Long Short-Term Memory Recurrent Neural Network-Based Acoustic Model Using Connectionist Temporal Classification on a Large-Scale Training Corpus

A Long Short-Term Memory(LSTM) Recurrent Neural Network(RNN) has driven tremendous improvements on an acoustic model based on Gaussian Mixture Model(GMM). However, these models based on a hybrid method require a forced aligned Hidden Markov Model(HMM) state sequence obtained from the GMM-based acoustic model. Therefore, it requires a long computation time for training both the GMM-based acoustic model and a deep learning-based acoustic model. In order to solve this problem, an acoustic model using CTC algorithm is proposed. CTC algorithm does not require the GMM-based acoustic model because it does not use the forced aligned HMM state sequence. However, previous works on a LSTM RNN-based acoustic model using CTC used a small-scale training corpus. In this paper, the LSTM RNN-based acoustic model using CTC is trained on a large-scale training corpus and its performance is evaluated. The implemented acoustic model has a performance of 6.18% and 15.01% in terms of Word Error Rate(WER) for clean speech and noisy speech, respectively. This is similar to a performance of the acoustic model based on the hybrid method.

Short-TermWind Power Prediction Based on Combinatorial Neural Networks

Wind power volatility not only limits the large-scale grid connection but also poses many challenges to safe grid operation.Accurate wind power prediction can mitigate the adverse effects of wind power volatility on wind power grid connections.For the characteristics of wind power antecedent data and precedent data jointly to determine the prediction accuracy of the prediction model,the short-term prediction of wind power based on a combined neural network is proposed.First,the Bi-directional Long Short Term Memory(BiLSTM)network prediction model is constructed,and the bi-directional nature of the BiLSTM network is used to deeply mine the wind power data information and find the correlation information within the data.Secondly,to avoid the limitation of a single prediction model when the wind power changes abruptly,the Wavelet Transform-Improved Adaptive Genetic Algorithm-Back Propagation(WT-IAGA-BP)neural network based on the combination of the WT-IAGA-BP neural network and BiLSTM network is constructed for the short-term prediction of wind power.Finally,comparing with LSTM,BiLSTM,WT-LSTM,WT-BiLSTM,WT-IAGA-BP,and WT-IAGA-BP&LSTM prediction models,it is verified that the wind power short-term prediction model based on the combination of WT-IAGA-BP neural network and BiLSTM network has higher prediction accuracy.

Conditional Random Field Tracking Model Based on a Visual Long Short Term Memory Network

In dense pedestrian tracking,frequent object occlusions and close distances between objects cause difficulty when accurately estimating object trajectories.In this study,a conditional random field tracking model is established by using a visual long short term memory network in the three-dimensional(3D)space and the motion estimations jointly performed on object trajectory segments.Object visual field information is added to the long short term memory network to improve the accuracy of the motion related object pair selection and motion estimation.To address the uncertainty of the length and interval of trajectory segments,a multimode long short term memory network is proposed for the object motion estimation.The tracking performance is evaluated using the PETS2009 dataset.The experimental results show that the proposed method achieves better performance than the tracking methods based on the independent motion estimation.

Short-Term Relay Quality Prediction Algorithm Based on Long and Short-Term Memory

The fraction defective of semi-finished products is predicted to optimize the process of relay production lines, by which production quality and productivity are increased, and the costs are decreased. The process parameters of relay production lines are studied based on the long-and-short-term memory network. Then, the Keras deep learning framework is utilized to build up a short-term relay quality prediction algorithm for the semi-finished product. A simulation model is used to study prediction algorithm. The simulation results show that the average prediction absolute error of the fraction is less than 5%. This work displays great application potential in the relay production lines.

State of Health Estimation of Lithium-Ion Batteries Using Support Vector Regression and Long Short-Term Memory

Lithium-ion batteries are the most widely accepted type of battery in the electric vehicle industry because of some of their positive inherent characteristics. However, the safety problems associated with inaccurate estimation and prediction of the state of health of these batteries have attracted wide attention due to the adverse negative effect on vehicle safety. In this paper, both machine and deep learning models were used to estimate the state of health of lithium-ion batteries. The paper introduces the definition of battery health status and its importance in the electric vehicle industry. Based on the data preprocessing and visualization analysis, three features related to actual battery capacity degradation are extracted from the data. Two learning models, SVR and LSTM were employed for the state of health estimation and their respective results are compared in this paper. The mean square error and coefficient of determination were the two metrics for the performance evaluation of the models. The experimental results indicate that both models have high estimation results. However, the metrics indicated that the SVR was the overall best model.

