Multi-scale persistent spatiotemporal transformer for long-term urban traffic flow prediction

Long-term urban traffic flow prediction is an important task in the field of intelligent transportation,as it can help optimize traffic management and improve travel efficiency.To improve prediction accuracy,a crucial issue is how to model spatiotemporal dependency in urban traffic data.In recent years,many studies have adopted spatiotemporal neural networks to extract key information from traffic data.However,most models ignore the semantic spatial similarity between long-distance areas when mining spatial dependency.They also ignore the impact of predicted time steps on the next unpredicted time step for making long-term predictions.Moreover,these models lack a comprehensive data embedding process to represent complex spatiotemporal dependency.This paper proposes a multi-scale persistent spatiotemporal transformer(MSPSTT)model to perform accurate long-term traffic flow prediction in cities.MSPSTT adopts an encoder-decoder structure and incorporates temporal,periodic,and spatial features to fully embed urban traffic data to address these issues.The model consists of a spatiotemporal encoder and a spatiotemporal decoder,which rely on temporal,geospatial,and semantic space multi-head attention modules to dynamically extract temporal,geospatial,and semantic characteristics.The spatiotemporal decoder combines the context information provided by the encoder,integrates the predicted time step information,and is iteratively updated to learn the correlation between different time steps in the broader time range to improve the model’s accuracy for long-term prediction.Experiments on four public transportation datasets demonstrate that MSPSTT outperforms the existing models by up to 9.5%on three common metrics.

Traffic Flow Prediction with Heterogenous Data Using a Hybrid CNN-LSTM Model

Predicting traffic flow is a crucial component of an intelligent transportation system.Precisely monitoring and predicting traffic flow remains a challenging endeavor.However,existingmethods for predicting traffic flow do not incorporate various external factors or consider the spatiotemporal correlation between spatially adjacent nodes,resulting in the loss of essential information and lower forecast performance.On the other hand,the availability of spatiotemporal data is limited.This research offers alternative spatiotemporal data with three specific features as input,vehicle type(5 types),holidays(3 types),and weather(10 conditions).In this study,the proposed model combines the advantages of the capability of convolutional(CNN)layers to extract valuable information and learn the internal representation of time-series data that can be interpreted as an image,as well as the efficiency of long short-term memory(LSTM)layers for identifying short-term and long-term dependencies.Our approach may utilize the heterogeneous spatiotemporal correlation features of the traffic flowdataset to deliver better performance traffic flow prediction than existing deep learning models.The research findings show that adding spatiotemporal feature data increases the forecast’s performance;weather by 25.85%,vehicle type by 23.70%,and holiday by 14.02%.

Kalman Filter-Based CNN-BiLSTM-ATT Model for Traffic Flow Prediction

To accurately predict traffic flow on the highways,this paper proposes a Convolutional Neural Network-Bi-directional Long Short-Term Memory-Attention Mechanism(CNN-BiLSTM-Attention)traffic flow prediction model based on Kalman-filtered data processing.Firstly,the original fluctuating data is processed by Kalman filtering,which can reduce the instability of short-term traffic flow prediction due to unexpected accidents.Then the local spatial features of the traffic data during different periods are extracted,dimensionality is reduced through a one-dimensional CNN,and the BiLSTM network is used to analyze the time series information.Finally,the Attention Mechanism assigns feature weights and performs Soft-max regression.The experimental results show that the data processed by Kalman filter is more accurate in predicting the results on the CNN-BiLSTM-Attention model.Compared with the CNN-BiLSTM model,the Root Mean Square Error(RMSE)of the Kal-CNN-BiLSTM-Attention model is reduced by 17.58 and Mean Absolute Error(MAE)by 12.38,and the accuracy of the improved model is almost free from non-working days.To further verify the model’s applicability,the experiments were re-run using two other sets of fluctuating data,and the experimental results again demonstrated the stability of the model.Therefore,the Kal-CNN-BiLSTM-Attention traffic flow prediction model proposed in this paper is more applicable to a broader range of data and has higher accuracy.

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%.

