利用长短期记忆网络LSTM对赤道太平洋海表面温度短期预报

海表面温度作为海洋中一个最重要的变量,对全球气候、海洋生态等有很大的影响,因此十分有必要对海表面温度(SST)进行预报。深度学习具备高效的数据处理能力,但目前利用深度学习对整个赤道太平洋的SST短期预报及预报技巧的研究仍较少。本文基于最优插值海表面温度(OISST)的日平均SST数据,利用长短期记忆(LSTM)网络构建了未来10天赤道太平洋(10°S~10°N,120°E~80°W)SST的逐日预报模型。LSTM预报模型利用1982~2010年的观测数据进行训练,2011~2020年的观测数据作为初值进行预报和检验评估。结果表明:赤道太平洋东部地区预报均方根误差(RMSE)大于中、西部,东部预报第1天RMSE为0.6℃左右,而中、西部均小于0.3℃。在不同的年际变化位相,预报RMSE在拉尼娜出现时期最大,正常年份次之,厄尔尼诺时期最小,RMSE在拉尼娜时期比在厄尔尼诺时期可达20%。预报偏差整体表现为东正、西负。相关预报技巧上,中部最好,可预报天数基本为10天以上,赤道冷舌附近可预报天数为4~7天,赤道西边部分地区可预报天数为3天。预报模型在赤道太平洋东部地区各月份预报技巧普遍低于西部地区,相比较而言各区域10、11月份预报技巧最低。总的来说,基于LSTM构建的SST预报模型能很好地捕捉到SST在时序上的演变特征,在不同案例中预报表现良好。同时该预报模型依靠数据驱动,能迅速且较好地预报未来10天以内的日平均SST的短期变化。

Device Anomaly Detection Algorithm Based on Enhanced Long Short-Term Memory Network

The problems in equipment fault detection include data dimension explosion,computational complexity,low detection accuracy,etc.To solve these problems,a device anomaly detection algorithm based on enhanced long short-term memory(LSTM)is proposed.The algorithm first reduces the dimensionality of the device sensor data by principal component analysis(PCA),extracts the strongly correlated variable data among the multidimensional sensor data with the lowest possible information loss,and then uses the enhanced stacked LSTM to predict the extracted temporal data,thus improving the accuracy of anomaly detection.To improve the efficiency of the anomaly detection,a genetic algorithm(GA)is used to adjust the magnitude of the enhancements made by the LSTM model.The validation of the actual data from the pumps shows that the algorithm has significantly improved the recall rate and the detection speed of device anomaly detection,with the recall rate of 97.07%,which indicates that the algorithm is effective and efficient for device anomaly detection in the actual production environment.

A multi-source information fusion layer counting method for penetration fuze based on TCN-LSTM

When employing penetration ammunition to strike multi-story buildings,the detection methods using acceleration sensors suffer from signal aliasing,while magnetic detection methods are susceptible to interference from ferromagnetic materials,thereby posing challenges in accurately determining the number of layers.To address this issue,this research proposes a layer counting method for penetration fuze that incorporates multi-source information fusion,utilizing both the temporal convolutional network(TCN)and the long short-term memory(LSTM)recurrent network.By leveraging the strengths of these two network structures,the method extracts temporal and high-dimensional features from the multi-source physical field during the penetration process,establishing a relationship between the multi-source physical field and the distance between the fuze and the target plate.A simulation model is developed to simulate the overload and magnetic field of a projectile penetrating multiple layers of target plates,capturing the multi-source physical field signals and their patterns during the penetration process.The analysis reveals that the proposed multi-source fusion layer counting method reduces errors by 60% and 50% compared to single overload layer counting and single magnetic anomaly signal layer counting,respectively.The model's predictive performance is evaluated under various operating conditions,including different ratios of added noise to random sample positions,penetration speeds,and spacing between target plates.The maximum errors in fuze penetration time predicted by the three modes are 0.08 ms,0.12 ms,and 0.16 ms,respectively,confirming the robustness of the proposed model.Moreover,the model's predictions indicate that the fitting degree for large interlayer spacings is superior to that for small interlayer spacings due to the influence of stress waves.

