The Framework Study and Development of a Design Wind Load Expert System for Buildings

碩士 === 淡江大學 === 土木工程學系碩士班 === 93 === It is a tedious task to refer to wind code while performing structural building designs because lots of parameters need to be considered. The decision of their suitable ranges and application conditions can be very confusion sometimes. Moreover, the process of wi...

Full description

Bibliographic Details
Main Authors: Hsaing-Jung Hsu, 許祥榕
Other Authors: 王人牧
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/72779212478402638924
Description
Summary:碩士 === 淡江大學 === 土木工程學系碩士班 === 93 === It is a tedious task to refer to wind code while performing structural building designs because lots of parameters need to be considered. The decision of their suitable ranges and application conditions can be very confusion sometimes. Moreover, the process of wind load calculation is prone to mistakes because of omitting or negligence. The comprehension of wind code and calculation of design wind loads are difficult and time-consuming for designers that have weak wind engineering concept. Therefore, Taiwan''s new wind code and the logic flow of calculation were coded in rules, and a rule-based expert system was developed. Users only need to input the basic building information to get design wind pressures and wind loads as well as the parameters and coefficients used during calculation. The advantage is that mistakes can be reduced to a minimum and engineers can get the design information needed rapidly to reduce the time for wind load analysis and calculation, so as to increase the exactitude and efficiency. The application areas of the system covers the evaluations of design wind loads for structural systems and design wind pressures for claddings. The subjects of building design wind load analysis include high-rise buildings, mid-rise buildings, low-rise buildings and open structures. The cladding wind pressure analysis takes account of exterior walls and roofs. The expert system''s inference strategy is backward chaining, also called goal-driven. Base on what the user wants, the system goes through the rules to uncover and inquire for the information needed. The user interface is built on Internet browsers using mostly JSP, and the knowledge base and inference engine are on a MS IIS server. The final results are listed in tables and various loading curves can be plotted using web-charting software.