A decision approach to mechanism type selection is presented, which employs LVQ neural network as classifier and decision-maker to recognize a satisfactory mechanism from a range of mechanisms achieving a required kinematic function. Through learning from correct samples extracted from different mechanisms, expert knowledge is acquired and expressed in the form of weight matrix by LVQ network. When selecting mechanism type, through digitizing the design requirements, converting into a characteristic factor set, and fed into the trained LVQ network, a satisfactory mechanism can be automatically recognized from a range of mechanisms with the same kinematic function. Under this approach, the problem of knowledge acquisition and expression can be effectively solved, and the rationality of the decision can be improved at some extent. It is verified this approach is feasible to perform mechanism type selection and possesses a better characteristic of pattern classification compared with BP neural network.
[1]
Rui-Feng Bo.
A New Approach to Mechanism Type Selection by Using Back-Propagation Neural Networks
,
2010,
2010 International Conference on Artificial Intelligence and Computational Intelligence.
[2]
Deyi Xue,et al.
Design candidate identification using neural network-based fuzzy reasoning
,
2000
.
[3]
Dongxing Cao.
CONCEPTUAL DESIGN OF MECHANICAL PRODUCT BASED ON FUNCTIONAL DECOMPOSITION
,
2001
.
[4]
Juite Wang,et al.
Ranking engineering design concepts using a fuzzy outranking preference model
,
2001,
Fuzzy Sets Syst..
[5]
Li Tie-nan,et al.
NEURAL NETWORK AND ITS APPLICATION
,
2003
.