Graphics Processing Units(GPUs)are used to accelerate computing-intensive tasks,such as neural networks,data analysis,high-performance computing,etc.In the past decade or so,researchers have done a lot of work on GPU ...Graphics Processing Units(GPUs)are used to accelerate computing-intensive tasks,such as neural networks,data analysis,high-performance computing,etc.In the past decade or so,researchers have done a lot of work on GPU architecture and proposed a variety of theories and methods to study the microarchitectural characteristics of various GPUs.In this study,the GPU serves as a co-processor and works together with the CPU in an embedded real-time system to handle computationally intensive tasks.It models the architecture of the GPU and further considers it based on some excellent work.The SIMT mechanism and Cache-miss situation provide a more detailed analysis of the GPU architecture.In order to verify the GPU architecture model proposed in this article,10 GPU kernel_task and an Nvidia GPU device were used to perform experiments.The experimental results showed that the minimum error between the kernel task execution time predicted by the GPU architecture model proposed in this article and the actual measured kernel task execution time was 3.80%,and the maximum error was 8.30%.展开更多
As a category of recurrent neural networks,echo state networks(ESNs)have been the topic of in-depth investigations and extensive applications in a diverse array of fields,with spectacular triumphs achieved.Nevertheles...As a category of recurrent neural networks,echo state networks(ESNs)have been the topic of in-depth investigations and extensive applications in a diverse array of fields,with spectacular triumphs achieved.Nevertheless,the traditional ESN and the majority of its variants are devised in the light of the second-order statistical information of data(e.g.,variance and covariance),while more information is neglected.In the context of information theoretic learning,correntropy demonstrates the capacity to grab more information from data.Therefore,under the guidelines of the maximum correntropy criterion,this paper proposes a correntropy-based echo state network(CESN)in which the first-order and higher-order information of data is captured,promoting robustness to noise.Furthermore,an incremental learning algorithm for the CESN is presented,which has the expertise to update the CESN when new data arrives,eliminating the need to retrain the network from scratch.Finally,experiments on benchmark problems and comparisons with existing works are provided to verify the effectiveness and superiority of the proposed CESN.展开更多
Accurate wind speed measurements on maritime vessels are crucial for weather forecasting,sea state prediction,and safe navigation.However,vessel motion and challenging environmental conditions often affect measurement...Accurate wind speed measurements on maritime vessels are crucial for weather forecasting,sea state prediction,and safe navigation.However,vessel motion and challenging environmental conditions often affect measurement precision.To address this issue,this study proposes an innovative framework for correcting and predicting shipborne wind speed.By integrating a main network with a momentum updating network,the proposed framework effectively extracts features from the time and frequency domains,thereby allowing for precise adjustments and predictions of shipborne wind speed data.Validation using real sensor data collected at the Qingdao Oceanographic Institute demonstrates that the proposed method outperforms existing approaches in single-and multi-step predictions compared to existing methods,achieving higher accuracy in wind speed forecasting.The proposed innovative approach offers a promising direction for future validation in more realistic maritime onboard scenarios.展开更多
The chaotic characteristics and maximum predictable time scale of the observation series of hourly water consumption in Hangzhou were investigated using the advanced algorithm presented here is based on the convention...The chaotic characteristics and maximum predictable time scale of the observation series of hourly water consumption in Hangzhou were investigated using the advanced algorithm presented here is based on the conventional Wolf's algorithm for the largest Lyapunov exponent. For comparison, the largest Lyapunov exponents of water consumption series with one-hour and 24-hour intervals were calculated respectively. The results indicated that chaotic characteristics obviously exist in the hourly water consumption system; and that observation series with 24-hour interval have longer maximum predictable scale than hourly series. These findings could have significant practical application for better prediction of urban hourly water consumption.展开更多
Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stab...Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stability of reservoir banks changes with the long-term dynamics of external disastercausing factors.Thus,assessing the time-varying reliability of reservoir landslides remains a challenge.In this paper,a machine learning(ML)based approach is proposed to analyze the long-term reliability of reservoir bank landslides in spatially variable soils through time series prediction.This study systematically investigated the prediction performances of three ML algorithms,i.e.multilayer perceptron(MLP),convolutional neural network(CNN),and long short-term memory(LSTM).Additionally,the effects of the data quantity and data ratio on the predictive power of deep learning models are considered.The results show that all three ML models can accurately depict the changes in the time-varying failure probability of reservoir landslides.The CNN model outperforms both the MLP and LSTM models in predicting the failure probability.Furthermore,selecting the right data ratio can improve the prediction accuracy of the failure probability obtained by ML models.展开更多
Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep lear...Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep learning has largely contributed to the elevation of the prediction performance.Currently,the most up-to-date review of advanced machine learning techniques for financial time series prediction is still lacking,making it challenging for finance domain experts and relevant practitioners to determine which model potentially performs better,what techniques and components are involved,and how themodel can be designed and implemented.This review article provides an overview of techniques,components and frameworks for financial time series prediction,with an emphasis on state-of-the-art deep learning models in the literature from2015 to 2023,including standalonemodels like convolutional neural networks(CNN)that are capable of extracting spatial dependencies within data,and long short-term memory(LSTM)that is designed for handling temporal dependencies;and hybrid models integrating CNN,LSTM,attention mechanism(AM)and other techniques.For illustration and comparison purposes,models proposed in recent studies are mapped to relevant elements of a generalized framework comprised of input,output,feature extraction,prediction,and related processes.Among the state-of-the-artmodels,hybrid models like CNNLSTMand CNN-LSTM-AM in general have been reported superior in performance to stand-alone models like the CNN-only model.Some remaining challenges have been discussed,including non-friendliness for finance domain experts,delayed prediction,domain knowledge negligence,lack of standards,and inability of real-time and highfrequency predictions.The principal contributions of this paper are to provide a one-stop guide for both academia and industry to review,compare and summarize technologies and recent advances in this area,to facilitate smooth and informed implementation,and to highlight future research directions.展开更多
