基于TLBGA-GRU神经网络的短期负荷预测

短期负荷预测在电网调度安排和电力市场交易中发挥着重要作用,预测精度高,有利于提高发电设备的利用率和经济调度的有效性。为充分挖掘负荷数据中时序性特征的联系,解决神经网络中由超参数的随机选取导致的预测精度下降问题,提出一种基于教与学的遗传算法（TLBGA）和门控循环单元（GRU）神经网络的短期负荷预测方法。利用灰色关联分析法对原始数据进行相关度分析,剔除冗余特征,使输入与输出保持较好的映射关系,在遗传算法中加入一种基于教与学优化的新型变异算子,用于防止其出现早熟收敛问题,从而提高解的质量。在此基础上,运用改进后的TLBGA算法对GRU神经网络模型进行超参数寻优,更新GRU的模型超参数并使其性能达到最佳状态,以提高负荷预测的精度。对欧洲某地区的电力负荷数据集和美国PJM电力市场公开负荷数据集进行预测,结果表明,该方法的预测精度分别达到了97.1%和97.2%,相比反向传播神经网络、循环神经网络及GRU神经网络模型,具有更高的预测精度。     

基于自注意力机制和双向GRU神经网络的深度知识追踪优化模型

针对现有深度知识追踪模型存在输入习题间复杂关系捕获能力弱、无法有效处理长序列输入数据等问题,提出了基于自注意力机制和双向GRU神经网络的深度知识追踪优化模型(KTSA-BiGRU)。首先,将学习者的历史学习交互序列数据映射为实值向量序列;其次,以实值向量序列作为输入训练双向GRU神经网络,利用双向GRU神经网络建模学习者的学习过程;最后,使用自注意力机制捕获练习题之间的关系,根据双向GRU神经网络输出的隐向量和注意力权重计算学习者正确回答下一问题的概率。实验在三个公共数据集上的性能分析优于现有的知识追踪模型,能提高深度知识追踪的预测精度。     

Modelling of diesel engine performance using advanced machine learning methods under scarce and exponential data set

Traditional methods on creating diesel engine models include the analytical methods like multi-zone models and the intelligent based models like artificial neural network (ANN) based models. However, those analytical models require excessive assumptions while those ANN models have many drawbacks such as the tendency to overfitting and the difficulties to determine the optimal network structure. In this paper, several emerging advanced machine learning techniques, including least squares support vector machine (LS-SVM), relevance vector machine (RVM), basic extreme learning machine (ELM) and kernel based ELM, are newly applied to the modelling of diesel engine performance. Experiments were carried out to collect sample data for model training and verification. Limited by the experiment conditions, only 24 sample data sets were acquired, resulting in data scarcity. Six-fold cross-validation is therefore adopted to address this issue. Some of the sample data are also found to suffer from the problem of data exponentiality, where the engine performance output grows up exponentially along the engine speed and engine torque. This seriously deteriorates the prediction accuracy. Thus, logarithmic transformation of dependent variables is utilized to pre-process the data. Besides, a hybrid of leave-one-out cross-validation and Bayesian inference is, for the first time, proposed for the selection of hyperparameters of kernel based ELM. A comparison among the advanced machine learning techniques, along with two traditional types of ANN models, namely back propagation neural network (BPNN) and radial basis function neural network (RBFNN), is conducted. The model evaluation is made based on the time complexity, space complexity, and prediction accuracy. The evaluation results show that kernel based ELM with the logarithmic transformation and hybrid inference is far better than basic ELM, LS-SVM, RVM, BPNN and RBFNN, in terms of prediction accuracy and training time.     

A SOM-based hybrid linear-neural model for short-term load forecasting

In this paper, a short-term load forecasting method is considered, which is based upon a flexible smooth transition autoregressive (STAR) model. The described model is a linear model with time varying coefficients, which are the outputs of a single hidden layer feedforward neural network. The hidden layer is responsible for partitioning the input space into multiple sub-spaces through multivariate thresholds and smooth transition between the sub-spaces. In this paper, we propose a new method to smartly initialize the weights of the hidden layer of the neural network before its training. A self-organizing map (SOM) network is applied to split the historical data dynamics into clusters, and the Ho-Kashyap algorithm is then used to obtain the separating planes' equations. Applied to the electricity markets, the proposed method is better able to model the smooth transitions between the different regimes, which are present in the load demand series because of market effects and season effects. We use data from three electricity markets to compare the prediction accuracy of the proposed method with traditional benchmarks and other recent models, and find our results to be competitive.     

