Artificial Intelligence in Engineering最新文献

This paper presents the software architecture for a next generation concurrent engineering environment that helps geographically separated designers and engineers to collaborate effectively. The paper highlights research in computer-supported collaboration work (CSCW) based on various models of group interaction, social communication theory, negotiation theory and distributed artificial intelligence concepts. The paper describes CAIRO (Collaborative Agent Interaction and synchROnization) system, a distributed conferencing architecture for managing designers and engineers in a distributed design meeting. The CAIRO system allows designers and engineers to work together in virtual teams by supporting multi-media interactions over computer networks. CAIRO aids the concurrent engineering effort by relaxing the physical, temporal and organizational constraints experienced in traditional design meeting environments. CAIRO provides both media synchronization, i.e. ensuring that all information exchanged between users is synchronized, and agent synchronization, i.e. ensuring effective structuring and control of a distributed conference. This paper also details the prototype CAIRO system with a detailed example, illustrating its use in concurrent design settings.

SEED-Config is a design environment intended to assist structural designers in collaboratively exploring and extending the design buildings. Its purpose is to help designers in rapidly synthesizing alternative structural design solutions at the conceptual level to the point where the structural concept is supportive of the overall design and possible conflicts are explored and resolved. The SEED-Config prototype consists of four subsystems. The Design Information Repository is built upon an information model that exploits the hierarchical nature of building descriptions to represent design information. It can save an overall design or parts of a design to a case library, thus supporting case-based reasoning. The Design Knowledge Manager allows for the browsing, editing, selecting and applying of technology nodes which encapsulate structural design knowledge. The Classification Reference Manager allows for the definition, management, inference and querying of classifications which are used to classify and index design solutions as they are generated. The Geometric Modeler reasons about topology and geometry.

When constraints are used to represent engineering requirements, enhanced support for collaboration becomes possible. More specifically, if important engineering design specifications are represented as groups of inequalities on continuous variables, solving these constraints results in spaces of feasible values. Such spaces improve efficiency through avoiding artificial conflicts, improving design flexibility, enhancing change management and assisting conflict resolution. This paper describes an implementation that employs algebraic reformulation, which includes an efficient approach for transformation to ternary expressions. This approach reduces limitations related to computational complexity that were inherent in previous implementations. Important features are then illustrated using a full-scale example. Carrying out collaborative design using solution spaces (CDSS) with this new implementation called SpaceSolver. Such software is expected to enhance future collaboration tools.

In concurrent design, there is a need to control the interaction between the participants and also to manage the compatibility of the various components being designed. This requires a carefully defined structure to control the changes and additions to the design of artefacts and a means of allowing all participants to see as much of the design information as possible. For situations where colocation is not possible, this is best achieved through a suitably designed computer system. This paper covers the development of the Describe environment which consists of a CAD system linked to an Object-Oriented Database (OODB) via AutoLISP and a C++ executable. The interaction model within Describe has been developed to be as simple and flexible as possible and is based on fresults from a design experiment. Within the system, design information is stored as complex, wrapped objects which allow the seamless integration of graphical and other design information. The resulting system automatically updates design calculations when dimensions are altered in the CAD system. The system has been developed in conjunction with practising designers and applied to the domain of beam/slab bridges.

Construction projects usually involve transient ‘virtual organisations’ made up of members of a project team (involving several disparate disciplines) working together on the design and construction of a facility. Team members are often non-co-located, particularly at the early stages of the design process, and tend to work independently while taking decisions that affect others. The adoption of concurrent engineering principles by the construction industry is increasingly being seen as vital for reducing the problems posed by the industry's fragmentation, and enhancing its competitiveness. An important aspect of concurrent engineering in construction is the need for an effective communications infrastructure able to transmit project information between members of the project team and across all stages in the constructed facility's lifecycle. This paper describes the development of such a communications infrastructure that is based on the concept of Telepresence. The intention is to create a persistent space to support interaction between project personnel throughout the design and construction phases of projects. The paper first highlights the key communications issues that need to be addressed, introduces ‘Telepresence’ and describes an initial prototype system. The approach being adopted in the development of an advanced Telepresence environment for construction project teams is also presented. The Telepresence environment is intended to help people who cannot be together to work together.

In this paper, a competitive neural network architecture is used as an intelligent fault diagnosis system to detect the fault sources in different subsystems or elements of a plant or any other device. The prioritized modification rule for connection weights is introduced and four different procedures are studied and compared from the viewpoint of their efficiency. It is shown that the fourth procedure is more convenient for human type decision-making. The output functions of different neurons are considered as the possibility of being fault sources for different units. The system starts from a vague initial state and the connection weights are modified during the learning procedures. The simulation results of different strategies are analyzed and compared. A typical CNC machine is considered as a case study.

This paper addresses a marked variation of the classical trim-loss problem. This problem has been dealt with a rule-based approach due to its inherent complexity in the case under consideration. A rule-based model was evolved as the basis for the expert system development. The paper emphasises on the practical application of a rule-based model for inventory issue procedure in a construction company. Due to the peculiar nature of the problem, issuing procedures for steel pipes are quite different from the normal inventory issue practices. Inventory Management procedures which are guided by historic practices, rules of thumb and mathematical models were found to be unsuitable for the case under study and hence motivates the use of a “rule-based system” that has evolved from the “expert” knowledge of inventory personnel. The paper provides some of the implementation modules and highlights the benefits of the rule-based system for the construction company under study.

This paper considers nonlinearly constrained tolerance allocation problems in which both tolerance and process selection are to be selected simultaneously so as to minimize the manufacturing cost. The tolerance allocation problem has been studied in the literature for decades, usually using mathematical programming or heuristic optimization approaches. The difficulties encountered for both methodologies are the number of constraints and the difficulty of satisfying the constraints. A penalty-guided genetic algorithm is presented for solving such mixed-integer tolerance allocation problems. It can efficiently and effectively search over promising feasible and infeasible regions to find the feasible optimal or near optimal solution. Genetic results are compared with the results obtained from 12 problems from the literature that dominate the previously mentioned solution techniques. Numerical examples indicate that the genetic algorithms perform well for the tolerance allocation problem considered in this paper. In particular, as reported, solutions obtained by genetic algorithms are as well as or better than the previously best-known solutions.

Stream sampling is essential for the performance assessment of a mineral processing plant. This process generates errors that are caused by stream material heterogeneity and incorrect cutter features. To control the sampling process efficiently, it is very important to evaluate and minimize the sampling errors. The objective of this paper is to introduce two expert systems for stream sampling in mineral processing plants. The first one is intended to inspect the correctness of sampling operations. It is called Sampling Correctness Inspector (SCI). The second one is destined for the evaluation of sampling errors. It is named Sampling Error Evaluator (SEE). These expert systems are validated successfully.