Rough Set Based Fuzzy Neural Network for Pattern Classification

A rough set based fuzzy neural network algorithm is proposed to solve the problem of pattern recognition. The least square algorithm (LSA) is used in the learning process of fuzzy neural network to obtain the performance of global convergence. In addition, the numbers of rules and the initial weights and structure of fuzzy neural networks are difficult to determine. Here rough sets are introduced to decide the numbers of rules and original weights. Finally, experiment results show the algorithm may get better effect than the BP algorithm.     

Mass concrete strength assessment method by Sonreb and Core combined method using artificial neural network

The Sonreb and Core (SRC) combined method is proposed to assess the concrete compression strength of mass concrete structures.Artificial neural network is employed together with the SRC combined method to obtain the optimal core number.The artificial neural network is trained based on data from different testing methods.The procedure of using artificial neural network to assess the concrete strength is described.It proves that the SRC combined method is superior in many aspects and artificial the presented neural network has a high efficiency and reliability.The combined method using artificial intelligence is promising in the strength assessment of mass concrete structures such as the dam,the anchor of the suspension bridge,etc.     

Walking Stability Control Method for Biped Robot on Uneven Ground Based on Deep Q-Network

A gait control method for a biped robot based on the deep Q-network (DQN) algorithm is proposed to enhance the stability of walking on uneven ground. This control strategy is an intelligent learning method of posture adjustment. A robot is taken as an agent and trained to walk steadily on an uneven surface with obstacles, using a simple reward function based on forward progress. The reward-punishment (RP) mechanism of the DQN algorithm is established after obtaining the offline gait which was generated in advance foot trajectory planning. Instead of implementing a complex dynamic model, the proposed method enables the biped robot to learn to adjust its posture on the uneven ground and ensures walking stability. The performance and effectiveness of the proposed algorithm was validated in the V-REP simulation environment. The results demonstrate that the biped robot''s lateral tile angle is less than 3° after implementing the proposed method and the walking stability is obviously improved.     

Color Reproduction on CRT Displays via BP Neural Networks Under Office Environment

A CRT characterization method based on color appearance matching is presented. A matching between Munsell color chips and CRT charts was obtained in vision perceiver in typical office environment and viewing condition by recommending. And neural networks were utilized to accomplish the color space conversion from CIE standard color space to CRT device color space. The neural networks related the color space conversion and color reproduction of soft/hard-copy directly to the influence of the illuminance and viewing condition in vision perceiver. The average color difference of training samples is 3.06 and that of testing samples is 5.17. The experiment results indicated that the neural networks can satisfy the requirements for the color appearance of hard-copy reproduction in CRT.     

Method of Dynamic Knowledge Representation and Learning Based on Fuzzy Petri Nets

A method of knowledge representation and learning based on fuzzy Petri nets was designed.In this way the parameters of weights,threshold value and certainty factor in knowledge model can be adjusted dynamically.The advantages of knowledge representation based on production rules and neural networks were integrated into this method.Just as production knowledge representation,this method has clear structure and specific parameters meaning.In addition,it has learning and parallel reasoning ability as neural networks knowledge representation does.The result of simulation shows that the learning algorithm can converge,and the parameters of weights,threshold value and certainty factor can reach the ideal level after training.     

Lung Nodule Image Retrieval Based on Convolutional Neural Networks and Hashing

Lung medical image retrieval based on content similarity plays an important role in computer-aided diagnosis of lung cancer. In recent years, binary hashing has become a hot topic in this field due to its compressed storage and fast query speed. Traditional hashing methods often rely on high-dimensional features based hand-crafted methods, which might not be optimally compatible with lung nodule images. Also, different hashing bits contribute to the image retrieval differently, and therefore treating the hashing bits equally affects the retrieval accuracy. Hence, an image retrieval method of lung nodule images is proposed with the basis on convolutional neural networks and hashing. First, a pre-trained and fine-tuned convolutional neural network is employed to learn multi-level semantic features of the lung nodules. Principal components analysis is utilized to remove redundant information and preserve informative semantic features of the lung nodules. Second, the proposed method relies on nine sign labels of lung nodules for the training set, and the semantic feature is combined to construct hashing functions. Finally, returned lung nodule images can be easily ranked with the query-adaptive search method based on weighted Hamming distance. Extensive experiments and evaluations on the dataset demonstrate that the proposed method can significantly improve the expression ability of lung nodule images, which further validates the effectiveness of the proposed method.     

