基于信号图像化和CNN-ResNet的配电网单相接地故障选线方法

配电网发生单相接地故障时,零序电流呈现较强的非线性与非平稳性,故障选线较为困难,针对此问题,提出一种基于信号图像化和卷积神经网络-残差网络的配电网单相接地故障选线方法。首先,利用排列熵优化变分模态分解算法的参数,将零序电流信号分解成一系列固有模态函数;其次,引入新的数据预处理方式,将固有模态函数转成二维图像,获得零序电流信号的时频特征图;最后,利用一维卷积神经网络提取零序电流信号的相关性和特征,利用残差网络提取时频特征图的特征,将两个网络融合,构建混合卷积神经网络结构,实现故障选线。仿真与实验结果表明,该方法能够在高阻接地、采样时间不同步、强噪声等情况下准确地选择出故障线路,可满足配电网对故障选线准确性和可靠性的需求。

State estimation for neural neutral-type networks with mixed time-varying delays and Markovian jumping parameters

This paper is concerned with a delay-dependent state estimator for neutral-type neural networks with mixed timevarying delays and Markovian jumping parameters.The addressed neural networks have a finite number of modes,and the modes may jump from one to another according to a Markov process.By construction of a suitable Lyapunov-Krasovskii functional,a delay-dependent condition is developed to estimate the neuron states through available output measurements such that the estimation error system is globally asymptotically stable in a mean square.The criterion is formulated in terms of a set of linear matrix inequalities（LMIs）,which can be checked efficiently by use of some standard numerical packages.

Reliability analysis of slope stability by neural network,principal component analysis,and transfer learning techniques

The prediction of slope stability is considered as one of the critical concerns in geotechnical engineering.Conventional stochastic analysis with spatially variable slopes is time-consuming and highly computation-demanding.To assess the slope stability problems with a more desirable computational effort,many machine learning(ML)algorithms have been proposed.However,most ML-based techniques require that the training data must be in the same feature space and have the same distribution,and the model may need to be rebuilt when the spatial distribution changes.This paper presents a new ML-based algorithm,which combines the principal component analysis(PCA)-based neural network(NN)and transfer learning(TL)techniques(i.e.PCAeNNeTL)to conduct the stability analysis of slopes with different spatial distributions.The Monte Carlo coupled with finite element simulation is first conducted for data acquisition considering the spatial variability of cohesive strength or friction angle of soils from eight slopes with the same geometry.The PCA method is incorporated into the neural network algorithm(i.e.PCA-NN)to increase the computational efficiency by reducing the input variables.It is found that the PCA-NN algorithm performs well in improving the prediction of slope stability for a given slope in terms of the computational accuracy and computational effort when compared with the other two algorithms(i.e.NN and decision trees,DT).Furthermore,the PCAeNNeTL algorithm shows great potential in assessing the stability of slope even with fewer training data.

NNL:a domain-specific language for neural networks

Recent years,neural networks(NNs)have received increasing attention from both academia and industry.So far significant diversity among existing NNs as well as their hardware platforms makes NN programming a daunting task.In this paper,a domain-specific language(DSL)for NNs,neural network language(NNL)is proposed to deliver productivity of NN programming and portable performance of NN execution on different hardware platforms.The productivity and flexibility of NN programming are enabled by abstracting NNs as a directed graph of blocks.The language describes 4 representative and widely used NNs and runs them on 3 different hardware platforms(CPU,GPU and NN accelerator).Experimental results show that NNs written with the proposed language are,on average,14.5%better than the baseline implementations across these 3 platforms.Moreover,compared with the Caffe framework that specifically targets the GPU platform,the code can achieve similar performance.

基于ANN模型的内冷型溶液除湿器性能研究

溶液除湿器因可被低品位热能驱动,且具有除湿效率高等优点而受到广泛关注,但其传质性能的预测准确度还有待提高.本文搭建了单通道内冷型溶液除湿实验平台,研究了不同参数对于除湿过程中传质性能的影响,同时,建立了基于MATLAB平台的人工神经网络(ANN)模型用于预测传质性能,并用上述实验数据对该ANN模型进行了验证.结果表明,ANN模型预测得出的Sh与实验Sh平均绝对相对偏差(MARD)为4.07%.与现有经验公式相比,建立的ANN模型预测精度更高.此外,还利用ANN模型研究了不同参数变化下的Sh的变化趋势,从而分析不同参数对除湿性能的影响.