The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries an...The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other fields.Furthermore,it is important to construct a digital twin system.However,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted accuracy.In this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)layer.Specifically,we fuse the spatial-temporal graph based on the interrelationship of spatial graphs.Simultaneously,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding methods.Subsequently,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)module.The module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like data.AFSTGCN dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted accuracy.Experiments on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.展开更多
Purpose: This study aimed to enhance the prediction of container dwell time, a crucial factor for optimizing port operations, resource allocation, and supply chain efficiency. Determining an optimal learning rate for ...Purpose: This study aimed to enhance the prediction of container dwell time, a crucial factor for optimizing port operations, resource allocation, and supply chain efficiency. Determining an optimal learning rate for training Artificial Neural Networks (ANNs) has remained a challenging task due to the diverse sizes, complexity, and types of data involved. Design/Method/Approach: This research used a RandomizedSearchCV algorithm, a random search approach, to bridge this knowledge gap. The algorithm was applied to container dwell time data from the TOS system of the Port of Tema, which included 307,594 container records from 2014 to 2022. Findings: The RandomizedSearchCV method outperformed standard training methods both in terms of reducing training time and improving prediction accuracy, highlighting the significant role of the constant learning rate as a hyperparameter. Research Limitations and Implications: Although the study provides promising outcomes, the results are limited to the data extracted from the Port of Tema and may differ in other contexts. Further research is needed to generalize these findings across various port systems. Originality/Value: This research underscores the potential of RandomizedSearchCV as a valuable tool for optimizing ANN training in container dwell time prediction. It also accentuates the significance of automated learning rate selection, offering novel insights into the optimization of container dwell time prediction, with implications for improving port efficiency and supply chain operations.展开更多
Traffic forecasting with high precision aids Intelligent Transport Systems(ITS)in formulating and optimizing traffic management strategies.The algorithms used for tuning the hyperparameters of the deep learning models...Traffic forecasting with high precision aids Intelligent Transport Systems(ITS)in formulating and optimizing traffic management strategies.The algorithms used for tuning the hyperparameters of the deep learning models often have accurate results at the expense of high computational complexity.To address this problem,this paper uses the Tree-structured Parzen Estimator(TPE)to tune the hyperparameters of the Long Short-term Memory(LSTM)deep learning framework.The Tree-structured Parzen Estimator(TPE)uses a probabilistic approach with an adaptive searching mechanism by classifying the objective function values into good and bad samples.This ensures fast convergence in tuning the hyperparameter values in the deep learning model for performing prediction while still maintaining a certain degree of accuracy.It also overcomes the problem of converging to local optima and avoids timeconsuming random search and,therefore,avoids high computational complexity in prediction accuracy.The proposed scheme first performs data smoothing and normalization on the input data,which is then fed to the input of the TPE for tuning the hyperparameters.The traffic data is then input to the LSTM model with tuned parameters to perform the traffic prediction.The three optimizers:Adaptive Moment Estimation(Adam),Root Mean Square Propagation(RMSProp),and Stochastic Gradient Descend with Momentum(SGDM)are also evaluated for accuracy prediction and the best optimizer is then chosen for final traffic prediction in TPE-LSTM model.Simulation results verify the effectiveness of the proposed model in terms of accuracy of prediction over the benchmark schemes.展开更多
Accurate prediction of cloud resource utilization is critical.It helps improve service quality while avoiding resource waste and shortages.However,the time series of resource usage in cloud computing systems often exh...Accurate prediction of cloud resource utilization is critical.It helps improve service quality while avoiding resource waste and shortages.However,the time series of resource usage in cloud computing systems often exhibit multidimensionality,nonlinearity,and high volatility,making the high-precision prediction of resource utilization a complex and challenging task.At present,cloud computing resource prediction methods include traditional statistical models,hybrid approaches combining machine learning and classical models,and deep learning techniques.Traditional statistical methods struggle with nonlinear predictions,hybrid methods face challenges in feature extraction and long-term dependencies,and deep learning methods incur high computational costs.The above methods are insufficient to achieve high-precision resource prediction in cloud computing systems.Therefore,we propose a new time series prediction model,called SDVformer,which is based on the Informer model by integrating the Savitzky-Golay(SG)filters,a novel Discrete-Variation Self-Attention(DVSA)mechanism,and a type-aware mixture of experts(T-MOE)framework.The SG filter is designed to reduce noise and enhance the feature representation of input data.The DVSA mechanism is proposed to optimize the selection of critical features to reduce computational complexity.The T-MOE framework is designed to adjust the model structure based on different resource characteristics,thereby improving prediction accuracy and adaptability.Experimental results show that our proposed SDVformer significantly outperforms baseline models,including Recurrent Neural Network(RNN),Long Short-Term Memory(LSTM),and Informer in terms of prediction precision,on both the Alibaba public dataset and the dataset collected by Beijing Jiaotong University(BJTU).Particularly compared with the Informer model,the average Mean Squared Error(MSE)of SDVformer decreases by about 80%,fully demonstrating its advantages in complex time series prediction tasks in cloud computing systems.展开更多
