An improved hybrid immune algorithm for mechanism kinematic chain isomorphism identification in intelligent design

In intelligent mechanism design, isomorphism identification of mechanism kinematic chains (IIMKC) is aimed at avoiding repeated mechanism design and is proved to be an NP-complete problem. In this paper, kinematic chains are represented by graphs. An improved hybrid immune algorithm, which integrates the clonal selection immune algorithm with genetic algorithm and the local search algorithm, is proposed to solve IIMKC problem. Moreover, the novel saving and updating operator is proposed to save the best antibodies and maintain a diverse repertoire of antibodies for improving performance of clonal selection. In addition, the pseudo-crossover operator is introduced to enhance the efficiency of genetic algorithm. Simulation results validate the high efficiency and robustness of the hybrid immune algorithm.

[1]  Yang Ping,et al.  Approach on complex neural-genetic algorithm modeling for isomorphism identification in conceptual design of mechanism , 2009, Comput. Syst. Sci. Eng..

[2]  Q. Li,et al.  An artificial neural network approach to mechanism kinematic chain isomorphism identification , 1999 .

[3]  Helena Szczerbicka,et al.  ARTIFICIAL IMMUNE SYSTEMS IN THE CONTEXT OF MISBEHAVIOR DETECTION , 2008, Cybern. Syst..

[4]  J. P. Cubillo,et al.  Comments on mechanism kinematic chain isomorphism identification using adjacent matrices , 2005 .

[5]  Ping Yang,et al.  A high-performance approach on mechanism isomorphism identification based on an adaptive hybrid genetic algorithm for digital intelligent manufacturing , 2009, Engineering with Computers.

[6]  Q. Li,et al.  Some further development on the eigensystem approach for graph isomorphism detection , 2005, J. Frankl. Inst..

[7]  Yang Ping,et al.  A mixed isomorphism approach for kinematic structure enumeration graphs based on intelligent design and manufacturing , 2007 .

[8]  Enrique Mérida Casermeiro,et al.  Improving Neural Networks for Mechanism Kinematic Chain Isomorphism Identification , 2007, Neural Processing Letters.

[9]  Chee Kheong Siew,et al.  Universal Approximation using Incremental Constructive Feedforward Networks with Random Hidden Nodes , 2006, IEEE Transactions on Neural Networks.

[11]  Lung-Wen Tsai,et al.  Mechanism Design: Enumeration of Kinematic Structures According to Function , 2001 .

[12]  Yuhu Yang,et al.  A new method to mechanism kinematic chain isomorphism identification , 2002 .

[13]  T. S. Mruthyunjaya,et al.  Kinematic structure of mechanisms revisited , 2003 .

[14]  Bing Yang,et al.  Isomorphism identification for epicyclic gear mechanism based on mapping property and ant algorithm , 2006, Engineering with Computers.

[15]  Ping Yang,et al.  A fast algorithm for kinematic chain isomorphism identification based on dividing and matching vertices , 2014 .

[16]  Zhen Huang,et al.  A new theory for the topological structure analysis of kinematic chains and its applications , 2007 .