Short Term Traffic Flow Prediction Using Hybrid Deep Learning

Traffic flow prediction in urban areas is essential in the IntelligentTransportation System (ITS). Short Term Traffic Flow (STTF) predictionimpacts traffic flow series, where an estimation of the number of vehicleswill appear during the next instance of time per hour. Precise STTF iscritical in Intelligent Transportation System. Various extinct systems aim forshort-term traffic forecasts, ensuring a good precision outcome which was asignificant task over the past few years. The main objective of this paper is topropose a new model to predict STTF for every hour of a day. In this paper,we have proposed a novel hybrid algorithm utilizing Principal ComponentAnalysis (PCA), Stacked Auto-Encoder (SAE), Long Short Term Memory(LSTM), and K-Nearest Neighbors (KNN) named PALKNN. Firstly, PCAremoves unwanted information from the dataset and selects essential features.Secondly, SAE is used to reduce the dimension of input data using onehotencoding so the model can be trained with better speed. Thirdly, LSTMtakes the input from SAE, where the data is sorted in ascending orderbased on the important features and generates the derived value. Finally,KNN Regressor takes information from LSTM to predict traffic flow. Theforecasting performance of the PALKNN model is investigated with OpenRoad Traffic Statistics dataset, Great Britain, UK. This paper enhanced thetraffic flow prediction for every hour of a day with a minimal error value.An extensive experimental analysis was performed on the benchmark dataset.The evaluated results indicate the significant improvement of the proposedPALKNN model over the recent approaches such as KNN, SARIMA, LogisticRegression, RNN, and LSTM in terms of root mean square error (RMSE)of 2.07%, mean square error (MSE) of 4.1%, and mean absolute error (MAE)of 2.04%.

ConvNeXt网络及Stacked BiLSTM-Self-Attention在轴承剩余寿命预测中的应用

在滚动轴承剩余使用寿命预测方面,采用传统方法时存在鲁棒性差、精度低等各种问题。近些年来深度学习的发展为解决这些问题提供了新的思路。为了进一步提高对轴承寿命的预测精度,提出了一种基于ConvNeXt网络、堆叠双向长短时记忆网络(SBiLSTM)和自注意力机制(Self-Attention)的滚动轴承寿命预测方法。首先,采用连续小波变换(CWT)构造了振动信号的时频图,以更好地捕捉信号的时域和频域特征;然后,将得到的时频图输入到构建的ConvNeXt网络中,通过卷积、池化和层归一化等操作,对时频图的关键特征进行了提取;最后,将提取后的特征输入到SBiLSTM-Self-Attention模块中,进一步提取了时序信息和特征权重分配数据,利用PHM2012挑战数据集进行了验证,通过实验分析了该方法的均方根误差(RMSE)和平均绝对误差(MAE)。研究结果表明:相较于现有技术方法,该方法的平均RMSE为0.031;与其他三种方法,即卷积神经网络(CNN)、深度残差双向门控循环单元(DRN-BiGRU)和深度卷积自注意力双向门控循环单元(DCNN-Self-Attention-BiGRU)相比,其平均RMSE值分别下降了79%、74%和55%,MAE值分别下降了78%、73%和53%,说明该方法在滚动轴承剩余寿命预测中有较好的性能。

Wind Speed Short-Term Prediction Based on Empirical Wavelet Transform, Recurrent Neural Network and Error Correction

Predicting wind speed accurately is essential to ensure the stability of the wind power system and improve the utilization rate of wind energy.However,owing to the stochastic and intermittent of wind speed,predicting wind speed accurately is difficult.A new hybrid deep learning model based on empirical wavelet transform,recurrent neural network and error correction for short-term wind speed prediction is proposed in this paper.The empirical wavelet transformation is applied to decompose the original wind speed series.The long short term memory network and the Elman neural network are adopted to predict low-frequency and high-frequency wind speed sub-layers respectively to balance the calculation efficiency and prediction accuracy.The error correction strategy based on deep long short term memory network is developed to modify the prediction errors.Four actual wind speed series are utilized to verify the effectiveness of the proposed model.The empirical results indicate that the method proposed in this paper has satisfactory performance in wind speed prediction.