Traffic datasets exhibit complex spatiotemporal characteristics,including significant fluctuations in traffic volume and intricate periodical patterns,which pose substantial challenges for the accurate forecasting and...Traffic datasets exhibit complex spatiotemporal characteristics,including significant fluctuations in traffic volume and intricate periodical patterns,which pose substantial challenges for the accurate forecasting and effective management of traffic conditions.Traditional forecasting models often struggle to adequately capture these complexities,leading to suboptimal predictive performance.While neural networks excel at modeling intricate and nonlinear data structures,they are also highly susceptible to overfitting,resulting in inefficient use of computational resources and decreased model generalization.This paper introduces a novel heuristic feature extraction method that synergistically combines the strengths of non-neural network algorithms with neural networks to enhance the identification and representation of relevant features from traffic data.We begin by evaluating the significance of various temporal characteristics using three distinct assessment strategies grounded in non-neural methodologies.These evaluated features are then aggregated through a weighted fusion mechanism to create heuristic features,which are subsequently integrated into neural network models for more accurate and robust traffic prediction.Experimental results derived from four real-world datasets,collected from diverse urban environments,show that the proposed method significantly improves the accuracy of long-term traffic forecasting without compromising performance.Additionally,the approach helps streamline neural network architectures,leading to a considerable reduction in computational overhead.By addressing both prediction accuracy and computational efficiency,this study not only presents an innovative and effective method for traffic condition forecasting but also offers valuable insights that can inform the future development of data-driven traffic management systems and transportation strategies.展开更多
Time series prediction has been successfully used in several application areas, such as meteoro-logical forecasting, market prediction, network traffic forecasting, etc. , and a number of techniques have been develop...Time series prediction has been successfully used in several application areas, such as meteoro-logical forecasting, market prediction, network traffic forecasting, etc. , and a number of techniques have been developed for modeling and predicting time series. In the traditional exponential smoothing method, a fixed weight is assigned to data history, and the trend changes of time series are ignored. In this paper, an uncertainty reasoning method, based on cloud model, is employed in time series prediction, which uses cloud logic controller to adjust the smoothing coefficient of the simple exponential smoothing method dynamically to fit the current trend of the time series. The validity of this solution was proved by experiments on various data sets.展开更多
To investigate the travel time prediction method of the freeway, a model based on the gradient boosting decision tree (GBDT) is proposed. Eleven variables (namely, travel time in current period T i , traffic flow in c...To investigate the travel time prediction method of the freeway, a model based on the gradient boosting decision tree (GBDT) is proposed. Eleven variables (namely, travel time in current period T i , traffic flow in current period Q i , speed in current period V i , density in current period K i , the number of vehicles in current period N i , occupancy in current period R i , traffic state parameter in current period X i , travel time in previous time period T i -1 , etc.) are selected to predict the travel time for 10 min ahead in the proposed model. Data obtained from VISSIM simulation is used to train and test the model. The results demonstrate that the prediction error of the GBDT model is smaller than those of the back propagation (BP) neural network model and the support vector machine (SVM) model. Travel time in current period T i is the most important variable among all variables in the GBDT model. The GBDT model can produce more accurate prediction results and mine the hidden nonlinear relationships deeply between variables and the predicted travel time.展开更多
Deficiencies of applying the traditional least squares support vector machine （LS-SVM） to time series online prediction were specified. According to the kernel function matrix＇s property and using the recursive cal...Deficiencies of applying the traditional least squares support vector machine （LS-SVM） to time series online prediction were specified. According to the kernel function matrix＇s property and using the recursive calculation of block matrix, a new time series online prediction algorithm based on improved LS-SVM was proposed. The historical training results were fully utilized and the computing speed of LS-SVM was enhanced. Then, the improved algorithm was applied to timc series online prediction. Based on the operational data provided by the Northwest Power Grid of China, the method was used in the transient stability prediction of electric power system. The results show that, compared with the calculation time of the traditional LS-SVM（75 1 600 ms）, that of the proposed method in different time windows is 40-60 ms, proposed method is above 0.8. So the improved method is online prediction. and the prediction accuracy（normalized root mean squared error） of the better than the traditional LS-SVM and more suitable for time series online prediction.展开更多
How to predict the bus arrival time accurately is a crucial problem to be solved in Internet of Vehicle. Existed methods cannot solve the problem effectively for ignoring the traffic delay jitter. In this paper,a thre...How to predict the bus arrival time accurately is a crucial problem to be solved in Internet of Vehicle. Existed methods cannot solve the problem effectively for ignoring the traffic delay jitter. In this paper,a three-stage mixed model is proposed for bus arrival time prediction. The first stage is pattern training. In this stage,the traffic delay jitter patterns(TDJP)are mined by K nearest neighbor and K-means in the historical traffic time data. The second stage is the single-step prediction,which is based on real-time adjusted Kalman filter with a modification of historical TDJP. In the third stage,as the influence of historical law is increasing in long distance prediction,we combine the single-step prediction dynamically with Markov historical transfer model to conduct the multi-step prediction. The experimental results show that the proposed single-step prediction model performs better in accuracy and efficiency than short-term traffic flow prediction and dynamic Kalman filter. The multi-step prediction provides a higher level veracity and reliability in travel time forecasting than short-term traffic flow and historical traffic pattern prediction models.展开更多
