摘要:

This article provides a review of two main research projects carried out at the French Scientific and Technical Center for Building (CSTB) in the area of artificial intelligence applied to building engineering. We first describe the software engineering techniques used to develop the computer environments. Then we present the architecture of an advanced multiservice computer system offering capabilities comprising retrieval facilities, hypertext editing of documents (e.g., unified codes of practice, regulations), object‐oriented database management, and multifunctional expert systems. Finally we discuss a major multiactor computeraided design project aiming to provide a framework able to combine the sundry contributions of the various actors involved in the building design process. The expected results of such projects are a panel of complementary capabilities either aiming at satisfying multiservice requirements for an isolated designer or at providing a single but cooperative and computer‐assisted process to different actors. Of course, merging the respective capabilities of these tools is a very promising perspect

ISSN:1093-9687    

DOI:10.1111/j.1467-8667.1990.tb00376.x

出版商:Blackwell Publishing Ltd    

年代:1990

数据来源: WILEY

摘要:

In the past few years literature on computational civil engineering has concentrated primarily on artificial intelligence (Al) applications involving expert system technology. This article discusses a different Al approach involving neural networks. Unlike their expert system counterparts, neural networks can be trained based on observed information. These systems exhibit a learning and memory capability similar to that of the human brain, a fact due to their simplified modeling of the brain's biological function. This article presents an introduction to neural network technology as it applies to structural engineering applications. Differing network types are discussed. A back‐propagation learning algorithm is presented. The article concludes with a demonstration of the potential of the neural network approach. The demonstration involves three structural engineering problems. The first problem involves pattern recognition; the second, a simple concrete beam design; and the third, a rectangular plate analysis. The pattern recognition problem demonstrates a solution which would otherwise be difficult to code in a conventional program. The concrete beam problem indicates that typical design decisions can be made by neural networks. The last problem demonstrates that numerically complex solutions can be estimated almost instantaneously with a neural networ

ISSN:1093-9687    

DOI:10.1111/j.1467-8667.1990.tb00377.x

出版商:Blackwell Publishing Ltd    

年代:1990

数据来源: WILEY

摘要:

The widespread acceptance of cold‐formed structural steel members as roof purlins in conventionally engineered building has been hindered by the lack of a useful and efficient analysis and design methodology. This article outlines a microcomputer‐based design methodology for roof Zpurlins, the purpose of which is to provide such a tool. The design methodology details used in the development of the interactive microcomputer program Z‐BM are described herein, along with a general program description and a sample program execution. Z‐BM is based on the American Iron and Steel Institute specifications and considers the combined effect of the Z‐purlin and corrugated roof deck in resisting applied loads. In addition to echoing all input data, for each loading combination, Z‐BM provides a complete stress summary, the maximum allowable axial load, plus the maximum deflection and i

ISSN:1093-9687    

DOI:10.1111/j.1467-8667.1990.tb00378.x

出版商:Blackwell Publishing Ltd    

年代:1990

数据来源: WILEY

摘要:

The paper presents an efficient method of flexibly automating test control and data acquisition functions for servo‐controlled testing. A systems analysis approach coupled with structured design is used in developing the system. With hardware that includes (a) a personal computer, (b) a servo‐controlled electrohydraulic testing machine equipped with a special type of digital function generator, and (c) a multipurpose data acquisition board, an automated test control and data acquisition system for servo‐controlled materials/structural/mechanical testing has been developed. The servo‐controlled materials/structural/mechanical testing system allows (1) automated control of the test, (2) rapid and reliable acqusition of several channels of data from static or fast‐rate tests subject to periodic or nonperiodic loading, and (3) realtime (or nearly real‐time) graphics of the test characteristics in order to allow the operator to monitor the test. The communication and postprocessing options allow user‐supplied software to be incorporated into the automated system for advanced non‐real‐time functions. These options include noise filtering, signal averaging, mathemtical analyses and processing, report quality graphics, and file transfer features. The system features have been validated through a series of load‐controlled direct tension tests on an aluminum specimen. The system has been used for successfully carrying out an elaborate research program on fracture toughness testing of fiber‐reinforced concrete in a flexural

ISSN:1093-9687    

DOI:10.1111/j.1467-8667.1990.tb00379.x

出版商:Blackwell Publishing Ltd    

年代:1990

数据来源: WILEY

摘要:

A central part of the construction of an expert system is the way in which the knowledge is represented. A computer can only store and manipulate knowledge in discrete chunks representing approximations to the real world, but the human mind can work on an infinitely variable sliding scale representing real‐world values. The greater the computerized discretization, the more precise the computer model becomes. The knowledge base within a computer expert system can take many forms depending on the type of domain problem, the language of execution and the nature of the system.The GWX expert system gives advice on groundwater control in excavations for construction projects. The system was written in Prolog and the knowledge base was created by breaking information down into very small pieces, each referring to a segment of data—fact segment—about precise details of the domain. The way in which fact segments have been devised for the GWX expert system is desc

ISSN:1093-9687    

DOI:10.1111/j.1467-8667.1990.tb00380.x

出版商:Blackwell Publishing Ltd    

年代:1990

数据来源: WILEY