基于PSO-CNN-LSTM的短期热负荷预测模型

为提高短期供热负荷预测精度,减少供热不均与供需失调所造成的能源浪费,提出一种基于粒子群(Particle swarm optimization,PSO)、经验模态分解(Empirical Mode Decomposition,EMD)、卷积神经网络(Convolutional Neural Network,CNN)和长短时记忆神经网络(Long Short Term Memory Network,LSTM)的混合预测模型。首先,针对供热负荷呈现非线性、复杂性等特点,采用EMD对供热负荷分解,从而实现弱化供热负荷复杂程度;其次,分别运用CNN与LSTM提取供热负荷空间特征与时域特征;最后,结合PSO算法对LSTM网络的超参数进行调整,寻找出最优参数。实验表明,结合EMD分解的PSOCNN-LSTM网络相比LSTM、CNN-LSTM、EMD-CNN-LSTM平均误差分别降低了44%、34%、24%、21%,拥有更高的预测精度和拟合效果。所提模型为集中供热负荷预测提供了一种新的思路,对于制定集中供热能源分配提供了参考意义。

基于两阶段中立型交叉效率和Shannon熵的公交服务效益评价:利益相关者视角

为了获得高区分度、更准确的公交服务效益评价结果,提出了一种基于两阶段中立型交叉效率和Shannon熵集结模型的组合评价方法。基于网络数据包络分析模型得到两阶段中立型交叉效率模型中的自评结果,在此基础上测算出互评结果,利用Shannon熵集结模型将自评与互评结果集结得出最终交叉效率值。从公交企业、公众和政府三方利益相关者的角度构建了公交服务效益评价指标体系,对2011~2016年间“长三角”地区案例进行研究。结果表明:大多数公交企业的运营状况均不佳,但综合效益总体上在缓慢提升;服务过程评价结果显著优于生产过程评价结果;小型规模公交企业综合效益最高,其次是中型规模企业,而大型规模企业最低;政府大量的补贴和公交企业过量的资源投入会对效益产生负面影响。

Machine Learning Approaches for the Solution of the Riemann Problem in Fluid Dynamics:a Case Study

We present our results by using a machine learning(ML)approach for the solution of the Riemann problem for the Euler equations of fluid dynamics.The Riemann problem is an initial-value problem with piecewise-constant initial data and it represents a mathematical model of the shock tube.The solution of the Riemann problem is the building block for many numerical algorithms in computational fluid dynamics,such as finite-volume or discontinuous Galerkin methods.Therefore,a fast and accurate approximation of the solution of the Riemann problem and construction of the associated numerical fluxes is of crucial importance.The exact solution of the shock tube problem is fully described by the intermediate pressure and mathematically reduces to finding a solution of a nonlinear equation.Prior to delving into the complexities of ML for the Riemann problem,we consider a much simpler formulation,yet very informative,problem of learning roots of quadratic equations based on their coefficients.We compare two approaches:(i)Gaussian process(GP)regressions,and(ii)neural network(NN)approximations.Among these approaches,NNs prove to be more robust and efficient,although GP can be appreciably more accurate(about 30\%).We then use our experience with the quadratic equation to apply the GP and NN approaches to learn the exact solution of the Riemann problem from the initial data or coefficients of the gas equation of state(EOS).We compare GP and NN approximations in both regression and classification analysis and discuss the potential benefits and drawbacks of the ML approach.

Existence of Periodic Solutions for Neutral-Type Neural Networks with Delays on Time Scales

In this paper, we employ a fixed point theorem due to Krasnosel’skii to attain the existence of periodic solutions for neutral-type neural networks with delays on a periodic time scale. Some new sufficient conditions are established to show that there exists a unique periodic solution by the contraction mapping principle.