It is crucial to predict the outputs of a thickening system,including the underflow concentration(UC)and mud pressure,for optimal control of the process.The proliferation of industrial sensors and the availability of ...It is crucial to predict the outputs of a thickening system,including the underflow concentration(UC)and mud pressure,for optimal control of the process.The proliferation of industrial sensors and the availability of thickening-system data make this possible.However,the unique properties of thickening systems,such as the non-linearities,long-time delays,partially observed data,and continuous time evolution pose challenges on building data-driven predictive models.To address the above challenges,we establish an integrated,deep-learning,continuous time network structure that consists of a sequential encoder,a state decoder,and a derivative module to learn the deterministic state space model from thickening systems.Using a case study,we examine our methods with a tailing thickener manufactured by the FLSmidth installed with massive sensors and obtain extensive experimental results.The results demonstrate that the proposed continuous-time model with the sequential encoder achieves better prediction performances than the existing discrete-time models and reduces the negative effects from long time delays by extracting features from historical system trajectories.The proposed method also demonstrates outstanding performances for both short and long term prediction tasks with the two proposed derivative types.展开更多
Three functional models, polynomial, spectral analysis, and modified AR model, are studied and compared in fitting and predicting clock deviation based on the data sequence derived from two-way satellite time and freq...Three functional models, polynomial, spectral analysis, and modified AR model, are studied and compared in fitting and predicting clock deviation based on the data sequence derived from two-way satellite time and frequency transfer. A robust equivalent weight is applied, which controls the significant influence of outlying observations. Some conclusions show that the prediction precision of robust estimation is better than that of LS. The prediction precision calculated from smoothed observations is higher than that calculated from sampling observations. As a count of the obvious period variations in the clock deviation sequence, the predicted values of polynomial model are implausible. The prediction precision of spectral analysis model is very low, but the principal periods can be determined. The prediction RMS of 6-hour extrapolation interval is Ins or so, when modified AR model is used.展开更多
A DRNN (diagonal recurrent neural network) and its RPE (recurrent prediction error) learning algorithm are proposed in this paper .Using of the simple structure of DRNN can reduce the capacity of calculation. The prin...A DRNN (diagonal recurrent neural network) and its RPE (recurrent prediction error) learning algorithm are proposed in this paper .Using of the simple structure of DRNN can reduce the capacity of calculation. The principle of RPE learning algorithm is to adjust weights along the direction of Gauss-Newton. Meanwhile, it is unnecessary to calculate the second local derivative and the inverse matrixes, whose unbiasedness is proved. With application to the extremely short time prediction of large ship pitch, satisfactory results are obtained. Prediction effect of this algorithm is compared with that of auto-regression and periodical diagram method, and comparison results show that the proposed algorithm is feasible.展开更多
The prediction methods and its applications of the nonlinear dynamic systems determined from chaotic time series of low-dimension are discussed mainly. Based on the work of the foreign researchers, the chaotic time se...The prediction methods and its applications of the nonlinear dynamic systems determined from chaotic time series of low-dimension are discussed mainly. Based on the work of the foreign researchers, the chaotic time series in the phase space adopting one kind of nonlinear chaotic model were reconstructed. At first, the model parameters were estimated by using the improved least square method. Then as the precision was satisfied, the optimization method was used to estimate these parameters. At the end by using the obtained chaotic model, the future data of the chaotic time series in the phase space was predicted. Some representative experimental examples were analyzed to testify the models and the algorithms developed in this paper. ne results show that if the algorithms developed here are adopted, the parameters of the corresponding chaotic model will be easily calculated well and true. Predictions of chaotic series in phase space make the traditional methods change from outer iteration to interpolations. And if the optimal model rank is chosen, the prediction precision will increase notably. Long term superior predictability of nonlinear chaotic models is proved to be irrational and unreasonable.展开更多
The accurate prediction of travel time along roadway provides valuable traffic information for travelers and traffic managers. Aiming at short-term travel time forecasting on urban arterials,a prediction model( PSOSVM...The accurate prediction of travel time along roadway provides valuable traffic information for travelers and traffic managers. Aiming at short-term travel time forecasting on urban arterials,a prediction model( PSOSVM) combining support vector machine( SVM) and particle swarm optimization( PSO) is developed. Travel time data collected with Bluetooth devices are used to calibrate the proposed model. Field experiments show that the PSO-SVM model 's error indicators are lower than the single SVM model and the BP neural network( BPNN) model. Particularly,the mean-absolute percentage error( MAPE) of PSO-SVM is only 9. 453 4 %which is less than that of the single SVM model( 12. 230 2 %) and the BPNN model( 15. 314 7 %). The results indicate that the proposed PSO-SVM model is feasible and more effective than other models for shortterm travel time prediction on urban arterials.展开更多
文摘Graphics Processing Units(GPUs)are used to accelerate computing-intensive tasks,such as neural networks,data analysis,high-performance computing,etc.In the past decade or so,researchers have done a lot of work on GPU architecture and proposed a variety of theories and methods to study the microarchitectural characteristics of various GPUs.In this study,the GPU serves as a co-processor and works together with the CPU in an embedded real-time system to handle computationally intensive tasks.It models the architecture of the GPU and further considers it based on some excellent work.The SIMT mechanism and Cache-miss situation provide a more detailed analysis of the GPU architecture.In order to verify the GPU architecture model proposed in this article,10 GPU kernel_task and an Nvidia GPU device were used to perform experiments.The experimental results showed that the minimum error between the kernel task execution time predicted by the GPU architecture model proposed in this article and the actual measured kernel task execution time was 3.80%,and the maximum error was 8.30%.
