Search

Based on quantum-inspired evolutionary algorithm (QEA), a novel approach of constructing multi-class least squares wavelet SVM (LS-WSVM) classifiers is presented, regularization parameters and kernel parameters of LS-WSVM can be optimized. Quantum-inspired evolutionary optimazition can get appropriate parameters of LS-WSVM with global search, so the LS-WSVM model for the multi-class classifiers is built. And then, classification is studied using LS-SVM with wavelet kernel and Gaussian kernel. The simulation results show that the approach for the multi-class LS-WSVM classifiers is effective, that can obtain the optimal parameters of LS-WSVM with global searching QEA, and improved LS-WSVM provides excellent precision for classification.

Based on least squares wavelet support vector machines (LS-WSVM) within the Bayesian evidence framework, a systematic method for fault diagnosis of power circuits is presented. In this paper, the Bayesian evidence framework is applied to select the optimal values of the regularization and kernel parameters of multi-class LS-WSVM classifiers. Also wavelet coefficients of output voltage signals of power circuits under faulty conditions are obtained with wavelet lifting decomposition, and then faulty feature vectors are extracted from the disposed wavelet coefficients. The faulty feature vectors are used to train the multi-class LS-WSVM classifiers, so the model of the power circuits fault diagnosis system is built. In push-pull circuits, this method is applied to diagnose the faults of the circuits with simulation; the results show that the fault diagnosis method of the power circuits with LS-WSVM within the Bayesian evidence framework is effective.

A combinational circuit design method that can be completely implemented on a FPGA is presented. For carrying out faster evolution in evolutionary circuit design, a sub population based MOEA (multiobjective evolutionary algorithm) is employed in which the reconfigurable circuit (RC) architecture is encoded by Cartesian Genetic Programming (CGP). For hardware implementation, the Celoxica RC1000 PCI is employed which includes Xilinx Virtex xcv 2000E FPGA chip. This PCI card is communicating with host PC and acting as an evolvable platform. MOEA adopted modules are designed into a FPGA chip for discussing the rationality of circuit design method. Results of direct evolution and results of incremental evolution is compared, it shows MOEA is most efficient in the aspect of speeding up the convergence of evolution.

The energy of node in wireless sensor networks (WSN) is severely constrained, it is very important to maximize the lifetime of the entire network. In order to enhance energy efficiency and prolong the system lifetime, a novel energy saving algorithm based on correlation function is presented. Firstly, the relationship between the energy consumption and the length of communication data is introduced and analysed. According to the geographic information and the threshold of the redundant radius, the mutual supportability of nodes in the redundant area was obtained through the correlation function, then, some non-cluster head nodes with high supportability can be turned off and become candidate active nodes till the next round. Thus, the cluster head can get exact fused data through less non-cluster head nodes. The energy saving algorithm can reduce communication between non-cluster head nodes and the cluster heads. Finally, the reliability of algorithm is tested with 12 sensor nodes, and the efficiency of energy saving is analysed. The results of test and simulation demonstrate that the energy saving algorithm is effective, and the system lifetime is prolonged.