Jing-Rong Li

Desktop virtual reality for maintenance training: an object oriented prototype system (V-REALISM)

Virtual reality (VR) provides an intuitive and immersive human-computer interface, which can be an efficient tool for maintenance training. In this paper, a low-cost VR application for maintenance training, which is an object oriented prototype desktop VR-enabled system for maintenance (V-REALISM), is presented. It has successfully integrated a novel disassembly sequence planning technique and evolutionary algorithms with virtual reality using the object oriented methodology. Details of the system architecture design and its implementation are presented. A case study on the maintenance of a centrifugal pump system in a virtual workshop environment was used to illustrate the performance of the V-REALISM. The result shows the potential of desktop VR to maintenance training in cost reduction.

An object-oriented intelligent disassembly sequence planner for maintenance

This paper presents a prototype object-oriented intelligent disassembly sequence planner for maintenance. A novel disassembly representation scheme known as disassembly constraint graph (DCG) has been proposed and implemented as a prototype system in this work. Using the DCG, all the possible disassembly operations that are needed for the maintenance of certain components or subassemblies can be deduced. Subsequently, a sequence-based optimisation technique, genetic algorithms, is employed to generate near optimal disassembly sequence from all the feasible combination of these disassembly operations. Based on the DCG, an object-oriented intelligent disassembly sequence planner for maintenance has been developed. The prototype system comprises three modules each having a set of objects. Users are able to view the entire disassembly process in accordance to the near optimal sequence generated by the planner via a graphical user interface. Two case studies, which are used to illustrate the effectiveness of the planner, are presented.

A rough set-based fault ranking prototype system for fault diagnosis

Fault diagnosis is a complex and difficult problem that concerns effective decision-making. Carrying out timely system diagnosis whenever a fault symptom is detected would help to reduce system down time and improve the overall productivity. Due to the knowledge and experience intensive nature of fault diagnosis, the diagnostic result very much depends on the preference of the decision makers on the hidden relations between possible faults and the presented symptom. In other words, fault diagnosis is to rank the possible faults accordingly to give the engineer a practical priority to carry out the maintenance work in an efficient and orderly manner. This paper presents a rough set-based prototype system that aims at ranking the possible faults for fault diagnosis. The novel approach engages rough theory as a knowledge extraction tool to work on the past diagnostic records, which is registered in a pair-wise comparison table. It attempts to extract a set of minimal diagnostic rules encoding the preference pattern of decision-making by domain experts. By means of the knowledge acquired, the ordering of possible faults for failure symptom can then be determined. The prototype system also incorporates a self-learning ability to accumulate the diagnostic knowledge. A case study is used to illustrate the functionality of the developed prototype. Result shows that the ranking outcome of the possible faults is reasonable and sensible.

Interactive visualization of complex dynamic virtual environments for industrial assemblies

Interactive visualization of assemblies in a virtual environment is important to the industrial applications like design verification and maintenance processes training. In general, the difficulty and the complication of the interactive visualization of industrial assemblies arise from the facts that an assembly model generally has a large number of primitives and high depth complexity; furthermore, dynamic environments are required for simulating assembly manipulations. In this paper, a dynamic data model for visualization of industrial assemblies is proposed. The data model integrates a spatial data set of hierarchical model representation and a dynamic transformation mechanism for runtime adaptation to the changes and movements of the assembly structures. Based on the dynamic spatial database, a visibility optimization approach that can efficiently reduce the overall geometry is presented. The complexity reduction is accomplished through view frustum culling, non-conservative occlusion culling, and geometry simplification. By which, significant speedup in user interaction rates has been achieved. Since all the techniques developed here are supported by mainstream graphics hardware, thus the solution is able to provide users with practical real-time visualization of complex dynamic virtual assembly environments with affordable graphics hardware.

A Tabu-enhanced genetic algorithm approach for assembly process planning

Over the past decade, much work has been done to optimize assembly process plans to improve productivity. Among them, genetic algorithms (GAs) are one of the most widely used techniques. Basically, GAs are optimization methodologies based on a direct analogy to Darwinian natural selection and genetics in biological systems. They can deal with complex product assembly planning. However, during the process, the neighborhood may converge too fast and limit the search to a local optimum prematurely. In a similar domain, Tabu search (TS) constitutes a meta-procedure that organizes and directs the operation of a search process. It is able to systematically impose and release constraints so as to permit the exploration of otherwise forbidden regions in a search space. This study attempts to combine the strengths of GAs and TS to realize a hybrid approach for optimal assembly process planning. More robust search behavior can possibly be obtained by incorporating the Tabu’s intensification and diversification strategies into GAs. The hybrid approach also takes into account assembly guidelines and assembly constraints in the derivation of near optimal assembly process plans. A case study on a cordless telephone assembly is used to demonstrate the approach. Results show that the assembly process plans obtained are superior to those derived by GA alone. The details of the hybrid approach and the case study are presented.