基金supported in part by the National Natural Science Foundation of China(62176109,62476115)the Fundamental Research Funds for the Central Universities(lzujbky-2023-ey07,lzujbky-2023-it14)+1 种基金the Natural Science Foundation of Gansu Province(24JRRA488)the Supercomputing Center of Lanzhou University
文摘As a category of recurrent neural networks,echo state networks(ESNs)have been the topic of in-depth investigations and extensive applications in a diverse array of fields,with spectacular triumphs achieved.Nevertheless,the traditional ESN and the majority of its variants are devised in the light of the second-order statistical information of data(e.g.,variance and covariance),while more information is neglected.In the context of information theoretic learning,correntropy demonstrates the capacity to grab more information from data.Therefore,under the guidelines of the maximum correntropy criterion,this paper proposes a correntropy-based echo state network(CESN)in which the first-order and higher-order information of data is captured,promoting robustness to noise.Furthermore,an incremental learning algorithm for the CESN is presented,which has the expertise to update the CESN when new data arrives,eliminating the need to retrain the network from scratch.Finally,experiments on benchmark problems and comparisons with existing works are provided to verify the effectiveness and superiority of the proposed CESN.
基金supported by the Major Innovation Project for the Integration of Science,Education,and Industry of Qilu University of Technology(Shandong Academy of Sciences)(Nos.2023HYZX01,2023JBZ02)the Open Project of Key Laboratory of Computing Power Network and Information Security,Ministry of Education,Qilu University of Technology(Shandong Academy of Sciences)(No.2023ZD007)+2 种基金the Talent Research Projects of Qilu University of Technology(Shandong Academy of Sciences)(No.2023RCKY136)the Technology and Innovation Major Project of the Ministry of Science and Technology of China(No.2022ZD0118600)the Jinan‘20 New Colleges and Universities’Funded Project(No.202333043)。
文摘Accurate wind speed measurements on maritime vessels are crucial for weather forecasting,sea state prediction,and safe navigation.However,vessel motion and challenging environmental conditions often affect measurement precision.To address this issue,this study proposes an innovative framework for correcting and predicting shipborne wind speed.By integrating a main network with a momentum updating network,the proposed framework effectively extracts features from the time and frequency domains,thereby allowing for precise adjustments and predictions of shipborne wind speed data.Validation using real sensor data collected at the Qingdao Oceanographic Institute demonstrates that the proposed method outperforms existing approaches in single-and multi-step predictions compared to existing methods,achieving higher accuracy in wind speed forecasting.The proposed innovative approach offers a promising direction for future validation in more realistic maritime onboard scenarios.
基金Project (No. 50078048) supported by the National Natural Science Foundation of China
文摘The chaotic characteristics and maximum predictable time scale of the observation series of hourly water consumption in Hangzhou were investigated using the advanced algorithm presented here is based on the conventional Wolf's algorithm for the largest Lyapunov exponent. For comparison, the largest Lyapunov exponents of water consumption series with one-hour and 24-hour intervals were calculated respectively. The results indicated that chaotic characteristics obviously exist in the hourly water consumption system; and that observation series with 24-hour interval have longer maximum predictable scale than hourly series. These findings could have significant practical application for better prediction of urban hourly water consumption.
基金supported by the National Natural Science Foundation of China(Grant No.52308340)the Innovative Projects of Universities in Guangdong(Grant No.2022KTSCX208)Sichuan Transportation Science and Technology Project(Grant No.2018-ZL-01).
文摘Historically,landslides have been the primary type of geological disaster worldwide.Generally,the stability of reservoir banks is primarily affected by rainfall and reservoir water level fluctuations.Moreover,the stability of reservoir banks changes with the long-term dynamics of external disastercausing factors.Thus,assessing the time-varying reliability of reservoir landslides remains a challenge.In this paper,a machine learning(ML)based approach is proposed to analyze the long-term reliability of reservoir bank landslides in spatially variable soils through time series prediction.This study systematically investigated the prediction performances of three ML algorithms,i.e.multilayer perceptron(MLP),convolutional neural network(CNN),and long short-term memory(LSTM).Additionally,the effects of the data quantity and data ratio on the predictive power of deep learning models are considered.The results show that all three ML models can accurately depict the changes in the time-varying failure probability of reservoir landslides.The CNN model outperforms both the MLP and LSTM models in predicting the failure probability.Furthermore,selecting the right data ratio can improve the prediction accuracy of the failure probability obtained by ML models.
