Ziying Pu, Shanghai Jiao Tong University, China
Zhihong Yao, Shanghai Jiao Tong University, China; Shanghai University, China
A new adaptive potential crossover operator, one process of the improved genetic algorithm, is proposed in this paper to overcome the drawbacks of high randomness and slow convergence speed of genetic algorithm. The new crossover operator is based on the reflectance and trans- mittance coefficients of particle penetrating the potential in quantum mechanics. The improved genetic algorithm, which is used in neural network training, includes the new crossover operator and the deterministic crowding mecha- nism. It has been demonstrated by simulation results and the pattern recognition experiment on blue-green algae1 that the approach not only has the properties of high con- vergence speed and good searching ability but also has ef- ficiency in pattern recognition.
Citation:
Ziying Pu, Zhihong Yao, Minrui Fei, Xiurong Yin, Hainan Kong, "Genetic Neural Network Based on Adaptive Potential Crossover Operator and its Application in Pattern Recognition of Blue-green Algae," isda, vol. 1, pp.971-978, Sixth International Conference on Intelligent Systems Design and Applications (ISDA'06) Volume 1, 2006