Almost sure exponential stability of neutral stochastic delayed cellular neural networks

In this paper, almost sure exponential stability of neutral delayed cellular neural networks which are in the noised environment is studied by decomposing the state space to sub-regions in view of the saturation linearity of output functions of neurons of the cellular neural networks. Some algebraic criteria are obtained and easily verified. Some examples are given to illustrate the correctness of the results obtained.

基于TNN-BL模型的低压配电网断零与缺相故障检测方法研究

低压配电网中因断零与缺相故障对电网公司造成的安全隐患和经济损失一直是电网公司迫切解决的难题,随着智能化检测设备在电网中普及,可利用智能电表采集的低压侧负载电压和各序电流数据开展故障检测。首先,建立基于Transformer神经网络(TNN)和双向长短期记忆(Bi-LSTM)的混合模型TNN-BL;其次,通过选择合适的损失函数和正则化函数完善模型以进一步提高模型检测性能;最后,采用南网数据集对模型性能进行试验验证。试验结果表明,该方法拥有更有效的特征提取能力,相比于其他故障检测方法具有更高的检测准确度和更强的鲁棒性。

Periodic Solution for Neutral Type Neural Networks

The principle aim of this paper is to explore the existence of periodic solution of neural networks model with neutral delay. Sufficient and realistic conditions are obtained by means of an abstract continuous theorem of k-set contractive operator and some analysis technique.

Aero-engine Blade Fatigue Analysis Based on Nonlinear Continuum Damage Model Using Neural Networks

Fatigue life and reliability of aero-engine blade are always of important significance to flight safety.The establishment of damage model is one of the key factors in blade fatigue research.Conventional linear Miner＇s sum method is not suitable for aero-engine because of its low accuracy.A back propagation neutral network（BPNN） based on the combination of Levenberg-Marquardt（LM） and finite element method（FEM） is used to describe process of nonlinear damage accumulation behavior in material and predict fatigue life of the blade.Fatigue tests of standard specimen made from TC4 are carried out to obtain material fatigue parameters and S-N curve.A nonlinear continuum damage model（CDM）,based on the BPNN with one hidden layer and ten neurons,is built to investigate the nonlinear damage accumulation behavior,in which the results from the tests are used as training set.Comparing with linear models and previous nonlinear models,BPNN has the lowest calculation error in full load range.It has significant accuracy when the load is below 500 MPa.Especially,when the load is 350 MPa,the calculation error of the BPNN is only 0.4%.The accurate model of the blade is built by using 3D coordinate measurement technology.The loading cycle in fatigue analysis is defined from takeoff to cruise in 10 min,and the load history is obtained from finite element analysis（FEA）.Then the fatigue life of the compressor blade is predicted by using the BPNN model.The final fatigue life of the aero-engine blade is 6.55 104 cycles（10 916 h） based on the BPNN model,which is effective for the virtual design of aero-engine blade.

Synchronization and Exponential Estimates of Complex Networks with Mixed Time-varying Coupling Delays

Exponential estimates and sufficient conditions for the exponential synchronization of complex dynamical networks with bounded time-varying delays are given in terms of linear matrix inequalities （LMIs）. A generalized complex networks model involving both neutral delays and retarded ones is presented. The exponential synchronization problem of the complex networks is converted equivalently into the exponential stability problem of a group of uncorrelated delay functional differential equations with mixed timevarying delays. By utilizing the free weighting matrix technique, a less conservative delay-dependent synchronization criterion is derived. An illustrative example is provided to demonstrate the effectiveness of the proposed method.

Study on Decision Method of Neutral Point Grounding Mode for Medium-Voltage Distribution Network

The neutral grounding mode of medium-voltage distribution network decides the reliability, overvoltage, relay protection and electrical safety. Therefore, a comprehensive consideration of the reliability, safety and economy is particularly important for the decision of neutral grounding mode. This paper proposes a new decision method of neutral point grounding mode for mediumvoltage distribution network. The objective function is constructed for the decision according the life cycle cost. The reliability of the neutral point grounding mode is taken into account through treating the outage cost as an operating cost. The safety condition of the neutral point grounding mode is preserved as the constraint condition of decision models, so the decision method can generate the most economical and reliable scheme of neutral point grounding mode within a safe limit. The example is used to verify the feasibility and effectiveness of the decision method.