A CAD-linked virtual assembly environment

This paper presents a novel approach to integrate VR and CAD for virtual assembly, namely, CAD linked virtual assembly environment. Compared with most existing VR and CAD integration approaches, the proposed approach is a kind of external integration with a CAD system through automation interfaces. A novel hybrid virtual assembly model is developed in this paper to support this external integration. It works as a combinator that links the precisely represented design dataset in CAD models and the corresponding hierarchically structured virtual assembly dataset in the VR application. With this model, a framework has been developed to embed and apply CAD assembly constraints on the virtual assembly structure. Constraint-based manipulation is then implemented in the virtual assembly environment. Moreover, with this framework, advanced computations, such as runtime DOF determination, disassembly directionality computation, and assembly/disassembly sequence generation, which rely on assembly constraint information, can be feasibly integrated into the virtual assembly application with much less effort. The approach is also more practical in terms of implementation due to two advantages: (1) It does not require the implementation of a complicated support for solid modelling in the virtual assembly application. (2) A virtual assembly environment can be automatically created and updated from the linked CAD models.

RMINE: A Rough Set Based Data Mining Prototype for the Reasoning of Incomplete Data in Condition-based Fault Diagnosis

Condition-based fault diagnosis aims at identifying the root cause of a system malfunction from vast amount of condition-based monitoring information and knowledge. The needs of extracting knowledge from vast amount of information have spurred the interest in data mining, which can be categorized into two stages data preparation and knowledge extraction. It has been established that most of the current data mining approaches to fault diagnosis focus on the latter stage. In reality, condition-based monitoring data may, most of the time, contain incomplete, or missing data, which have an adverse effect on the knowledge or diagnostic rules extracted. Several approaches to deal with missing data can be found in literature. Unfortunately, all of which have serious drawbacks. In this paper, a novel approach to the treatment of incomplete data is proposed for the data preparation stage, while a rough set based approach has been developed to pre-process the data for the extraction of diagnostic rules. The two-stage data mining technique is implemented into a prototype system, RMINE, which also possesses a self-learning ability to cope with the changing condition-based data. A real industrial case study of a pump system is used to demonstrate the fault diagnosis process using RMINE. The result has shown the potential of RMINE as a general data mining prototype to condition-based fault diagnosis.

Graphics-assisted cutter orientation correction for collision-free five-axis machining

Automated cutter orientation correction is very important for achieving collision-free five-axis NC machining, in particular to the machining situations with complex collisions, i.e. multiple-points colliding or fully engaged colliding. In the current paper, we present a novel approach to identify admissible tool orientations for collision avoidance in five-axis ball-end milling with complex machining environment. The proposed approach is built upon two techniques: graphics-assisted cubic mapping; and instantaneous visibility and accessibility cones computation. For each colliding cutter location (CL), the graphics-assisted cubic mapping algorithm computes the cutters instantaneous visibility cone with respect to machining environment and the occlusion depth along each obstructed direction. With the given instantaneous visibility and occlusion depth information, an algorithm of spherical region shrinking is then developed to compute the cutters accessibility cone, which represents the aggregate of all admissible directions of the cutter for current machining point. Therefore, a collision-free cutter orientation can be ensured by adjusting the cutter with the accessibility cone. The mapping computation is performed very efficiently by taking the advantage of rapid performance from graphics hardware. By employing instantaneous visibility and accessibility computation, the accuracy and robustness in handling complex collision situations is improved. Also the required computational memory usage is greatly reduced.

A rough set based data mining approach for house of quality analysis

As the first phase of quality function deployment (QFD) and the only interface between the customers and product development team, house of quality (HOQ) plays the most important role in developing quality products that are able to satisfy customer needs. No matter in what shape or form HOQ can be built, the key to this process is to find out the hidden relationship between customers’ requirements and product design specifications. This paper presents a general rough set based data mining approach for HOQ analysis. It utilises the historical information of customer needs and the design specifications of the product that was purchased, employs the basic rough set notions to reveal the interrelationships between customer needs and design specifications automatically. Due to the data reduction nature of the approach, a minimal set of customer needs that are crucial for the decision on the correlated design specifications is derived. The end result of the approach is in the form of a minimal rule set, which not only fulfils the goal of HOQ, but can be used as supporting data for marketing purposes. A case study on the product of electrically powered bicycles is included to illustrate the approach and its efficiency.

An integrated disassembly constraint generation approach for product design evaluation

Computer-aided design (CAD) has been a de facto industry standard approach for common product design. Analysing the manufacturing factors of the product at its design stage can greatly reduce its cost at the later stages of product life cycle. This article presents an integrated approach of disassembly constraint generation, based on which an object-oriented prototype is designed and developed. The approach is implemented as an embedded application on commercial CAD software platform so as to provide a tight and seamless integration to CAD process. It utilises the CAD model information directly and allows designers to evaluate the on-going design for its disassembly properties. Moreover, since the modules and objects are well encapsulated, it is possible for further disassembly planning applications to build on top of or integrate to it. The case study on a widely used shaft assembly demonstrated the effectiveness of the integrated approach.