Real-time UAV path planning based on LSTM network

To address the shortcomings of single-step decision making in the existing deep reinforcement learning based unmanned aerial vehicle(UAV)real-time path planning problem,a real-time UAV path planning algorithm based on long shortterm memory(RPP-LSTM)network is proposed,which combines the memory characteristics of recurrent neural network(RNN)and the deep reinforcement learning algorithm.LSTM networks are used in this algorithm as Q-value networks for the deep Q network(DQN)algorithm,which makes the decision of the Q-value network has some memory.Thanks to LSTM network,the Q-value network can use the previous environmental information and action information which effectively avoids the problem of single-step decision considering only the current environment.Besides,the algorithm proposes a hierarchical reward and punishment function for the specific problem of UAV real-time path planning,so that the UAV can more reasonably perform path planning.Simulation verification shows that compared with the traditional feed-forward neural network(FNN)based UAV autonomous path planning algorithm,the RPP-LSTM proposed in this paper can adapt to more complex environments and has significantly improved robustness and accuracy when performing UAV real-time path planning.

Track correlation algorithm based on CNN-LSTM for swarm targets

The rapid development of unmanned aerial vehicle(UAV) swarm, a new type of aerial threat target, has brought great pressure to the air defense early warning system. At present, most of the track correlation algorithms only use part of the target location, speed, and other information for correlation.In this paper, the artificial neural network method is used to establish the corresponding intelligent track correlation model and method according to the characteristics of swarm targets.Precisely, a route correlation method based on convolutional neural networks (CNN) and long short-term memory (LSTM)Neural network is designed. In this model, the CNN is used to extract the formation characteristics of UAV swarm and the spatial position characteristics of single UAV track in the formation,while the LSTM is used to extract the time characteristics of UAV swarm. Experimental results show that compared with the traditional algorithms, the algorithm based on CNN-LSTM neural network can make full use of multiple feature information of the target, and has better robustness and accuracy for swarm targets.

基于LSTM-AEKF算法的锂离子电池SOC估计

针对扩展卡尔曼滤波(Extended Kalman filter,EKF)算法与长短期记忆网络(Long Short-Term Memory,LSTM)不能准确估计锂离子电池荷电状态(State of Charge,SOC)的问题,本文提出了一种基于二阶戴维宁(Thevenin)的等效电路模型,采用自适应扩展卡尔曼滤波(Adaptve Extended Kalman filter,AEKF)与LSTM相结合的SOC估计算法,即LSTM-AEKF算法。在二阶RC等效电路模型的基础上建立整数阶模型,并采用EKF算法辨识模型参数,采用LSTM-AEKF算法估计SOC,与AEKF算法、LSTM算法进行比较。根据马里兰大学公开数据集进行测试,结果表明,与传统方法相比,LSTM-AEKF算法估计SOC的平均绝对误差(Mean Absolute Error,MAE)与均方根误差(Root Mean Square Error,RMSE)分别下降了1.23%、1.5%,基于二阶RC模型的LSTM-AEKF算法可以有效估计SOC。

Navigation jamming signal recognition based on long short-term memory neural networks

This paper introduces the time-frequency analyzed long short-term memory(TF-LSTM) neural network method for jamming signal recognition over the Global Navigation Satellite System(GNSS) receiver. The method introduces the long shortterm memory(LSTM) neural network into the recognition algorithm and combines the time-frequency(TF) analysis for signal preprocessing. Five kinds of navigation jamming signals including white Gaussian noise(WGN), pulse jamming, sweep jamming, audio jamming, and spread spectrum jamming are used as input for training and recognition. Since the signal parameters and quantity are unknown in the actual scenario, this work builds a data set containing multiple kinds and parameters jamming to train the TF-LSTM. The performance of this method is evaluated by simulations and experiments. The method has higher recognition accuracy and better robustness than the existing methods, such as LSTM and the convolutional neural network(CNN).