Novel delay-dependent stability analysis of Takagi-Sugeno fuzzy uncertain neural networks with time varying delays

This paper presents the stability analysis for a class of neural networks with time varying delays that are represented by the Takagi^ugeno IT-S） model. The main results given here focus on the stability criteria using a new Lyapunov functional. New relaxed conditions and new linear matrix inequality-based designs are proposed that outperform the previous results found in the literature. Numerical examples are provided to show that the achieved conditions are less conservative than the existing ones in the literature.

The improved neutral network and its application for valuing rock mass mechanical parameter

The artificial neutral network(ANN) has the ability that self-study and self-remember, its 3 layers BP network has been applied extensively, but sometimes because of serious multi-correlation between the variables, and a few observations while many variables, there usually will result into paralyzing in study, and the neutral network further development is restricted in the system to some extent. The partial least square regression(PLS) has its advantage of building the calculation model between the variables with strong multi-correlation, especially much effective on a few data and many variables. So a new and effective method-improved neutral network has been introduced-the neutral network based on the PLS. The results of example show the improved method has a few calculations and high accuracy, and provide a new way for valuing the rock mass mechanical parameters.

An Improved SPSA Algorithm for System Identification Using Fuzzy Rules for Training Neural Networks

Simultaneous perturbation stochastic approximation （SPSA） belongs to the class of gradient-free optimization methods that extract gradient information from successive objective function evaluation. This paper describes an improved SPSA algorithm, which entails fuzzy adaptive gain sequences, gradient smoothing, and a step rejection procedure to enhance convergence and stability. The proposed fuzzy adaptive simultaneous perturbation approximation （FASPA） algorithm is particularly well suited to problems involving a large number of parameters such as those encountered in nonlinear system identification using neural networks （NNs）. Accordingly, a multilayer perceptron （MLP） network with popular training algorithms was used to predicate the system response. We found that an MLP trained by FASPSA had the desired accuracy that was comparable to results obtained by traditional system identification algorithms. Simulation results for typical nonlinear systems demonstrate that the proposed NN architecture trained with FASPSA yields improved system identification as measured by reduced time of convergence and a smaller identification error.

Efficient Training of Multi-Layer Neural Networks to Achieve Faster Validation

Artificial neural networks(ANNs)are one of the hottest topics in computer science and artificial intelligence due to their potential and advantages in analyzing real-world problems in various disciplines,including but not limited to physics,biology,chemistry,and engineering.However,ANNs lack several key characteristics of biological neural networks,such as sparsity,scale-freeness,and small-worldness.The concept of sparse and scale-free neural networks has been introduced to fill this gap.Network sparsity is implemented by removing weak weights between neurons during the learning process and replacing them with random weights.When the network is initialized,the neural network is fully connected,which means the number of weights is four times the number of neurons.In this study,considering that a biological neural network has some degree of initial sparsity,we design an ANN with a prescribed level of initial sparsity.The neural network is tested on handwritten digits,Arabic characters,CIFAR-10,and Reuters newswire topics.Simulations show that it is possible to reduce the number of weights by up to 50%without losing prediction accuracy.Moreover,in both cases,the testing time is dramatically reduced compared with fully connected ANNs.

Using Neural Networks to Predict Secondary Structure for Protein Folding

Protein Secondary Structure Prediction (PSSP) is considered as one of the major challenging tasks in bioinformatics, so many solutions have been proposed to solve that problem via trying to achieve more accurate prediction results. The goal of this paper is to develop and implement an intelligent based system to predict secondary structure of a protein from its primary amino acid sequence by using five models of Neural Network (NN). These models are Feed Forward Neural Network (FNN), Learning Vector Quantization (LVQ), Probabilistic Neural Network (PNN), Convolutional Neural Network (CNN), and CNN Fine Tuning for PSSP. To evaluate our approaches two datasets have been used. The first one contains 114 protein samples, and the second one contains 1845 protein samples.