基于PSO-GRU神经网络的青椒生长期需水预测

青椒生长期内需水量与气温、气压、相对湿度等因子之间存在复杂的非线性关系,需水量变化呈现出时序性和周期性的规律,为提高青椒生长期日均需水量的预测精度,提出一种PSO-GRU青椒生长期日均需水预测模型。以2014—2018年实验所得的青椒需水和气象环境等数据为数据源,将日均气温、气压、风速等六维数据作为特征集,需水量作为标签,采用GRU神经网络作为需水预测的训练模型,并针对GRU超参数容易陷入局部最优的问题,利用粒子群算法(PSO）优化GRU模型的超参数,通过仿真实验对青椒生长期日均需水量进行预测,并与RNN,LSTM和GRU等模型进行对比,验证PSO-GRU模型的优越性。仿真实验结果表明：PSO-GRU模型的预测精度和拟合效果显著提高,RMSE为0.505,MAE为0.388,MAPE为7.73,R2为0.888。PSO-GRU模型可为制定灌溉计划提供依据,有利于节水灌溉,推动农业种植水利信息化。     

基于组合模型的短时交通流预测方法

准确、实时的交通流预测对交通规划、交通管理和交通控制具有重要意义.然而,由于道路网络拓扑结构约束和交通流随时间动态变化的空时相关特性,交通流预测仍然具有挑战性.为了同时捕获交通流的空间和时间相关性,提出一种将图卷积网络(GCN)和门控循环单元(GRU)相结合的组合模型方法.利用GCU能够灵活处理图结构数据的优点来捕捉各个路段的空间特征,继而发挥GRU在处理时间序列方面的优势挖掘交通流的内在时间规律,空时融合后得到最终预测结果.利用美国交通研究数据实验室的高速公路交通数据对该模型进行仿真验证,结果表明,所提出的GCN-GRU组合模型方法具有更高的预测精度,预测结果优于自回归积分滑动平均(ARIMA)模型和GRU模型等基准预测方法.     

基于迭代误差补偿的核极端学习机模型在短期电力负荷预测中的应用

针对BP神经网络方法制约短期电力负荷预测精度的问题,提出一种基于迭代误差补偿的核极端学习机(KELM-IEC)预测模型。首先,建立短期电力负荷预测模型的输入指标体系,选择月份、日期、星期、周数、是否为节假日、日平均气温、前一日的最大负荷量等影响电力负荷的7个因素作为预测模型的输入;其次,基于新型神经网络模型--核极端学习机(KELM),建立负荷预测模型,引入支持向量机(SVM)的核函数映射作为极端学习机(ELM)的隐含层节点映射,有效结合ELM结构简单、训练简便与SVM泛化能力强的优势,提高负荷预测精度;最后,基于时间序列预测中迭代误差补偿(IEC)技术,建立IEC模型,再次利用KELM对负荷预测模型的预测误差进行学习,从而对预测结果进行补偿和修正,进一步减小模型预测误差,提高预测性能。采用两组实际电力负荷数据进行仿真实验,其中,KELM-IEC模型与BP神经网络模型相比,平均绝对百分误差(MAPE)分别降低了74.39%和34.73%,最大绝对误差(ME)分别降低了58.34%和39.58%;同时与KELM模型相比,平均绝对百分误差分别降低了18.60%和4.29%,最大绝对误差分别降低了0.08%和11.21%,说明误差补偿策略的必要性。实验结果表明,KELM-IEC预测模型能够有效地提高短期电力负荷预测的精度,有利于改善电力系统的计划、运营和管理,保障生产和生活用电,提高经济效益和社会效益。     