LSTM-DPPO based deep reinforcement learning controller for path following optimization of unmanned surface vehicle

To solve the path following control problem for unmanned surface vehicles(USVs),a control method based on deep reinforcement learning(DRL)with long short-term memory(LSTM)networks is proposed.A distributed proximal policy opti-mization(DPPO)algorithm,which is a modified actor-critic-based type of reinforcement learning algorithm,is adapted to improve the controller performance in repeated trials.The LSTM network structure is introduced to solve the strong temporal cor-relation USV control problem.In addition,a specially designed path dataset,including straight and curved paths,is established to simulate various sailing scenarios so that the reinforcement learning controller can obtain as much handling experience as possible.Extensive numerical simulation results demonstrate that the proposed method has better control performance under missions involving complex maneuvers than trained with limited scenarios and can potentially be applied in practice.

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.

Motion simulation of moorings using optimized LSTM neural network

Mooring arrays have been widely deployed in sustained ocean observation in high resolution to measure finer dynamic features of marine phenomena.However,the irregular posture changes and nonlinear response of moorings under the effect of ocean currents face huge challenges for the deployment of mooring arrays,which may cause the deviations of measurements and yield a vacuum of observation in the upper ocean.We developed a data-driven mooring simulation model based on LSTM(long short-term memory)neural network,coupling the ocean current with position data from moorings to predict the motion of moorings,including single-step output prediction and multi-step prediction.Based on the predictive information,the formation of the mooring array can be adjusted to improve the accuracy and integrity of measurements.Moreover,we proposed the cuckoo search(CS)optimization algorithm to tune the parameters of LSTM,which improves the robustness and generalization of the model.We utilize the datasets observed from moorings anchored in the Kuroshio Extension region to train and validate the simulation model.The experimental results demonstrate that the model can remarkably improve prediction accuracy and yield stable performance.Moreover,compared with other optimization algorithms,CS is more efficient and performs better in simulating the motion of moorings.

基于注意力机制的LSTM时变水声信道深度学习预测

This paper investigates the channel prediction algorithm of the time-varying channels in underwater acoustic(UWA)communication systems using the long short-term memory(LSTM)model with the attention mechanism.AttLstmPreNet is a deep learning model that combines an attention mechanism with LSTM-type models to capture temporal information with different scales from historical UWA channels.The attention mechanism is used to capture sparsity in the time-delay scales and coherence in the gep-time scale under the LSTM framework.The soft attention mechanism is introduced before the LSTM to support the model to focus on the features of input sequences and help improve the learning capacity of the proposed model.The performance of the proposed model is validated using different simulation time-varying UWA channels.Compared with the adaptive channel predictors and the plain LSTM model,the proposed model is better in terms of channel prediction accuracy.

Maximizing Supermarket Profits:Data-Driven Strategies for Pricing,Sales,and Forecasting

The actual circumstances of daily life are crucial for the purchasing and pricing strategies of supermarkets.Developing strategies based on these circumstances can assist businesses in ensuring profits and fostering win-win cooperation.This paper explores methods to maximize profit through purchasing and sales strategies.Initially,the relevant data for various categories of vegetables is integrated.Through histograms,their sales patterns are directly understood,highlighting the most popular vegetables.Upon analyzing each vegetable category,it becomes evident that their sales data do not conform to normal distributions.Therefore,Spearman correlation coefficients are calculated,revealing strong correlations between certain categories,such as aquatic roots and edible fungi.A line chart depicting the top ten selling vegetables indicates a noticeable periodicity.Traditional fitting methods struggle to adequately model the sales of each vegetable category and their relationship with cost-plus pricing.To address this,additional factors such as holidays,weeks,and months are incorporated using techniques like random forest regression.This approach yields cost-plus pricing dependence curves that better capture the relationship,while effectively managing noise.Regarding sales volume prediction,the original data displays significant volatility,necessitating the handling of outliers using the threshold method.For missing data,linear interpolation is employed to mitigate the impact of continuous missing values on prediction accuracy.Subsequently,Adam-optimized long short-term memory(LSTM)networks are utilized to forecast incoming quantities for the next seven days.By extrapolating from normal sales volume,market capacity is estimated,allowing for additional sales through discount strategies.This framework has the potential to increase original income by 1.1 times.