Sub-pixel mapping method based on BP neural network

A new sub-pixel mapping method based on BP neural network is proposed in order to determine the spatial distribution of class components in each mixed pixel.The network was used to train a model that describes the relationship between spatial distribution of target components in mixed pixel and its neighboring information.Then the sub-pixel scaled target could be predicted by the trained model.In order to improve the performance of BP network,BP learning algorithm with momentum was employed.The experiments were conducted both on synthetic images and on hyperspectral imagery(HSI).The results prove that this method is capable of estimating land covers fairly accurately and has a great superiority over some other sub-pixel mapping methods in terms of computational complexity.     

Modeling and Simulation of Time Series Prediction Based on Dynamic Neural Network

Molding and simulation of time series prediction based on dynamic neural network(NN) are studied.Prediction model for non-linear and time-varying system is proposed based on dynamic Jordan NN. Aiming at the intrinsic defects of back-propagation (BP) algorithm that cannot update network weights incrementally, a hybrid algorithm combining the temporal difference (TD) method with BP algorithm to train Jordan NN is put forward.The proposed method is applied to predict the ash content of clean coal in jigging production real-time and multistep. A practical example is also given and its application results indicate that the method has better performance than others and also offers a beneficial reference to the prediction of nonlinear time series.     

Improved BP Neural Network for Transformer Fault Diagnosis

The back propagation （BP）-based artificial neural nets （ANN） can identify complicated relationships among dissolved gas contents in transformer oil and corresponding fault types, using the highly nonlinear mapping nature of the neural nets. An efficient BP-ALM （BP with Adaptive Learning Rate and Momentum coefficient） algorithm is proposed to reduce the training time and avoid being trapped into local minima, where the learning rate and the momentum coefficient are altered at iterations. We developed a system of transformer fault diagnosis based on Dissolved Gases Analysis （DGA） with a BP-ALM algorithm. Training patterns were selected from the results of a Refined Three-Ratio method （RTR）. Test results show that the system has a better ability of quick learning and global convergence than other methods and a superior performance in fault diagnosis compared to convectional BP-based neural networks and RTR.     

Transient Air-Fuel Ratio Control in a CNG Engine Using Fuzzy Neural Networks

The fuzzy neural networks has been used as means of precisely controlling the air-fuel ratio of a leanburn compressed natural gas (CNG) engine. A control algorithm, without based on engine model, has been utilized to construct a feedforward/feedback control scheme to regulate the air-fuel ratio. Using fuzzy neural networks, a fuzzy neural hybrid controller is obtained based on PI controller. The new controller, which can adjust parameters online, has been tested in transient air-fuel ratio control of a CNG engine.     

Inverse Control of Nonlinear Servo System Based on Neural Networks

ＩｎｖｅｒｓｅＣｏｎｔｒｏｌｏｆＮｏｎｌｉｎｅａｒＳｅｒｖｏＳｙｓｔｅｍＢａｓｅｄｏｎＮｅｕｒａｌＮｅｔｗｏｒｋｓＷＡＮＧＣｈａｎｇｈｏｎｇＸＵＬｉｘｉｎＧＡＯＸｉａｏｚｈｉＺＨＵＡＮＧＸｉａｎｙｉ（王常虹）（徐立新）（高晓智）（庄显义）（Ｄｅｐｔ．...     

非线性连续时间神经网络的熵测度理论

能量函数是神经网络的基本测度函数,但能量最小化原则并不是普遍适用于各种神经网络模型。从本质上讲,神经网络的智能信息处理过程就是系统不确定性减小的过程。基于上述思想,文中深入地研究并揭示了神经网络模型及其能量函数的动态机理和系统熵流之间的联系,建立了连续时间神经网络熵测度理论与方法以及熵学习算法,提出了了基于熵测度的神经优化方法。     

Adaptive learning with guaranteed stability for discrete-time recurrent neural networks

To avoid unstable learning, a stable adaptive learning algorithm was proposed for discrete-time recurrent neural networks. Unlike the dynamic gradient methods, such as the backpropagation through time and the real time recurrent learning, the weights of the recurrent neural networks were updated online in terms of Lyapunov stability theory in the proposed learning algorithm, so the learning stability was guaranteed. With the inversion of the activation function of the recurrent neural networks, the proposed learning algorithm can be easily implemented for solving varying nonlinear adaptive learning problems and fast convergence of the adaptive learning process can be achieved. Simulation experiments in pattern recognition show that only 5 iterations are needed for the storage of a 15×15 binary image pattern and only 9 iterations are needed for the perfect realization of an analog vector by an equilibrium state with the proposed learning algorithm.     