MEEMD-DBA-based short term traffic flow prediction

Aiming at the problem that ensemble empirical mode decomposition(EEMD)method can not completely neutralize the added noise in the decomposition process,which leads to poor reconstruction of decomposition results and low accuracy of traffic flow prediction,a traffic flow prediction model based on modified ensemble empirical mode decomposition(MEEMD),double-layer bidirectional long-short term memory(DBiLSTM)and attention mechanism is proposed.Firstly,the intrinsic mode functions(IMFs)and residual components(Res)are obtained by using MEEMD algorithm to decompose the original traffic data and separate the noise in the data.Secondly,the IMFs and Res are put into the DBiLSTM network for training.Finally,the attention mechanism is used to enhance the extraction of data features,then the obtained results are reconstructed and added.The experimental results show that in different scenarios,the MEEMD-DBiLSTM-attention(MEEMD-DBA)model can reduce the data reconstruction error effectively and improve the accuracy of the short-term traffic flow prediction.

Overview of machine learning-based traffic flow prediction

Traffic flow prediction is an important component of intelligent transportation systems.Recently,unprecedented data availability and rapid development of machine learning techniques have led to tremendous progress in this field.This article first introduces the research on traffic flow prediction and the challenges it currently faces.It then proposes a classification method for literature,discussing and analyzing existing research on using machine learning methods to address traffic flow prediction from the perspectives of the prediction preparation process and the construction of prediction models.The article also summarizes innovative modules in these models.Finally,we provide improvement strategies for current baseline models and discuss the challenges and research directions in the field of traffic flow prediction in the future.

Short-term inbound rail transit passenger flow prediction based on BILSTM model and influence factor analysis

Accurate and real-time passenger flow prediction of rail transit is an important part of intelligent transportation systems(ITS).According to previous studies,it is found that the prediction effect of a single model is not good for datasets with large changes in passenger flow characteristics and the deep learning model with added influencing factors has better prediction accuracy.In order to provide persuasive passenger flow forecast data for ITS,a deep learning model considering the influencing factors is proposed in this paper.In view of the lack of objective analysis on the selection of influencing factors by predecessors,this paper uses analytic hierarchy processes(AHP)and one-way ANOVA analysis to scientifically select the factor of time characteristics,which classifies and gives weight to the hourly passenger flow through Duncan test.Then,combining the time weight,BILSTM based model considering the hourly travel characteristics factors is proposed.The model performance is verified through the inbound passenger flow of Ningbo rail transit.The proposed model is compared with many current mainstream deep learning algorithms,the effectiveness of the BILSTM model considering influencing factors is validated.Through comparison and analysis with various evaluation indicators and other deep learning models,the results show that the R2 score of the BILSTM model considering influencing factors reaches 0.968,and the MAE value of the BILSTM model without adding influencing factors decreases by 45.61%.

Short-Term Traffic Flow Prediction Based on LSTM-XGBoost Combination Model

According to the time series characteristics of the trajectory history data,we predicted and analyzed the traffic flow.This paper proposed a LSTMXGBoost model based urban road short-term traffic flow prediction in order to analyze and solve the problems of periodicity,stationary and abnormality of time series.It can improve the traffic flow prediction effect,achieve efficient traffic guidance and traffic control.The model combined the characteristics of LSTM(Long Short-Term Memory)network and XGBoost(Extreme Gradient Boosting)algorithms.First,we used the LSTM model that increases dropout layer to train the data set after preprocessing.Second,we replaced the full connection layer with the XGBoost model.Finally,we depended on the model training to strengthen the data association,avoided the overfitting phenomenon of the fully connected layer,and enhanced the generalization ability of the prediction model.We used the Kears based on TensorFlow to build the LSTM-XGBoost model.Using speed data samples of multiple road sections in Shenzhen to complete the model verification,we achieved the comparison of the prediction effects of the model.The results show that the combined prediction model used in this paper can not only improve the accuracy of prediction,but also improve the practicability,real-time and scalability of the model.

