis code. For the impulse-dominated problem, the impact load is a square wave step function concentrated load while the blast loads from the detonation of an explosive are a series of distributed pressure loads approximated as triangular impulse loads with linear decay and varying arrival and duration times. The 3-D numerical model has been generated using the PATRAN3 modeling code and converted to the ADINA finite element input data deck using the ADINA translator and careful inclusion of appropriate material properties as well as initial and boundary conditions. Monolithic single-layered four-noded quad shell elements were sufficient to model the bottom floor and the left- and right-horizontal and vertical sponsons as well as the lower front glacis. Although several simplifying assumptions and approximations are made during the generation of the basic floor model, material properties, and the forcing functions, the investigation gives valuable insight into the response behavior of a generic hull bottom floor to externally applied coupled blast and impact loads and provides an inexpensive nondestructive method of evaluation of the structural integrity of modern vehicles subjected to spatially varying transient loads.

Automated Reuse of Solutions in Manufacturing Process Planning Through a Case-Based Approach

This research focuses on the development of an object-oriented case-based process planner which combines the advantages of the variant and generative approaches to process planning. The case-based process planner operates on general 3D prismatic parts, represented by a collection of features (eg: slots, pockets, holes, etc.). Each feature subplan is developed by the case-based planner. Then the feature subplans are combined into the global process plan for the part via a hierarchical plan merging mechanism. Abstracted feature subplans correspond to cases, which are used in subsequent planning operations to solve new problems. The abstracting and storing of feature subplans as cases is the primary mechanism by which the planner learns from its previous experiences to become more effective and efficient. The computer-aided process planner is designed to be extensible and flexible through the effective use of object-oriented principles.

A Generalized Method for the Classification and Extraction of Form Features

Feature-extraction techniques address the primary limitation of feature-recognition approaches, namely their lack of generalization. This paper presents a boundary-based procedure for the classification and sequential extraction of form features from the CAD models of objects with planar surfaces. Form features are first classified based on their effect on the boundary elements of a basic shape. Geometric reasoning is then used to obtain generalized properties of the form-features’ classes. Finally, form-features’ classes are sequentially extracted based on the recognized properties. At the onset of each extraction stage, the object is viewed as an initial basic shape that has been iteratively altered through the introduction of form features.

A Modeling and Design Method for Flexible Manufacturing System Controller Software

“Entity-life modeling” (ELM) is a general method for the modeling and design of concurrent software. One area where it can applied is for the control software for flexible manufacturing systems (FMS). The application of ELM allows the integration and overall control of the operation of the numerically controlled machinery, the material transportation and storage facilities into automated factories within the context of computer integrated manufacturing (CIM). The development of flexible manufacturing has been hampered by the lack of such a general approach to FMS software design. ELM is based on the principle that processes and objects in the software are patterned after concurrent “threads of events” and objects in the problem domain. A job in an FMS represents such a thread with events such as “pick from storage”, “place on stand”, etc., which all occur sequentially and with certain time intervals. Several job threads are in progress simultaneously as different jobs are being processed. In the software, a thread of events is represented by a process, such as an Ada task. The direct coupling between the analysis and an efficient control-system software implementation is an advantage over other analysis models, such as Petri nets. A simple FMS is used as an example. It consists of a storage facility and a number of numerically controlled workstations. A conveyor belt is used for the transportation of parts between storage and workstations and between workstations.

Fuzzy Control of an Inverted Pendulum

An inverted pendulum system has been designed and constructed as a physical model of inherently unstable mechanical systems. The vertical upright position of a pendulum is controlled by changing the horizontal position of a cart to which the pendulum is hinged. The stability of the system has been investigated when a fuzzy controller is used to produce the control signal, while making a single measurement. It has been shown that by using simple fuzzy rules to allow real time computation with a single angular position measurement, the system can not be made absolutely stable. However, the stability and performance of the system have been considerably improved by shrinking the membership functions of angular position, computed angular velocity and control signal when inverted pendulum is very close to the vertical upright position.

An Interactive CAPP Kernel Based on a Blackboard System Architecture

This paper describes the ideas and developments that lead to a new Computer Aided Process Planning (CAPP) approach : a blackboard-style, interactive CAPP system, in which the human process planner is assisted by expert modules, each capable of performing a process planning task. At all times, the human operator has full control over all planning activities. An important reasoning mechanism, that is inevitably linked with the interactive CAPP kernel, is opportunistic process planning. The implementation of this mechanism in an automatic feature based CAPP system will also be elaborated in this paper.

Rapid Prototyping of Fuzzy Logic Controllers

The rapid prototyping of fuzzy logic controllers is accomplished by using the tools Matlab, Simulink, Fuzzy Logic Toolkit, and Real-Time Workshop. Device drivers were developed for Simulink for interfacing with DT2801 and DT2821 data acquisition boards. The fuzzy logic inference engine for the Fuzzy Logic Toolkit was modified to allow the systems to work as independent programs and to be downloadable to DSP (Digital Signal Processing) boards. Simulink is used to graphically implement fuzzy logic controllers. The Real-Time Workshop is used to compile blocks from Simulink into C code, then into an independent executable program, both on the PC and a dSpace DSP (Digital Signal Processing) board. Graphical interfaces are created and debugged by using dSPACE’s tools, Cockpit and Trace. By combining these tools, real-time fuzzy logic controllers are developed in laboratory environments.