模糊神经网络在机载相机稳像中的应用

针对解决微型飞行器空中拍摄的图像抖动问题,采用自组织递归区间二型模糊神经网络的函数逼近及泛化能力对微型飞行器上的相机振动规律进行模拟,预测机载相机的振动矢量.该自组织递归区间二型模糊神经网络的初始规则数为零,所有规则都是通过结构和参数同时在线学习来产生,网络结构学习采用的是在线区间二型模糊群集,提高自组织递归区间二型模糊神经网络的稳定性及计算精度.仿真结果表明:将自组织递归区间二型模糊神经网络与双BP神经网络进行对比,利用自组织递归区间二型模糊神经网络对微型飞行器相机振动矢量进行预测的精度高.     

基于鲁棒稳定高阶动态神经网络的非线性系统的辨识

将高阶动态神经网络作为非线性系统的辨识模型,运用Lyapunov稳定性理论,提出一种有效的鲁棒稳定学习规则及相应的学习网络结构,从而确保在对非线性系统辨识时,即使存在建模偏差,辨识误差和动态神经网络的参数能一致最终有界(UUB)稳定,解决了动态神经网络的学习稳定性问题．仿真结果也证明了该辨识方法的有效性．     

基于BP神经网络整定的PID温度控制

展示了一种基于BP神经网络的PID控制器,利用神经网络的自学习特性,将神经网络与PID控制方法相结合,采用3层前向网络,动态BP算法,实现对温度控制系统的在线智能控制,显示了BP神经网络PID控制方法很强的鲁棒性,同时也显示了神经网络在解决高度非线性和严重不确定系统方面的能力．仿真结果表明,此种PID在温度控制中能够取得较满意的效果．     

悉尼自适应交通控制系统线圈数据短时多步预测双层模型

为了进一步改善悉尼自适应交通控制系统(Sydney coordinated adaptive traffic system,SCATS)线圈数据短时多步预测的效果,在对SCATS线圈数据进行预处理的基础上,设计了一种基于动态神经网络的短时多步预测双层模型,包括基于NARX(Nonlinear autoregressive model with exogenous inputs)神经网络的多步预测方法以及基于FTD(Focused time-delay)神经网络的可预测步数在线估计方法,并采用某特大城市SCATS线圈实测数据进行了验证和对比分析。结果表明:本文方法能够进一步降低SCATS线圈数据短时多步预测的误差。     

在线时间序列预测方法及其应用

为解决非线性复杂时间序列在线预测问题,提出了一种基于过程神经网络模型的在线预测方法.首先,在历史数据的基础上建立双并联离散过程神经网络模型;然后,根据在线更新的数据样本,采用递推极限学习算法对过程神经网络隐层到输出层的权值进行相应的更新;最后,应用权值更新后的过程神经网络模型对时间序列进行预测.文中给出了具体的过程神经网络学习算法与权值更新机制,并以混沌时间序列与液体火箭发动机的状态预测为例对方法进行了验证.研究结果表明:该方法在预测精度和适应能力上较单一的离线模型有显著提高,可以为非线性复杂时间序列在线预测问题提供一种有效的解决方法.     

基于OS-ELM的光伏发电中长期功率预测

为了进一步提高光伏出力预测的精度,提出了一种基于在线序列极限学习机的光伏发电中长期功率预测方法. 结合在线序列极限学习机学习速度快、泛化能力强的特点,通过对大量气象数据和历史发电数据综合处理,对光伏发电系统的输出功率进行预测. 同时,由于实时数据的不断输入,该方法能够对预测模型进行在线更新. 算例仿真研究表明,该预测方法与反向传播神经网络、支持向量机方法相比,能够有效提高预测精度,满足在线应用的需求,具有较好的应用前景.     

基于执行依赖启发式动态规划的船舶减摇鳍在线最优控制

针对船舶线性横摇系统,设计了一种基于执行依赖启发式动态规划(ADHDP)方法的在线学习最优减摇鳍控制器.在设计过程中直接使用输入输出数据获取系统状态值.利用评价网络来逼近针对船舶减摇鳍控制系统设计的性能指标函数,并通过执行网络获得最优控制律,这两个网络都是多层前馈神经网络,即反向传播(BP)神经网络.在训练过程中,这两个神经网络不仅可以使用实时测量数据,也可以减少船舶横摇模型的内部误差和不确定性干扰的影响,从而提高系统的鲁棒性.最后,仿真结果表明所提出的ADHDP控制器对于降低船舶横摇有很好的控制效果.