基金funded by the Natural Science Foundation of Fujian Province,China (Grant No.2022J05291)Xiamen Scientific Research Funding for Overseas Chinese Scholars.
文摘Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep learning has largely contributed to the elevation of the prediction performance.Currently,the most up-to-date review of advanced machine learning techniques for financial time series prediction is still lacking,making it challenging for finance domain experts and relevant practitioners to determine which model potentially performs better,what techniques and components are involved,and how themodel can be designed and implemented.This review article provides an overview of techniques,components and frameworks for financial time series prediction,with an emphasis on state-of-the-art deep learning models in the literature from2015 to 2023,including standalonemodels like convolutional neural networks(CNN)that are capable of extracting spatial dependencies within data,and long short-term memory(LSTM)that is designed for handling temporal dependencies;and hybrid models integrating CNN,LSTM,attention mechanism(AM)and other techniques.For illustration and comparison purposes,models proposed in recent studies are mapped to relevant elements of a generalized framework comprised of input,output,feature extraction,prediction,and related processes.Among the state-of-the-artmodels,hybrid models like CNNLSTMand CNN-LSTM-AM in general have been reported superior in performance to stand-alone models like the CNN-only model.Some remaining challenges have been discussed,including non-friendliness for finance domain experts,delayed prediction,domain knowledge negligence,lack of standards,and inability of real-time and highfrequency predictions.The principal contributions of this paper are to provide a one-stop guide for both academia and industry to review,compare and summarize technologies and recent advances in this area,to facilitate smooth and informed implementation,and to highlight future research directions.
基金supported by the China Scholarship Council and the CERNET Innovation Project under grant No.20170111.
文摘The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other fields.Furthermore,it is important to construct a digital twin system.However,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted accuracy.In this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)layer.Specifically,we fuse the spatial-temporal graph based on the interrelationship of spatial graphs.Simultaneously,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding methods.Subsequently,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)module.The module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like data.AFSTGCN dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted accuracy.Experiments on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.
文摘Purpose: This study aimed to enhance the prediction of container dwell time, a crucial factor for optimizing port operations, resource allocation, and supply chain efficiency. Determining an optimal learning rate for training Artificial Neural Networks (ANNs) has remained a challenging task due to the diverse sizes, complexity, and types of data involved. Design/Method/Approach: This research used a RandomizedSearchCV algorithm, a random search approach, to bridge this knowledge gap. The algorithm was applied to container dwell time data from the TOS system of the Port of Tema, which included 307,594 container records from 2014 to 2022. Findings: The RandomizedSearchCV method outperformed standard training methods both in terms of reducing training time and improving prediction accuracy, highlighting the significant role of the constant learning rate as a hyperparameter. Research Limitations and Implications: Although the study provides promising outcomes, the results are limited to the data extracted from the Port of Tema and may differ in other contexts. Further research is needed to generalize these findings across various port systems. Originality/Value: This research underscores the potential of RandomizedSearchCV as a valuable tool for optimizing ANN training in container dwell time prediction. It also accentuates the significance of automated learning rate selection, offering novel insights into the optimization of container dwell time prediction, with implications for improving port efficiency and supply chain operations.
文摘Traffic forecasting with high precision aids Intelligent Transport Systems(ITS)in formulating and optimizing traffic management strategies.The algorithms used for tuning the hyperparameters of the deep learning models often have accurate results at the expense of high computational complexity.To address this problem,this paper uses the Tree-structured Parzen Estimator(TPE)to tune the hyperparameters of the Long Short-term Memory(LSTM)deep learning framework.The Tree-structured Parzen Estimator(TPE)uses a probabilistic approach with an adaptive searching mechanism by classifying the objective function values into good and bad samples.This ensures fast convergence in tuning the hyperparameter values in the deep learning model for performing prediction while still maintaining a certain degree of accuracy.It also overcomes the problem of converging to local optima and avoids timeconsuming random search and,therefore,avoids high computational complexity in prediction accuracy.The proposed scheme first performs data smoothing and normalization on the input data,which is then fed to the input of the TPE for tuning the hyperparameters.The traffic data is then input to the LSTM model with tuned parameters to perform the traffic prediction.The three optimizers:Adaptive Moment Estimation(Adam),Root Mean Square Propagation(RMSProp),and Stochastic Gradient Descend with Momentum(SGDM)are also evaluated for accuracy prediction and the best optimizer is then chosen for final traffic prediction in TPE-LSTM model.Simulation results verify the effectiveness of the proposed model in terms of accuracy of prediction over the benchmark schemes.
