摘要
在论述人工神经网络模型的特点和性质的基础上,综述国内机械工程领域研究与应用人工种经网络技术的现状与发展,介绍最常用的模型种类,举例说明这类模型在机械制造和机械设计各环节中的实际应用,并阐述了完善模型结构、改进建模方法方面的基础性研究工作。从内容、深度、实际效果和软硬件环境方面评述当前在人工种经网络模型研究与应用中的不足之处,指出今后努力的方向和潜在的应用领域。
As a new-type structuralized mod-el, artificial neural network (ANN ) stores infor-mation of objects by means of the topologicalstructures. This distributed, holographic infor-mation storage intrusts ANN the properties oferror torlerence, error prevention, associationand parallel operation. Desides, non-linear map-ping and self-learning abilities are also obtainedby the models of ANN. After a description of thenature and the characterristics of ANN models,in this paper the current state and the develop-ments in the application of ANN models to me-chanical engineering are reviewed, some widelyused ANN models are introduced, the practicaluse in all aspects of mechanical design and man-ufacturing are exampled, and the foundamentalresearch work on ANN for optimizing modelstructure and improving modelling methodologyare summarized. Finally, the defects in researchand application of ANN are discussed in the as-pects of contents,depth,effects and the environ-memt of hardware and software, the potentialapplications and the future work are also pro-posed.
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
1997年第2期5-10,共6页
China Mechanical Engineering
关键词
人工神经网络
机械工程
机械制造
ANN mechanical engineer-ing manufacturing theoretical model