Traffic flow prediction based on BILSTM model and data denoising scheme

Accurate prediction of road traffic flow is a significant part in the intelligent transportation systems.Accurate prediction can alleviate traffic congestion,and reduce environmental pollution.For the management department,it can make effective use of road resources.For individuals,it can help people plan their own travel paths,avoid congestion,and save time.Owing to complex factors on the road,such as damage to the detector and disturbances from environment,the measured traffic volume can contain noise.Reducing the influence of noise on traffic flow prediction is a piece of very important work.Therefore,in this paper we propose a combination algorithm of denoising and BILSTM to effectively improve the performance of traffic flow prediction.At the same time,three denoising algorithms are compared to find the best combination mode.In this paper,the wavelet(WL) denoising scheme,the empirical mode decomposition(EMD) denoising scheme,and the ensemble empirical mode decomposition(EEMD) denoising scheme are all introduced to suppress outliers in traffic flow data.In addition,we combine the denoising schemes with bidirectional long short-term memory(BILSTM)network to predict the traffic flow.The data in this paper are cited from performance measurement system(PeMS).We choose three kinds of road data(mainline,off ramp,on ramp) to predict traffic flow.The results for mainline show that data denoising can improve prediction accuracy.Moreover,prediction accuracy of BILSTM+EEMD scheme is the highest in the three methods(BILSTM+WL,BILSTM+EMD,BILSTM+EEMD).The results for off ramp and on ramp show the same performance as the results for mainline.It is indicated that this model is suitable for different road sections and long-term prediction.

Optimal Logistics Activities Based Deep Learning Enabled Traffic Flow Prediction Model

Traffic flow prediction becomes an essential process for intelligent transportation systems(ITS).Though traffic sensor devices are manually controllable,traffic flow data with distinct length,uneven sampling,and missing data finds challenging for effective exploitation.The traffic data has been considerably increased in recent times which cannot be handled by traditional mathematical models.The recent developments of statistic and deep learning(DL)models pave a way for the effectual design of traffic flow prediction(TFP)models.In this view,this study designs optimal attentionbased deep learning with statistical analysis for TFP(OADLSA-TFP)model.The presentedOADLSA-TFP model intends to effectually forecast the level of traffic in the environment.To attain this,the OADLSA-TFP model employs attention-based bidirectional long short-term memory(ABLSTM)model for predicting traffic flow.In order to enhance the performance of the ABLSTM model,the hyperparameter optimization process is performed using artificial fish swarm algorithm(AFSA).A wide-ranging experimental analysis is carried out on benchmark dataset and the obtained values reported the enhancements of the OADLSA-TFP model over the recent approaches mean square error(MSE),root mean square error(RMSE),and mean absolute percentage error(MAPE)of 120.342%,10.970%,and 8.146%respectively.

Attention-based spatio-temporal graph convolutional network considering external factors for multi-step traffic flow prediction

Traffic flow prediction is an important part of the intelligent transportation system. Accurate multi-step traffic flow prediction plays an important role in improving the operational efficiency of the traffic network. Since traffic flow data has complex spatio-temporal correlation and non-linearity, existing prediction methods are mainly accomplished through a combination of a Graph Convolutional Network (GCN) and a recurrent neural network. The combination strategy has an excellent performance in traffic prediction tasks. However, multi-step prediction error accumulates with the predicted step size. Some scholars use multiple sampling sequences to achieve more accurate prediction results. But it requires high hardware conditions and multiplied training time. Considering the spatiotemporal correlation of traffic flow and influence of external factors, we propose an Attention Based Spatio-Temporal Graph Convolutional Network considering External Factors (ABSTGCN-EF) for multi-step traffic flow prediction. This model models the traffic flow as diffusion on a digraph and extracts the spatial characteristics of traffic flow through GCN. We add meaningful time-slots attention to the encoder-decoder to form an Attention Encoder Network (AEN) to handle temporal correlation. The attention vector is used as a competitive choice to draw the correlation between predicted states and historical states. We considered the impact of three external factors (daytime, weekdays, and traffic accident markers) on the traffic flow prediction tasks. Experiments on two public data sets show that it makes sense to consider external factors. The prediction performance of our ABSTGCN-EF model achieves 7.2%–8.7% higher than the state-of-the-art baselines.

