摘要
本文较系统地介绍了钢筋混凝土蓄水池智能ＣＡＤ──ＷＳＣＡＤ的构成与特点，讨论了ＷＳＣＡＤ中所采用的神经网络法．针对神经网络所存在的训练速度慢、系统结构通用性差等缺点，提出了一种表示蓄水池设计事例的层状神经网络模型．这种网络模型的提出为智能化的钢筋混凝土蓄水池标准图库的建立及其应用奠定了基础．
This paper describes the structure and characteristics of an intelligent CAD system forthe structural design of reinforced concrete pools──WSCAD,and discusses the neural net-work methods adopted in the system, Because of the low learning rate and structural versatil-ity of neural netmowks a layered neural network system expressing the design cases of con-crete pools is presented. Based on the neural network system,an intellingent database of thepool standard working drawings can be developed.