基于LSTM与XGBOOST混合模型的孕妇产后出血预测

孕妇产后大出血是造成全球孕妇死亡的重要因素之一,在我国位居孕妇死亡原因首位,然而对产后出血的提前判定一直以来都是医学上一个难题.电子病历的普及,以及机器学习和深度学习技术的发展,为预知孕妇产后大出血提供了基于大数据的解决办法.本文提出利用孕妇的电子病历数据,构建基于LSTM和XGBoost的混合模型来预测孕妇产后大出血.实验结果表明,利用基于LSTM和XGBoost的混合模型对孕妇产后大出血进行预测是可行的,能够为医生判断孕妇产后出血情况提供参考,为孕妇分娩时是否需要备血方案提供决策支持,对降低产后大出血致死率具有积极意义.

LSTM neural network for solar radio spectrum classification

A solar radio spectrometer records solar radio radiation in the radio waveband. Such solar radio radiation spanning multiple frequency channels and over a short time period could provide a solar radio spectrum which is a two dimensional image. The vertical axis of a spectrum represents frequency channel and the horizontal axis signifies time. Intrinsically, time dependence exists between neighboring columns of a spectrum since solar radio radiation varies continuously over time. Thus, a spectrum can be treated as a time series consisting of all columns of a spectrum, while treating it as a general image would lose its time series property. A recurrent neural network(RNN) is designed for time series analysis. It can explore the correlation and interaction between neighboring inputs of a time series by augmenting a loop in a network.This paper makes the first attempt to utilize an RNN, specifically long short-term memory(LSTM), for solar radio spectrum classification. LSTM can mine well the context of a time series to acquire more information beyond a non-time series model. As such, as demonstrated by our experimental results, LSTM can learn a better representation of a spectrum, and thus contribute better classification.

Dynamic Hand Gesture Recognition Based on Short-Term Sampling Neural Networks

Hand gestures are a natural way for human-robot interaction.Vision based dynamic hand gesture recognition has become a hot research topic due to its various applications.This paper presents a novel deep learning network for hand gesture recognition.The network integrates several well-proved modules together to learn both short-term and long-term features from video inputs and meanwhile avoid intensive computation.To learn short-term features,each video input is segmented into a fixed number of frame groups.A frame is randomly selected from each group and represented as an RGB image as well as an optical flow snapshot.These two entities are fused and fed into a convolutional neural network(Conv Net)for feature extraction.The Conv Nets for all groups share parameters.To learn longterm features,outputs from all Conv Nets are fed into a long short-term memory(LSTM)network,by which a final classification result is predicted.The new model has been tested with two popular hand gesture datasets,namely the Jester dataset and Nvidia dataset.Comparing with other models,our model produced very competitive results.The robustness of the new model has also been proved with an augmented dataset with enhanced diversity of hand gestures.

基于LSTM神经网络的SQL注入攻击检测研究

由于Web技术的不断发展,尤其是5G技术的逐渐普及,万物联网的时代已经到来.网络的便利让数据的流通更加的及时和有价值,各种Web应用不仅方便了人们的日常生活,推进了社会的进步,更是带来了巨大的经济效益.因此,许多不法分子以攻击他人Web应用的方式来获取盈利,而作为OWASP(Open WebApplication Security Project)公布的报告中排名第一的注入类漏洞,是不法分子最常攻击的方向,带来了巨大的安全威胁.近年来,众多安全研究者都着力于找寻出更高效,准确度更高的方式来防御SQL注入攻击,本文将会对当前研究状况进行总结分析,并提出一种新基于LSTM神经网络的SQL注入语句分类识别模型.