基于CNN和GRU的混合股指预测模型研究

针对股票数据共线性和非线性的特点,提出了一种基于卷积神经网络（Convolutional Neural Network,CNN）和门控循环单元（Gated Recurrent Unit,GRU）神经网络的混合预测模型,并对沪深300指数、上证综指和深证成指进行了预测。该模型首先采用CNN提取特征向量,对原始数据进行降维,然后利用GRU神经网络学习特征动态变化规律进行股指预测。仿真结果表明,与GRU神经网络、长短时记忆（Long-Short-Term Memory,LSTM）神经网络和CNN相比,该模型能够挖掘历史数据中蕴含的信息,有效提高股指预测的准确率,并可为股指交易提供一些参考。     

基于结合混沌纵横交叉的粒子群算法优化极限学习机的短期负荷预测

为了解决传统的单一负荷预测模型精度低以及常规智能算法在解决高维、多模复杂问题时容易陷入局部最优的问题,提出了一种结合混沌纵横交叉的粒子群算法（CC-PSO）优化极限学习机(ELM)的短期负荷预测模型。ELM的泛化能力与其输入权值和隐含层偏置密切相关,采用结合混沌纵横交叉的粒子群算法优化ELM的输入权值与隐含层偏置,提高了ELM的泛化能力和预测精度。选择广东某地区实际电网负荷数据进行分析,研究结果表明,相对于BP神经网络和支持向量机,ELM具有更高的泛化能力和预测精度;CC-PSO相对于粒子群和遗传算法具有更高的全局搜索能力,CC-PSO-ELM模型具有较高的负荷预测精度。     

基于正则化GRU模型的洪水预测

针对传统神经网络模型在洪水预测过程中存在准确性低、过拟合等问题,本文以赣江流域外洲水文站每月平均水位为研究对象,提出基于正则化GRU神经网络的洪水预测模型来提高洪水预报精度.选用relu函数作为整个神经网络的输出层激活函数,将弹性网正则化引入到GRU模型中,对网络中输入权重w实施正则化处理,以提升GRU模型的泛化性能,并将该模型应用于外洲水文站每月平均水位的拟合及预测.实验对比表明,弹性网正则化优化后的模型预测拟合程度较高,合格率提高了9.3%,计算出的均方根误差较小.     

基于NARX神经网络的热负荷预测中关键影响因素分析

在区域供热（DH）网络中,精确预测热负荷已被认为是提高效率和节省成本的重要环节。为了提高预测精度,研究不同影响因素对热负荷预测的影响极为重要。使用引入不同影响因素的数据集训练得到带外部输入的非线性自回归（NARX）神经网络模型,并比较其预测性能,以讨论直接太阳辐射和风速对热负荷预测的影响程度。实验结果表明,直接太阳辐射和风速都是热负荷预测中的关键影响因素。只引入风速时,预测模型的平均绝对百分比误差（MAPE）和均方根误差（RMSE）均低于只引入直接太阳辐射,同时引入风速和直接太阳辐射能够得到最佳的模型预测性能,但是对于MAPE和RMSE降低的贡献不大。     

基于FWADE-ELM的短时交通流预测方法

受道路环境和人为因素影响,实际交通系统可视为一个复杂的非线性动力系统,交通流数据具有较强的非线性、时变性和易受随机噪声影响等特征.针对复杂环境下的短时交通流预测问题,提出一种基于烟花差分进化混合算法-极限学习机的短时交通流预测方法.采用奇异谱分析方法滤除原始交通流数据中包含的噪声成分,降噪后的交通流数据用于训练极限学习机(ELM)网络预测模型.进行相空间重构,利用C-C算法确定ELM网络的结构和关键参数.通过融合烟花算法和差分进化算法提出一种烟花差分进化混合算法,可有效提高基本算法的整体优化性能.将改进的混合优化算法用于优化ELM网络的权阈值(结构为9-11-1,维数为110),建立短时交通流预测模型.测试与应用结果表明,所构建的短时交通流预测模型具有较高的预测精度和较强的泛化能力(均方误差为7.75,平均绝对百分比误差为0.086 7),预测值与实际值的拟合程度较好.     

