In order to analyze the influence rule of experimental parameters on the energy-absorption characteristics and effectively forecast energy-absorption characteristic of thin-walled structure, the forecast model of GA-B...In order to analyze the influence rule of experimental parameters on the energy-absorption characteristics and effectively forecast energy-absorption characteristic of thin-walled structure, the forecast model of GA-BP hybrid algorithm was presented by uniting respective applicability of back-propagation artificial neural network （BP-ANN） and genetic algorithm （GA）. The detailed process was as follows. Firstly, the GA trained the best weights and thresholds as the initial values of BP-ANN to initialize the neural network. Then, the BP-ANN after initialization was trained until the errors converged to the required precision. Finally, the network model, which met the requirements after being examined by the test samples, was applied to energy-absorption forecast of thin-walled cylindrical structure impacting. After example analysis, the GA-BP network model was trained until getting the desired network error only by 46 steps, while the single BP-ANN model achieved the same network error by 992 steps, which obviously shows that the GA-BP hybrid algorithm has faster convergence rate. The average relative forecast error （ARE） of the SEA predictive results obtained by GA-BP hybrid algorithm is 1.543%, while the ARE of the SEA predictive results obtained by BP-ANN is 2.950%, which clearly indicates that the forecast precision of the GA-BP hybrid algorithm is higher than that of the BP-ANN.展开更多
Based on the work of previous researchers, a new unbiased optimization algorithm—the dynamic lattice searching method with two-phase local search and interior operation (DLS-TPIO)—is proposed in this paper. This alg...Based on the work of previous researchers, a new unbiased optimization algorithm—the dynamic lattice searching method with two-phase local search and interior operation (DLS-TPIO)—is proposed in this paper. This algorithm is applied to the optimization of Lennard-Jones (LJ) clusters with N=2–650, 660, and 665–680. For each case, the putative global minimum reported in the Cambridge Cluster Database (CCD) is successfully found. Furthermore, for LJ533 and LJ536, the potential energies obtained in this study are superior to the previous best results. In DLS-TPIO, a combination of the interior operation, two-phase local search method and dynamic lattice searching method is adopted. At the initial stage of the optimization, the interior operation reduces the energy of the cluster, and gradually makes the configuration ordered by moving some surface atoms with high potential energy to the interior of the cluster. Meanwhile, the two-phase local search method guides the search to the more promising region of the configuration space. In this way the success rate of the algorithm is significantly increased. At the final stage of the optimization, in order to decrease energy of the cluster further, the positions of surface atoms are further optimized by using the dynamic lattice searching method. In addition, a simple new method to identify the central atom of icosahedral configurations is also presented. DLS-TPIO has higher computing speed and success rates than some well-known unbiased optimization methods in the literature.展开更多
基金Project(50175110) supported by the National Natural Science Foundation of ChinaProject(2009bsxt019) supported by the Graduate Degree Thesis Innovation Foundation of Central South University, China
文摘In order to analyze the influence rule of experimental parameters on the energy-absorption characteristics and effectively forecast energy-absorption characteristic of thin-walled structure, the forecast model of GA-BP hybrid algorithm was presented by uniting respective applicability of back-propagation artificial neural network （BP-ANN） and genetic algorithm （GA）. The detailed process was as follows. Firstly, the GA trained the best weights and thresholds as the initial values of BP-ANN to initialize the neural network. Then, the BP-ANN after initialization was trained until the errors converged to the required precision. Finally, the network model, which met the requirements after being examined by the test samples, was applied to energy-absorption forecast of thin-walled cylindrical structure impacting. After example analysis, the GA-BP network model was trained until getting the desired network error only by 46 steps, while the single BP-ANN model achieved the same network error by 992 steps, which obviously shows that the GA-BP hybrid algorithm has faster convergence rate. The average relative forecast error （ARE） of the SEA predictive results obtained by GA-BP hybrid algorithm is 1.543%, while the ARE of the SEA predictive results obtained by BP-ANN is 2.950%, which clearly indicates that the forecast precision of the GA-BP hybrid algorithm is higher than that of the BP-ANN.
基金supported by the National Natural Science Foundation of China (60773194, 61070235)
文摘Based on the work of previous researchers, a new unbiased optimization algorithm—the dynamic lattice searching method with two-phase local search and interior operation (DLS-TPIO)—is proposed in this paper. This algorithm is applied to the optimization of Lennard-Jones (LJ) clusters with N=2–650, 660, and 665–680. For each case, the putative global minimum reported in the Cambridge Cluster Database (CCD) is successfully found. Furthermore, for LJ533 and LJ536, the potential energies obtained in this study are superior to the previous best results. In DLS-TPIO, a combination of the interior operation, two-phase local search method and dynamic lattice searching method is adopted. At the initial stage of the optimization, the interior operation reduces the energy of the cluster, and gradually makes the configuration ordered by moving some surface atoms with high potential energy to the interior of the cluster. Meanwhile, the two-phase local search method guides the search to the more promising region of the configuration space. In this way the success rate of the algorithm is significantly increased. At the final stage of the optimization, in order to decrease energy of the cluster further, the positions of surface atoms are further optimized by using the dynamic lattice searching method. In addition, a simple new method to identify the central atom of icosahedral configurations is also presented. DLS-TPIO has higher computing speed and success rates than some well-known unbiased optimization methods in the literature.