AI Based Traffic Flow Prediction Model for Connected and Autonomous Electric Vehicles

There is a paradigm shift happening in automotive industry towards electric vehicles as environment and sustainability issues gainedmomentum in the recent years among potential users.Connected and Autonomous Electric Vehicle(CAEV)technologies are fascinating the automakers and inducing them to manufacture connected autonomous vehicles with self-driving features such as autopilot and self-parking.Therefore,Traffic Flow Prediction(TFP)is identified as a major issue in CAEV technologies which needs to be addressed with the help of Deep Learning(DL)techniques.In this view,the current research paper presents an artificial intelligence-based parallel autoencoder for TFP,abbreviated as AIPAE-TFP model in CAEV.The presented model involves two major processes namely,feature engineering and TFP.In feature engineering process,there are multiple stages involved such as feature construction,feature selection,and feature extraction.In addition to the above,a Support Vector Data Description(SVDD)model is also used in the filtration of anomaly points and smoothen the raw data.Finally,AIPAE model is applied to determine the predictive values of traffic flow.In order to illustrate the proficiency of the model’s predictive outcomes,a set of simulations was performed and the results were investigated under distinct aspects.The experimentation outcomes verified the effectual performance of the proposed AIPAE-TFP model over other methods.

Pedestrian wind flow prediction using spatial-frequency generative adversarial network

Pedestrian wind flow is a critical factor in designing livable residential environments under growing complex urban conditions.Predicting pedestrian wind flow during the early design stages is essential but currently suffers from inefficiencies in numerical simulations.Deep learning,particularly generative adversarial networks(GAN),has been increasingly adopted as an alternative method to provide efficient prediction of pedestrian wind flow.However,existing GAN-based wind flow prediction schemes have limitations due to the lack of considering the spatial and frequency characteristics of wind flow images.This study proposes a novel approach termed SFGAN,which embeds spatial and frequency characteristics to enhance pedestrian wind flow prediction.In the spatial domain,Gaussian blur is employed to decompose wind flow into components containing wind speed and distinguished flow edges,which are used as the embedded spatial characteristics.Detailed information of wind flow is obtained through discrete wavelet transformation and used as the embedded frequency characteristics.These spatial and frequency characteristics of wind flow are jointly utilized to enforce consistency between the predicted wind flow and ground truth during the training phase,thereby leading to enhanced predictions.Experimental results demonstrate that SFGAN clearly improves wind flow prediction,reducing Wind_MAE,Wind_RMSE and the Fréchet Inception Distance(FID)score by 5.35%,6.52%and 12.30%,compared to the previous best method,respectively.We also analyze the effectiveness of incorporating the spatial and frequency characteristics of wind flow in predicting pedestrian wind flow.SFGAN reduces errors in predicting wind flow at large error intervals and performs well in wake regions and regions surrounding buildings.The enhanced predictions provide a better understanding of performance variability,bringing insights at the early design stage to improve pedestrian wind comfort.The proposed spatial-frequency loss term is general and can be flexibly integrated with other generative models to enhance performance with only a slight computational cost.

Traffic Flow Prediction with Heterogeneous Spatiotemporal Data Based on a Hybrid Deep Learning Model Using Attention-Mechanism

A significant obstacle in intelligent transportation systems(ITS)is the capacity to predict traffic flow.Recent advancements in deep neural networks have enabled the development of models to represent traffic flow accurately.However,accurately predicting traffic flow at the individual road level is extremely difficult due to the complex interplay of spatial and temporal factors.This paper proposes a technique for predicting short-term traffic flow data using an architecture that utilizes convolutional bidirectional long short-term memory(Conv-BiLSTM)with attention mechanisms.Prior studies neglected to include data pertaining to factors such as holidays,weather conditions,and vehicle types,which are interconnected and significantly impact the accuracy of forecast outcomes.In addition,this research incorporates recurring monthly periodic pattern data that significantly enhances the accuracy of forecast outcomes.The experimental findings demonstrate a performance improvement of 21.68%when incorporating the vehicle type feature.

An Innovative Approach for the Short-term Traffic Flow Prediction

Traffic flow prediction plays an important role in intelligent transportation applications,such as traffic control,navigation,path planning,etc.,which are closely related to people's daily life.In the last twenty years,many traffic flow prediction approaches have been proposed.However,some of these approaches use the regression based mechanisms,which cannot achieve accurate short-term traffic flow predication.While,other approaches use the neural network based mechanisms,which cannot work well with limited amount of training data.To this end,a light weight tensor-based traffic flow prediction approach is proposed,which can achieve efficient and accurate short-term traffic flow prediction with continuous traffic flow data in a limited period of time.In the proposed approach,first,a tensor-based traffic flow model is proposed to establish the multi-dimensional relationships for traffic flow values in continuous time intervals.Then,a CANDECOMP/PARAFAC decomposition based algorithm is employed to complete the missing values in the constructed tensor.Finally,the completed tensor can be directly used to achieve efficient and accurate traffic flow prediction.The experiments on the real dataset indicate that the proposed approach outperforms many current approaches on traffic flow prediction with limited amount of traffic flow data.