Deep Learning Network for Energy Storage Scheduling in Power Market Environment Short-Term Load Forecasting Model

In the electricity market,fluctuations in real-time prices are unstable,and changes in short-term load are determined by many factors.By studying the timing of charging and discharging,as well as the economic benefits of energy storage in the process of participating in the power market,this paper takes energy storage scheduling as merely one factor affecting short-term power load,which affects short-term load time series along with time-of-use price,holidays,and temperature.A deep learning network is used to predict the short-term load,a convolutional neural network(CNN)is used to extract the features,and a long short-term memory(LSTM)network is used to learn the temporal characteristics of the load value,which can effectively improve prediction accuracy.Taking the load data of a certain region as an example,the CNN-LSTM prediction model is compared with the single LSTM prediction model.The experimental results show that the CNN-LSTM deep learning network with the participation of energy storage in dispatching can have high prediction accuracy for short-term power load forecasting.

基于LSTM-NeuralProphet模型的城市需水预测方法研究

城市水资源规划和管理是确保城市可持续发展和居民生活基本需求得到满足的关键环节,城市短期需水预测是城市水资源规划和管理的基础。由于气温、降水量和蒸发量等随季节变化明显,直接影响不同季节的用水峰值、高峰期,导致传统基于时间序列算法的固定时隙预测无法适应时隙的变化,从而不能保证预测精度。针对固定时隙预测精度低的问题,研究了基于四季24 h时间分辨率和夏季15 min时间分辨率的双时间尺度城市短期需水预测模型。该模型使用Anomaly-Transformer模型进行异常值检测,并通过分段曲线拟合对异常值校正,采用主成分分析法对城市短期需水影响因子进行分析提取主成分,在AutoML的标准模型分析中选取三个效果最好的模型作为Stacking模型的基学习器再结合长短期记忆网络(Long Short-Term Memory,LSTM)和Optune框架超参数优化后的NeuralProphet模型对双时间尺度的城市短期需水量进行预测,同时加入安全网机制,以保证LSTM-NeuralProphet模型的精确度。与其他模型(LSTM模型、NeuralProphet模型、BP神经网络模型)相比,LSTM-NeuralProphet模型的平均绝对误差在四季24 h时间分辨率的数据集上降低了0.18%~1.96%,在夏季15 min时间分辨率的数据集上降低了0.45%~11.90%。实验结果表明,LSTM-NeuralProphet模型具有更好的拟合效果和更高的预测精度,能较准确地预测双时间尺度下的城市需水量,可以较好地应用于城市短期需水预测研究中。

Ensembling Neural Networks for User’s Indoor Localization Using Magnetic Field Data from Smartphones

Predominantly the localization accuracy of the magnetic field-based localization approaches is severed by two limiting factors:Smartphone heterogeneity and smaller data lengths.The use of multifarioussmartphones cripples the performance of such approaches owing to the variability of the magnetic field data.In the same vein,smaller lengths of magnetic field data decrease the localization accuracy substantially.The current study proposes the use of multiple neural networks like deep neural network(DNN),long short term memory network(LSTM),and gated recurrent unit network(GRN)to perform indoor localization based on the embedded magnetic sensor of the smartphone.A voting scheme is introduced that takes predictions from neural networks into consideration to estimate the current location of the user.Contrary to conventional magnetic field-based localization approaches that rely on the magnetic field data intensity,this study utilizes the normalized magnetic field data for this purpose.Training of neural networks is carried out using Galaxy S8 data while the testing is performed with three devices,i.e.,LG G7,Galaxy S8,and LG Q6.Experiments are performed during different times of the day to analyze the impact of time variability.Results indicate that the proposed approach minimizes the impact of smartphone variability and elevates the localization accuracy.Performance comparison with three approaches reveals that the proposed approach outperforms them in mean,50%,and 75%error even using a lesser amount of magnetic field data than those of other approaches.

Synthetic well logs generation via Recurrent Neural Networks

To supplement missing logging information without increasing economic cost, a machine learning method to generate synthetic well logs from the existing log data was presented, and the experimental verification and application effect analysis were carried out. Since the traditional Fully Connected Neural Network(FCNN) is incapable of preserving spatial dependency, the Long Short-Term Memory(LSTM) network, which is a kind of Recurrent Neural Network(RNN), was utilized to establish a method for log reconstruction. By this method, synthetic logs can be generated from series of input log data with consideration of variation trend and context information with depth. Besides, a cascaded LSTM was proposed by combining the standard LSTM with a cascade system. Testing through real well log data shows that: the results from the LSTM are of higher accuracy than the traditional FCNN; the cascaded LSTM is more suitable for the problem with multiple series data; the machine learning method proposed provides an accurate and cost effective way for synthetic well log generation.