基于多时空图卷积网络的交通流预测

交通流预测在交通管理和城市规划的应用中具有重要意义,然而现有的预测方法无法充分挖掘其潜在的复杂时空相关性,为进一步挖掘路网道路网络数据的时空特性以提高预测精度,提出一种多时空图卷积网络(multi-spatial-temporal graph convolutional network,MST-GCN)模型。首先,利用切比雪夫图卷积(ChebNet)结合门控循环单元(GRU)构建时空组件以深度挖掘节点的时空相关性;其次,分别提取周相关、日相关、邻近时间的序列数据,输入三个时空组件以深度挖掘不同时间窗口间的时间相关性;最后,将时空组件与编码器—解码器网络结构(encoder-decoder)融合组建MST-GCN模型。利用加利福尼亚州交通局(Caltrans)性能评估系统中高速公路数据集PEMS04和PEMS08进行实验,结果表明新模型的性能明显优于门控循环单元模型和最近提出的扩散卷积循环神经网络(DCRNN)、时间图卷积网络(T-GCN)、基于注意力机制的时空图卷积神经网络(ASTGCN)和时空同步图卷积网络(STSGCN)模型。     

基于两级筛选机制及深度学习组合模型实现短时交通流预测

准确实时的短时交通流预测对现代交通管理服务体系的构建至关重要.为了充分挖掘并利用不同路段短时交通流交互作用而表现出的时空特性,构建由自相关函数、互相关函数和KNN算法组成的两级筛选机制评估与目标路段的相关性优化路段组合,实现空间信息深度挖掘;提出一种GCN-GRU组合预测模型,利用图卷积网络(GCN)全局处理路段拓扑信息的优势进一步捕捉短时交通流的空间特性,并借助门控循环单元(GRU)对时间信息的长时记忆能力提取其时间特性.利用实测高速公路短时交通流数据进行验证,仿真结果表明,采用两级筛选机制对路段进行有效筛选并引入深度学习组合模型,预测性能明显改善,优于堆栈式自编码网络(SAEs)和GRU等经典模型.     

Non-tuned machine learning approach for hydrological time series forecasting

Stream-flow forecasting is a crucial task for hydrological science. Throughout the literature, traditional and artificial intelligence models have been applied to this task. An attempt to explore and develop better expert models is an ongoing endeavor for this hydrological application. In addition, the accuracy of modeling, confidence and practicality of the model are the other significant problems that need to be considered. Accordingly, this study investigates modern non-tuned machine learning data-driven approach, namely extreme learning machine (ELM). This data-driven approach is containing single layer feedforward neural network that selects the input variables randomly and determine the output weights systematically. To demonstrate the reliability and the effectiveness, one-step-ahead stream-flow forecasting based on three time-scale pattern (daily, mean weekly and mean monthly) for Johor river, Malaysia, were implemented. Artificial neural network (ANN) model is used for comparison and evaluation. The results indicated ELM approach superior the ANN model level accuracies and time consuming in addition to precision forecasting in tropical zone. In measureable terms, the dominance of ELM model over ANN model was indicated in accordance with coefficient determination (R ²) root-mean-square error (RMSE) and mean absolute error (MAE). The results were obtained for example the daily time scale R ² = 0.94 and 0.90, RMSE = 2.78 and 11.63, and MAE = 0.10 and 0.43, for ELM and ANN models respectively.

     

基于深度残差网络和GRU的SqueezeNet模型的交通路标识别

Existing traffic road sign recognition methods are all based on convolutional neural networks. As the number of the model network layers increases, the recognition accuracy will also be improved, but there are still some problems such as the reduction of efficiency and the increase of the number of parameters. Therefore, an improved SqueezeNet model combining deep residual network with GRU neural network (SqueezeNet IR GRU) is proposed. In order to enhance the learning efficiency, ELU function is used as the activation function. To avoid the disappearance of gradients when the network layer is too deep, a deep residual network is introduced to guarantee the stability of the model, GRU neural network that can memorize the important past features is utilized. Experiments were performed on the Cafir 10 and GTSRB datasets, and their recognition accuracy rates are above 99.13% and 88.25%respectively. The experimental results show that the SqueezeNet IR GRU model not only reduces the parameter amount greatly, but also its convergence, stability and recall rate are all much better than others.     

