摘要:

The computer-based support of conceptual and procedural reasoning and of data manipulation is the central theme of this paper. The human conceptual mode of reasoning is supported by a computer-based thesaurus of concepts, and the procedural mode by a hierarchically organised set of specialists operated on by an inference engine. Support for data manipulation is provided by giving the user access to a set of functions for generating models and determining their behaviour. The resulting computer-based system is termed a knowledge-based support system (KBSS), rather than an expert system, because we emphasise the need for the computer to support the human's powerful reasoning methods. A KBSS has been developed in the domain of control engineering and is available commercially from Oxford Computer Consultants.

DOI:10.1049/ise.1992.0010

出版商:IEE    

年代:1992

数据来源: IET

摘要:

As the deployment of expert systems has spread into more complex and sophisticated environments, so inherent technological limitations have been observed. As a technique for overcoming this complexity barrier, researchers have started to build systems composed of multiple co-operating components. These systems fall into two distinct categories; those that solve a particular problem, such as speech recognition or vehicle monitoring, and those that are general to some extent. GRATE is a general framework, which enables an application builder to construct multi-agent systems for the domain of industrial process control. Unlike other frameworks, GRATE embodies a significant amount of inbuilt knowledge related to co-operation and control which can be utilised during system building. This approach offers a paradigm shift for the construction of multi-agent systems, in which the role of configuring pre-existing knowledge becomes an integral component. Instead of starting afresh, the designer can utilise the in-built knowledge and augment it, if necessary, with domain-specific information. The GRATE architecture has a clear separation of concerns, and has been applied to real-world problems in the domains of electricity transportation management and diagnosis of a particle accelerator beam controller.

DOI:10.1049/ise.1992.0011

出版商:IEE    

年代:1992

数据来源: IET

摘要:

We describe a system called the device modeling environment (DME), which helps the user to analyse the behaviour of the device being designed. Given a description of the device design, the DME formulates a behaviour model of the device, analyses its behaviour through simulation, and provides an explanation of the behaviour. The DME can model both continuous and discontinuous physical phenomena. When it detects that the situation has evolved to a point where the original model is no longer applicable, it is able to modify the model and to continue simulation. We describe the representation of physical knowledge, the model formulation and simulation procedure of the DME. Our goal is to make the DME an interactive environment, which can assist designers even at an early stage of the design process by providing immediate feedback about the implications of design decisions.

DOI:10.1049/ise.1992.0012

出版商:IEE    

年代:1992

数据来源: IET

摘要:

One of the major aims of design theory is to constructexplanatory modelsthat further enhance our understanding of design as a cognitive act. The search for such models raises the question as to whether there are, in fact,laws of designthat are valid across the ‘sciences of the artificial’. In this paper, we present what we believe are two suchqualitativelaws. Theideasembedded in these laws have long been part of the ‘folk theory’ of design. Their main significance as stated here lies in the fact that the laws have been derived systematically from, and are supported by, well knownknowledge levelconstructs; and in the conciseness, unambiguity and testability of their presentation.

DOI:10.1049/ise.1992.0013

出版商:IEE    

年代:1992

数据来源: IET

摘要:

The analysis and design of an intelligent control system, based on the fuzzy set theory and the well known PID control principle, are carried out. The structures offuzzy PID controllersare built up utilising expert heuristics to perform PID control actions. The properties of such controllers are investigated in terms of their non-linear equivalence, stability analysis and robustness examination. The design procedures for this type of controller are formulated using the linguistic phase-plane approach, and realised by means of the expert system and CAD techniques. The simulation of an experimental heating process regulated by a fuzzy PID controller is presented as a practical example to manifest the correctness and usefulness of the proposed analysis and design methods.

DOI:10.1049/ise.1992.0014

出版商:IEE    

年代:1992

数据来源: IET

摘要:

The long lead time for jig and fixture design is a serious bottle-neck in modern manufacturing. Efforts to automate the fixture design process using intelligent CAD systems have not been taken beyond the prototype stage, mainly because of impediments which include the requirement of complex data structures to represent the workpiece and its attributes. Two deficiencies in existing systems for fixture design are addressed; design for a sequence of operations and design with tolerance considerations. Tolerance and process specifications are obtained from the user during an interactive session. A knowledge-based design system, AutoFix II, designs a fixture for the workpiece. The fixture is optimised for a particular datum specification and sequence of operations. It is analysed and presented via the CAD system. AutoFix II is automated to minimise user-induced errors.

DOI:10.1049/ise.1992.0015

出版商:IEE    

年代:1992

数据来源: IET