Owing to the original Ott-Grebogi-Yorke(OGY) method can only be applied to the discrete dynamical systems or continuous dynamical systems which can be described by Poincare mapping, a novel method for controlling chaos is proposed by resorting to echo state network. The network was trained by the input and output sample which were generated by OGY method. The chaos controller can drive the chaotic attractor embedded in the unstable orbit to the unstable fixed point. The simulation results of Henon mapping verify that neural network controller can drive the chaos system to unstable fixed point in short time. The neural network controller greatly speeds up the stabilization process of controlling chaos system.
Stacked autoencoder is a typical deep neural network. The hidden layers will compress the input data with a better representation than the raw data. Stacked autoencoder has several hidden layers. However, the number of hidden layers is always experiential. In this paper, different hidden layers number autoencoders are discussed. Different depths of stacked autoencoder have different learning capability. The deeper stacked autoencoders have better learning capability which needs more training iterations and time.
For the re-evolution of the mobile robot behavior in unknown environments, the mapping relation was constructed between input of sensors and output of actuators based on echo state network. An algorithm of adaptive behavior learning was presented based on echo state network for evolutionary robotics. The composite architecture with responsive behavior and behavior learning was adopted. The responsive behavior was drived by the samples composed with sensor information and decision. The weights of echo state network were optimized via (μ+λ)-evolution strategy. The new control rules were generated via evolutionary algorithms, and new samples were added to the database constantly. The high intelligent behaviors of robot were transmitted to responsive behaviors. The experimental results indicate that the proposed approach has a better adaptability.
Utilization of hydrogen in the intermediate temperature solid oxide fuel cell (IT-SOFC) and micro gas turbine (MGT) hybrid system has the advantages of low costs, non-pollution, and wide range of sources. In this paper, the detailed model of SOFC/MGT hybrid system fueled by hydrogen was implemented to investigate the system operating performance as well as the influence of fuel inlet temperature and flow rate. At the beginning, the framework and functions of the SOFC/MGT hybrid system were introduced, and the topping cycle was arrested. Then, the numerical analysis was performed by using MATLAB software. The result of the simulation was generated to explain the performance and the electrical efficiency which can reach up to 75.61% at the design point. With an increasing fuel inlet temperature, both the output power and efficiency of the SOFC, MGT and hybrid system rose slightly. The rising fuel flow rate increased the output power of SOFC, MGT and the whole system, but made the system efficiency decrease.
Inspired by multiple information processing mechanisms of the human nervous system, a fusion simplified pulse coupled neural network (FSPCNN) model for infrared (IR) image segmentation is proposed in this paper. In the method based on FSPCNN, the time decay factor is set adaptively based on Stevens’ power law, and the synaptic weight is generated adaptively based on lateral inhibition (LI), without manual intervention. Meanwhile, according to fast linking mechanism, the similarity between adjacent iteration results is used to implement the automatic selection of optimal segmentation result and control iteration. Experimental results indicate that the proposed method has favorable robustness and segmentation performance.
In this paper, we propose an improved vehicle re-identification method based on the combination between the AlignedRelD and the Stochastic Weight Averaging (SWA). AlignedRelD extracts a global feature and local features of a vehicle's image and performs joint learning. Local automatic alignment is achieved by computing the shortest path between the two sets of local features, so that global feature learning can benefit from local feature learning. By running an optimizer with a high constant learning rate, the SWA averages the weight of the model to ensure that a better weight combination can be found. Our method achieves rank-1 accuracy of 94.4% on VeRi-776 and 95.1% on VehiclelD(small), outperforming state-of-the-art methods by a large margin. In order to better solve the task of vehicle re-identification in residential area, we have made the Oeasy-Parking dataset and experimented with our methods, and achieved good results.
As an important part of emotion research, facial expression recognition is a necessary condition for intelligent interaction between human and machine, which has an important research significance and a potential commercial value. Convolutional neural network (CNN) is an effective method to recognize facial emotions, which can perform feature extraction and classification simultaneously, and can automatically discover multiple levels of representations in data. Due to the fact that there are millions of parameters involved in training the convolutional neural network model, and there is a large demand for marked samples, transfer learning is often used to fine-tune the pre-trained model for a small target sample set. However, there are often some content differences between data sets during deep transfer learning, which will affect the recognition ability of feature extraction. In order to improve the facial expression recognition ability in transfer learning, a hybrid transfer learning model based on an improved convolution restricted boltzmann machine (CRBM) model and a CNN model is proposed in this paper. This method is fused by two learning abilities of these two models. When the pre-trained CNN model is transferred to a small target set, the CRBM is used to replace the full connection layer in the CNN model, and the CRBM layer and the sofmax layer will be retrained on the target set. The added CRBM layer can not only fully connects all feature maps, but can also learn about the unique statistical characteristics about the target set, which eliminates the influence of content differences between data sets, and extracts higher-order statistical features of facial expression images from the target set. The proposed method is evaluated based on four publicly available facial expression databases: JAFFE, FER2013, SFEW and RAF-DB. The new method can achieve better performance than most state-of-the-art methods, and it can effectively prevent the negative influence of transfer learning features between different data sets.
The paper presents the application of a fuzzy logic controlled genetic algorithm (FCGA) to environmental/economic dispatch. The authors first propose an improved genetic algorithm with two fuzzy controllers based on some heuristics to adaptively adjust the crossover probability and mutation rate during the optimisation process. The implementation of the fuzzy crossover and mutation controllers is described. The proposed FCGA can be applied to a wide range of optimisation problems. The validity of the proposed algorithm is illustrated on environmental/economic dispatch of a six-generator system. Its performance is compared with conventional GA and the Newton–Raphson method. The results are very encouraging.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)