基于条件变分自编码的密码攻击算法

使用密码猜测算法是评估用户密码强度和安全性的有效方法,提出一种基于条件变分自编码密码猜测算法PassCVAE。算法基于条件变分自编码模型,将用户个人信息作为条件特征,训练密码攻击模型。在编码器端,分别使用双向循环神经网络（GRU）和文本卷积神经网络（TextCNN）,实现对密码序列和用户个人信息的编码和特征的抽象提取;在解码器端使用两层GRU神经网络,实现对用户个人信息和密码数据隐编码的解码,生成密码序列。该算法可以有效地拟合密码数据的分布和字符组合规律,生成高质量的猜测密码数据。多组实验结果表明,提出的PassCVAE算法优于现有的主流密码猜测算法。     

Dynamic filter weights neural network model integrated with differential evolution for day-ahead price forecasting in energy market

In this paper a new dynamic model for forecasting electricity prices from 1 to 24 h in advance is proposed. The model is a dynamic filter weight Adaline using a sliding mode weight adaptation technique. The filter weights for this neuron constitute of first order dynamic filter with adjustable parameters. Sliding mode invariance conditions determine a least square characterization of the adaptive weights average dynamics whose stability features may be studied using standard time varying linear system results. The approach is found to exhibit robustness characteristics and first convergence properties. Comparison of results with a local linear wavelet neural network model is also presented in this paper. The hourly electricity prices of California and Spanish energy markets are taken as experimental data and the Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) are computed to find out the forecasting performance of both the models. In both the cases the MAPE and RMSE are found to be within the tolerable limits. As dynamic filter weight neural network gives better results in comparison to local linear wavelet neural network, the former has been further integrated with differential evolution algorithm to enhance the performance.     

Using deep belief network to construct the agricultural information system based on Internet of Things

To study the impact of the agricultural information system based on the Internet of Things (IoT) on the income of agricultural products, an agricultural information system was constructed based on the agricultural IoT technology, and its impact on the income of agricultural products was discussed through the deep belief network. First, the relevant theories of agricultural IoT were introduced. Then, an agricultural information system based on agricultural IoT technology was constructed, and a deep belief network model was proposed. The vegetable prices and influencing factors were collected. The data were distributed in the range of 0–1 after normalization. The collinearity between the data was eliminated through principal component analysis. Then, the principal component analysis of vegetable prices and influencing factors from 2015 to 2019 was performed. A total of 96 sample data of calibration set and 24 sample data of test machine were collected. The optimal number of hidden layers of the deep belief network model and the number of nodes contained in the hidden layer were obtained through experiments. The results show that the first, second, and third hidden layers have 8, 6, and 10 nodes, respectively; the prediction accuracy of the deep belief network model is more accurate than that of the BP neural network and wavelet neural network. Besides, the absolute value of the prediction error of the deep belief model is within 0.1, which has good prediction accuracy. In short, the deep belief model has a good development prospect in agricultural product price forecasting, and it can provide relevant reference for the establishment and research of other agricultural product price forecasting models.

     

Uncertainty evaluation and model selection of extreme learning machine based on Riemannian metric

Considering the uncertainty of hidden neurons, choosing significant hidden nodes, called as model selection, has played an important role in the applications of extreme learning machines(ELMs). How to define and measure this uncertainty is a key issue of model selection for ELM. From the information geometry point of view, this paper presents a new model selection method of ELM for regression problems based on Riemannian metric. First, this paper proves theoretically that the uncertainty can be characterized by a form of Riemannian metric. As a result, a new uncertainty evaluation of ELM is proposed through averaging the Riemannian metric of all hidden neurons. Finally, the hidden nodes are added to the network one by one, and at each step, a multi-objective optimization algorithm is used to select optimal input weights by minimizing this uncertainty evaluation and the norm of output weight simultaneously in order to obtain better generalization performance. Experiments on five UCI regression data sets and cylindrical shell vibration data set are conducted, demonstrating that the proposed method can generally obtain lower generalization error than the original ELM, evolutionary ELM, ELM with model selection, and multi-dimensional support vector machine. Moreover, the proposed algorithm generally needs less hidden neurons and computational time than the traditional approaches, which is very favorable in engineering applications.