文摘Accurate prediction of cloud resource utilization is critical.It helps improve service quality while avoiding resource waste and shortages.However,the time series of resource usage in cloud computing systems often exhibit multidimensionality,nonlinearity,and high volatility,making the high-precision prediction of resource utilization a complex and challenging task.At present,cloud computing resource prediction methods include traditional statistical models,hybrid approaches combining machine learning and classical models,and deep learning techniques.Traditional statistical methods struggle with nonlinear predictions,hybrid methods face challenges in feature extraction and long-term dependencies,and deep learning methods incur high computational costs.The above methods are insufficient to achieve high-precision resource prediction in cloud computing systems.Therefore,we propose a new time series prediction model,called SDVformer,which is based on the Informer model by integrating the Savitzky-Golay(SG)filters,a novel Discrete-Variation Self-Attention(DVSA)mechanism,and a type-aware mixture of experts(T-MOE)framework.The SG filter is designed to reduce noise and enhance the feature representation of input data.The DVSA mechanism is proposed to optimize the selection of critical features to reduce computational complexity.The T-MOE framework is designed to adjust the model structure based on different resource characteristics,thereby improving prediction accuracy and adaptability.Experimental results show that our proposed SDVformer significantly outperforms baseline models,including Recurrent Neural Network(RNN),Long Short-Term Memory(LSTM),and Informer in terms of prediction precision,on both the Alibaba public dataset and the dataset collected by Beijing Jiaotong University(BJTU).Particularly compared with the Informer model,the average Mean Squared Error(MSE)of SDVformer decreases by about 80%,fully demonstrating its advantages in complex time series prediction tasks in cloud computing systems.
基金supported by the Shandong Province Higher Education Young Innovative Talents Cultivation Programme Project:TJY2114Jinan City-School Integration Development Strategy Project:JNSX2023015the Natural Science Foundation of Shandong Province:ZR2021M F074.
文摘Traffic datasets exhibit complex spatiotemporal characteristics,including significant fluctuations in traffic volume and intricate periodical patterns,which pose substantial challenges for the accurate forecasting and effective management of traffic conditions.Traditional forecasting models often struggle to adequately capture these complexities,leading to suboptimal predictive performance.While neural networks excel at modeling intricate and nonlinear data structures,they are also highly susceptible to overfitting,resulting in inefficient use of computational resources and decreased model generalization.This paper introduces a novel heuristic feature extraction method that synergistically combines the strengths of non-neural network algorithms with neural networks to enhance the identification and representation of relevant features from traffic data.We begin by evaluating the significance of various temporal characteristics using three distinct assessment strategies grounded in non-neural methodologies.These evaluated features are then aggregated through a weighted fusion mechanism to create heuristic features,which are subsequently integrated into neural network models for more accurate and robust traffic prediction.Experimental results derived from four real-world datasets,collected from diverse urban environments,show that the proposed method significantly improves the accuracy of long-term traffic forecasting without compromising performance.Additionally,the approach helps streamline neural network architectures,leading to a considerable reduction in computational overhead.By addressing both prediction accuracy and computational efficiency,this study not only presents an innovative and effective method for traffic condition forecasting but also offers valuable insights that can inform the future development of data-driven traffic management systems and transportation strategies.
文摘Time series prediction has been successfully used in several application areas, such as meteoro-logical forecasting, market prediction, network traffic forecasting, etc. , and a number of techniques have been developed for modeling and predicting time series. In the traditional exponential smoothing method, a fixed weight is assigned to data history, and the trend changes of time series are ignored. In this paper, an uncertainty reasoning method, based on cloud model, is employed in time series prediction, which uses cloud logic controller to adjust the smoothing coefficient of the simple exponential smoothing method dynamically to fit the current trend of the time series. The validity of this solution was proved by experiments on various data sets.
基金The National Natural Science Foundation of China(No.51478114,51778136)
文摘To investigate the travel time prediction method of the freeway, a model based on the gradient boosting decision tree (GBDT) is proposed. Eleven variables (namely, travel time in current period T i , traffic flow in current period Q i , speed in current period V i , density in current period K i , the number of vehicles in current period N i , occupancy in current period R i , traffic state parameter in current period X i , travel time in previous time period T i -1 , etc.) are selected to predict the travel time for 10 min ahead in the proposed model. Data obtained from VISSIM simulation is used to train and test the model. The results demonstrate that the prediction error of the GBDT model is smaller than those of the back propagation (BP) neural network model and the support vector machine (SVM) model. Travel time in current period T i is the most important variable among all variables in the GBDT model. The GBDT model can produce more accurate prediction results and mine the hidden nonlinear relationships deeply between variables and the predicted travel time.
基金Project (SGKJ[200301-16]) supported by the State Grid Cooperation of China
文摘Deficiencies of applying the traditional least squares support vector machine （LS-SVM） to time series online prediction were specified. According to the kernel function matrix＇s property and using the recursive calculation of block matrix, a new time series online prediction algorithm based on improved LS-SVM was proposed. The historical training results were fully utilized and the computing speed of LS-SVM was enhanced. Then, the improved algorithm was applied to timc series online prediction. Based on the operational data provided by the Northwest Power Grid of China, the method was used in the transient stability prediction of electric power system. The results show that, compared with the calculation time of the traditional LS-SVM（75 1 600 ms）, that of the proposed method in different time windows is 40-60 ms, proposed method is above 0.8. So the improved method is online prediction. and the prediction accuracy（normalized root mean squared error） of the better than the traditional LS-SVM and more suitable for time series online prediction.