Bitcoin Candlestick Prediction with Deep Neural Networks Based on Real Time Data

Currently,Bitcoin is the world’s most popular cryptocurrency.The price of Bitcoin is extremely volatile,which can be described as high-benefit and high-risk.To minimize the risk involved,a means of more accurately predicting the Bitcoin price is required.Most of the existing studies of Bitcoin prediction are based on historical(i.e.,benchmark)data,without considering the real-time(i.e.,live)data.To mitigate the issue of price volatility and achieve more precise outcomes,this study suggests using historical and real-time data to predict the Bitcoin candlestick—or open,high,low,and close(OHLC)—prices.Seeking a better prediction model,the present study proposes time series-based deep learning models.In particular,two deep learning algorithms were applied,namely,long short-term memory(LSTM)and gated recurrent unit(GRU).Using real-time data,the Bitcoin candlesticks were predicted for three intervals:the next 4 h,the next 12 h,and the next 24 h.The results showed that the best-performing model was the LSTM-based model with the 4-h interval.In particular,this model achieved a stellar performance with a mean absolute percentage error(MAPE)of 0.63,a root mean square error(RMSE)of 0.0009,a mean square error(MSE)of 9e-07,a mean absolute error(MAE)of 0.0005,and an R-squared coefficient(R2)of 0.994.With these results,the proposed prediction model has demonstrated its efficiency over the models proposed in previous studies.The findings of this study have considerable implications in the business field,as the proposed model can assist investors and traders in precisely identifying Bitcoin sales and buying opportunities.

A Novel MegaBAT Optimized Intelligent Intrusion Detection System in Wireless Sensor Networks

Wireless Sensor Network(WSN),whichfinds as one of the major components of modern electronic and wireless systems.A WSN consists of numerous sensor nodes for the discovery of sensor networks to leverage features like data sensing,data processing,and communication.In thefield of medical health care,these network plays a very vital role in transmitting highly sensitive data from different geographic regions and collecting this information by the respective network.But the fear of different attacks on health care data typically increases day by day.In a very short period,these attacks may cause adversarial effects to the WSN nodes.Furthermore,the existing Intrusion Detection System(IDS)suffers from the drawbacks of limited resources,low detection rate,and high computational overhead and also increases the false alarm rates in detecting the different attacks.Given the above-mentioned problems,this paper proposes the novel MegaBAT optimized Long Short Term Memory(MBOLT)-IDS for WSNs for the effective detection of different attacks.In the proposed framework,hyperpara-meters of deep Long Short-Term Memory(LSTM)were optimized by the meta-heuristic megabat algorithm to obtain a low computational overhead and high performance.The experimentations have been carried out using(Wireless Sensor NetworkDetection System)WSN-DS datasets and performance metrics such as accuracy,recall,precision,specificity,and F1-score are calculated and compared with the other existing intelligent IDS.The proposed framework provides outstanding results in detecting the black hole,gray hole,scheduling,flooding attacks and significantly reduces the time complexity,which makes this system suitable for resource-constraint WSNs.

结合LSTM自编码器与集成学习的井漏智能识别方法

为了解决传统的井漏智能识别模型因井漏样本数量受限导致其识别准确率低的问题,提出了一种长短期记忆(long short-term memory,LSTM)网络与自编码器(auto-encoder,AE)相结合、集成LSTM-AE的井漏智能识别方法。首先,采用正常样本训练多个包含不同隐藏层神经元数目的LSTM-AE模型,利用重构得分筛选出识别效果较好的几个模型作为基识别器;然后,采用集成学习对多个基识别器的识别结果进行加权融合,解决单一模型因对样本局部特征过度学习导致的误报与漏报问题,提高模型的识别准确率。从某油田18口井的钻井数据中选取了6000组正常钻进状态下的立压、出口流量、池体积数据,对集成LSTM-AE模型进行训练和测试,结果表明,提出方法的识别准确率达到了94.7%,优于其他常用的智能模型的识别结果,为井漏识别提供了一种新的技术途径。