基于改进LSTM网络的犯罪态势预测方法

为了利用历史数据对犯罪态势进行更加准确的预测,提出一种基于改进长短期记忆(long short-term memory,LSTM)网络的犯罪态势预测方法.首先统计某区域在每一个时间步长内发生犯罪事件的数量,作为一个时间步长值,再由多个时间步长组成一个时间序列,结合均方差滤波对统计的序列数据做标准化处理.其次建立包括输入层、隐藏层、全连接层和输出层的LSTM网络,在训练阶段将以上一段时间步长的预测值作为输入改为以实际值作为输入,根据修正的网络参数循环进行后续的预测,再对网络输出进行标准化逆处理得到预测结果.将2016年美国洛杉矶地区统计的全部犯罪记录作为实验数据,得到了态势拟合度较高的实验结果,与改进前相比,预测结果的均方根误差(root mean square error,RMSE)从139.65降低到了85.88,验证了基于改进LSTM网络对犯罪态势预测的有效性和准确性,并且通过与其他现有方法的对比,进一步证明了本方法在时间性能和准确性上的优越性.

基于简化型LSTM神经网络的时间序列预测方法

针对标准长短期记忆(long short-term memory,LSTM)神经网络用于时间序列预测具有耗时长、复杂度高等问题,提出简化型LSTM神经网络并应用于时间序列预测.首先,通过耦合输入门与遗忘门实现对标准LSTM神经网络的结构简化;其次,从门结构控制方程中消除输入信号与偏差实现进一步精简;然后,采用梯度下降算法更新简化型LSTM神经网络的参数;最后,通过2个时间序列基准数据集及污水处理过程出水生化需氧量(biochemical oxygen demand,BOD)质量浓度预测进行实验验证.结果表明:在不显著降低预测精度的情况下,所设计的模型能够缩短训练时间,减少LSTM神经网络的计算复杂度,实现时间序列的预测.

Behavior recognition based on the fusion of 3D-BN-VGG and LSTM network

In order to effectively solve the problems of low accuracy,large amount of computation and complex logic of deep learning algorithms in behavior recognition,a kind of behavior recognition based on the fusion of 3 dimensional batch normalization visual geometry group(3D-BN-VGG)and long short-term memory(LSTM)network is designed.In this network,3D convolutional layer is used to extract the spatial domain features and time domain features of video sequence at the same time,multiple small convolution kernels are stacked to replace large convolution kernels,thus the depth of neural network is deepened and the number of network parameters is reduced.In addition,the latest batch normalization algorithm is added to the 3-dimensional convolutional network to improve the training speed.Then the output of the full connection layer is sent to LSTM network as the feature vectors to extract the sequence information.This method,which directly uses the output of the whole base level without passing through the full connection layer,reduces the parameters of the whole fusion network to 15324485,nearly twice as much as those of 3D-BN-VGG.Finally,it reveals that the proposed network achieves 96.5%and 74.9%accuracy in the UCF-101 and HMDB-51 respectively,and the algorithm has a calculation speed of 1066 fps and an acceleration ratio of 1,which has a significant predominance in velocity.

A multi-target stance detection based on Bi-LSTM network with position-weight

In the task of multi-target stance detection,there are problems the mutual influence of content describing different targets,resulting in reduction in accuracy.To solve this problem,a multi-target stance detection algorithm based on a bidirectional long short-term memory(Bi-LSTM)network with position-weight is proposed.First,the corresponding position of the target in the input text is calculated with the ultimate position-weight vector.Next,the position information and output from the Bi-LSTM layer are fused by the position-weight fusion layer.Finally,the stances of different targets are predicted using the LSTM network and softmax classification.The multi-target stance detection corpus of the American election in 2016 is used to validate the proposed method.The results demonstrate that the Bi-LSTM network with position-weight achieves an advantage of 1.4%in macro average F1 value in the comparison of recent algorithms.