基金National Science and Technology Major Project(2016ZX03001025-003)Special Found for Beijing Common Construction Project
文摘How to predict the bus arrival time accurately is a crucial problem to be solved in Internet of Vehicle. Existed methods cannot solve the problem effectively for ignoring the traffic delay jitter. In this paper,a three-stage mixed model is proposed for bus arrival time prediction. The first stage is pattern training. In this stage,the traffic delay jitter patterns(TDJP)are mined by K nearest neighbor and K-means in the historical traffic time data. The second stage is the single-step prediction,which is based on real-time adjusted Kalman filter with a modification of historical TDJP. In the third stage,as the influence of historical law is increasing in long distance prediction,we combine the single-step prediction dynamically with Markov historical transfer model to conduct the multi-step prediction. The experimental results show that the proposed single-step prediction model performs better in accuracy and efficiency than short-term traffic flow prediction and dynamic Kalman filter. The multi-step prediction provides a higher level veracity and reliability in travel time forecasting than short-term traffic flow and historical traffic pattern prediction models.
基金supported by National Key Research and Development Program of China(2019YFC0605300)the National Natural Science Foundation of China(61873299,61902022,61972028)+2 种基金Scientific and Technological Innovation Foundation of Shunde Graduate School,University of Science and Technology Beijing(BK21BF002)Macao Science and Technology Development Fund under Macao Funding Scheme for Key R&D Projects(0025/2019/AKP)Macao Science and Technology Development Fund(0015/2020/AMJ)。
文摘It is crucial to predict the outputs of a thickening system,including the underflow concentration(UC)and mud pressure,for optimal control of the process.The proliferation of industrial sensors and the availability of thickening-system data make this possible.However,the unique properties of thickening systems,such as the non-linearities,long-time delays,partially observed data,and continuous time evolution pose challenges on building data-driven predictive models.To address the above challenges,we establish an integrated,deep-learning,continuous time network structure that consists of a sequential encoder,a state decoder,and a derivative module to learn the deterministic state space model from thickening systems.Using a case study,we examine our methods with a tailing thickener manufactured by the FLSmidth installed with massive sensors and obtain extensive experimental results.The results demonstrate that the proposed continuous-time model with the sequential encoder achieves better prediction performances than the existing discrete-time models and reduces the negative effects from long time delays by extracting features from historical system trajectories.The proposed method also demonstrates outstanding performances for both short and long term prediction tasks with the two proposed derivative types.
基金Supported by the National Natural Science Foundations of China （No. 40474001, No. 40274002, No. 40604003）.
文摘Three functional models, polynomial, spectral analysis, and modified AR model, are studied and compared in fitting and predicting clock deviation based on the data sequence derived from two-way satellite time and frequency transfer. A robust equivalent weight is applied, which controls the significant influence of outlying observations. Some conclusions show that the prediction precision of robust estimation is better than that of LS. The prediction precision calculated from smoothed observations is higher than that calculated from sampling observations. As a count of the obvious period variations in the clock deviation sequence, the predicted values of polynomial model are implausible. The prediction precision of spectral analysis model is very low, but the principal periods can be determined. The prediction RMS of 6-hour extrapolation interval is Ins or so, when modified AR model is used.
文摘A DRNN (diagonal recurrent neural network) and its RPE (recurrent prediction error) learning algorithm are proposed in this paper .Using of the simple structure of DRNN can reduce the capacity of calculation. The principle of RPE learning algorithm is to adjust weights along the direction of Gauss-Newton. Meanwhile, it is unnecessary to calculate the second local derivative and the inverse matrixes, whose unbiasedness is proved. With application to the extremely short time prediction of large ship pitch, satisfactory results are obtained. Prediction effect of this algorithm is compared with that of auto-regression and periodical diagram method, and comparison results show that the proposed algorithm is feasible.
文摘The prediction methods and its applications of the nonlinear dynamic systems determined from chaotic time series of low-dimension are discussed mainly. Based on the work of the foreign researchers, the chaotic time series in the phase space adopting one kind of nonlinear chaotic model were reconstructed. At first, the model parameters were estimated by using the improved least square method. Then as the precision was satisfied, the optimization method was used to estimate these parameters. At the end by using the obtained chaotic model, the future data of the chaotic time series in the phase space was predicted. Some representative experimental examples were analyzed to testify the models and the algorithms developed in this paper. ne results show that if the algorithms developed here are adopted, the parameters of the corresponding chaotic model will be easily calculated well and true. Predictions of chaotic series in phase space make the traditional methods change from outer iteration to interpolations. And if the optimal model rank is chosen, the prediction precision will increase notably. Long term superior predictability of nonlinear chaotic models is proved to be irrational and unreasonable.
基金Sponsored by the National Natural Science Foundation of China(Grant No.71101109)
文摘The accurate prediction of travel time along roadway provides valuable traffic information for travelers and traffic managers. Aiming at short-term travel time forecasting on urban arterials,a prediction model( PSOSVM) combining support vector machine( SVM) and particle swarm optimization( PSO) is developed. Travel time data collected with Bluetooth devices are used to calibrate the proposed model. Field experiments show that the PSO-SVM model 's error indicators are lower than the single SVM model and the BP neural network( BPNN) model. Particularly,the mean-absolute percentage error( MAPE) of PSO-SVM is only 9. 453 4 %which is less than that of the single SVM model( 12. 230 2 %) and the BPNN model( 15. 314 7 %). The results indicate that the proposed PSO-SVM model is feasible and more effective than other models for shortterm travel time prediction on urban arterials.