基于改进麻雀搜索算法优化LSTM的滚动轴承故障诊断

为了对滚动轴承的工作状态及故障类别进行准确的诊断,本文采用长短时记忆(LSTM)神经网络作为分类器对滚动轴承数据集进行分类诊断。首先,从滚动轴承原始运行振动信号中提取时域和频域特征参数,组成具有高维特征参数的数据集;使用核主成分分析(KPCA)方法对高维特征集进行降维处理,选取重要性程度高的特征构成输入特征向量。然后,针对LSTM神经网络在滚动轴承故障诊断中存在的超参数难以确定的问题,提出一种基于自适应t分布策略的麻雀搜索算法优化LSTM神经网络的故障诊断方法(tSSA–LSTM)。最后,使用凯斯西储大学滚动轴承数据中心的数据进行故障诊断精度测试、泛化性能测试及噪声环境下故障诊断性能测试等多个仿真实验,并将本文提出的诊断模型与麻雀搜索算法优化长短时记忆神经网络(SSA–LSTM)、遗传算法优化长短时记忆神经网络(GA–LSTM)、粒子群算法优化长短时记忆神经网络(PSO–LSTM)及传统LSTM诊断模型进行对比。结果表明:tSSA可以更有效地对LSTM的隐含层神经元数量、周期次数、学习率等超参数进行合理优化;所提方法的平均诊断准确率达到98.86%,交叉验证平均诊断结果为98.57%;所提方法在噪声干扰下的故障诊断准确率也优于对比方法。因此,本文提出的tSSA–LSTM模型不仅可以更精准地诊断滚动轴承故障状态,而且具有更强的泛化能力及抗干扰能力,有效地提高了滚动轴承故障诊断的性能。

基于字词向量融合的民航智慧监管短文本分类

为解决民航监管事项所产生的检查记录仅依靠人工进行分类分析导致效率低的问题,提出一种基于数据增强与字词向量融合的双通道特征提取的短文本分类模型,探讨民航监管事项的分类,包括与人、设备设施环境、制度程序和机构职责等相关问题。为解决类别不平衡问题,采用数据增强算法在原始文本上进行变换,生成新的样本,使各个类别的样本数量更加均衡。将字向量和词向量按字融合拼接,得到具有词特征信息的字向量。将字词融合的向量分别送入到文本卷积神经网络(TextCNN)和双向长短期记忆(BiLSTM)模型中进行不同维度的特征提取,从局部的角度和全局的角度分别提取特征,并在民航监管事项检查记录数据集上进行试验。结果表明:该模型准确率为0.9837,F 1值为0.9836。与一些字嵌入模型和词嵌入模型相对比,准确率提升0.4%。和一些常用的单通道模型相比,准确率提升3%,验证了双通道模型提取的特征具有全面性和有效性。

融合多源异构气象数据的光伏功率预测模型

高精度光伏功率预测对提高电力系统运行效率具有重要意义。光伏功率受多种因素影响,其中云层的变化是最主要的不确定因素。传统光伏功率预测方法没有充分考虑云的3维结构和气象要素对光伏功率的影响。因此,该文提出一种融合多源异构气象数据的多源变量光伏功率预测模型(MPPM)。MPPM的核心包括时空条件扩散模型(STCDM)、注意力堆叠LSTM网络(ASLSTM)和多维特征融合模块(MFFM)。STCDM模型通过对2维卫星云图进行精确预测,消除了云层边界处的模糊现象。ASLSTM模型则提取了3维天气研究与预报模式(WRF)气象要素特征。MFFM模块将2维卫星云图特征和3维WRF气象要素特征进行融合,以得到未来1 h光伏功率预测结果。该文分别利用STCDM模型和MPPM模型开展卫星云图预测实验和光伏功率预测实验。实验结果显示,STCDM模型预测1 h内卫星云图的结构相似性指数(SSIM)达到0.914,MPPM模型预测1 h内光伏功率的相关系数(CORR)达到0.949,优于所有对比算法。

结合太阳辐射量计算与CNN-LSTM组合的光伏功率预测方法研究

为了提高模型预测性能,提出一种综合太阳辐射模型及深度学习的光伏功率预测模型。首先,利用太阳辐射机理建立太阳辐射模型(SRM),估算出水平面上总辐射值,再由斜面辐照度转换方法计算出光伏组件所接收的斜面辐射值。其次,通过皮尔逊相关分析法筛选出对光伏功率影响较大的主要因素,将斜面辐射计算值及主要影响因素作为输入,采用卷积神经网络(CNN)和长短期记忆网络(LSTM)建立光伏功率SRM-CNN-LSTM预测模型。分别利用春夏秋冬四季典型日的数据开展对比实验,结果表明:与几种其他方法相比,该文方法具有更好的预测效果。