FlowDNN:a physics-informed deep neural network for fast and accurate flow prediction

For flow-related design optimization problems,e.g.,aircraft and automobile aerodynamic design,computational fluid dynamics(CFD)simulations are commonly used to predict flow fields and analyze performance.While important,CFD simulations are a resource-demanding and time-consuming iterative process.The expensive simulation overhead limits the opportunities for large design space exploration and prevents interactive design.In this paper,we propose Flow DNN,a novel deep neural network(DNN)to efficiently learn flow representations from CFD results.Flow DNN saves computational time by directly predicting the expected flow fields based on given flow conditions and geometry shapes.Flow DNN is the first DNN that incorporates the underlying physical conservation laws of fluid dynamics with a carefully designed attention mechanism for steady flow prediction.This approach not only improves the prediction accuracy,but also preserves the physical consistency of the predicted flow fields,which is essential for CFD.Various metrics are derived to evaluate Flow DNN with respect to the whole flow fields or regions of interest(RoIs)(e.g.,boundary layers where flow quantities change rapidly).Experiments show that Flow DNN significantly outperforms alternative methods with faster inference and more accurate results.It speeds up a graphics processing unit(GPU)accelerated CFD solver by more than 14000×,while keeping the prediction error under 5%.

Inatorial forecasting method considering macro and micro characteristics of chaotic traffic flow

Traffic flow prediction is an effective strategy to assess traffic conditions and alleviate traffic congestion. Influenced by external non-stationary factors and road network structure, traffic flow sequences have macro spatiotemporal characteristics and micro chaotic characteristics. The key to improving the model prediction accuracy is to fully extract the macro and micro characteristics of traffic flow time sequences. However, traditional prediction model by only considers time features of traffic data, ignoring spatial characteristics and nonlinear characteristics of the data itself, resulting in poor model prediction performance. In view of this, this research proposes an intelligent combination prediction model taking into account the macro and micro features of chaotic traffic data. Firstly, to address the problem of time-consuming and inefficient multivariate phase space reconstruction by iterating nodes one by one, an improved multivariate phase space reconstruction method is proposed by filtering global representative nodes to effectively realize the high-dimensional mapping of chaotic traffic flow. Secondly, to address the problem that the traditional combinatorial model is difficult to adequately learn the macro and micro characteristics of chaotic traffic data, a combination of convolutional neural network(CNN) and convolutional long short-term memory(ConvLSTM) is utilized for capturing nonlinear features of traffic flow more comprehensively. Finally,to overcome the challenge that the combined model performance degrades due to subjective empirical determined network parameters, an improved lightweight particle swarm is proposed for improving prediction accuracy by optimizing model hyperparameters. In this paper, two highway datasets collected by the Caltrans Performance Measurement System(PeMS)are taken as the research objects, and the experimental results from multiple perspectives show that the comprehensive performance of the method proposed in this research is superior to those of the prevalent methods.

Self-attention Based Multimodule Fusion Graph Convolution Network for Traffic Flow Prediction

With rapid economic development,the per capita ownership of automobiles in our country has begun to rise year by year.More researchers have paid attention to using scientific methods to solve traffic flow problems.Traffic flow prediction is not simply affected by the number of vehicles,but also contains various complex factors,such as time,road conditions,and people flow.However,the existing methods ignore the complexity of road conditions and the correlation between individual nodes,which leads to the poor performance.In this study,a deep learning model SAMGCN is proposed to effectively capture the correlation between individual nodes to improve the performance of traffic flow prediction.First,the theory of spatiotemporal decoupling is used to divide each time of each node into finer particles.Second,multimodule fusion is used to mine the potential periodic relationships in the data.Finally,GRU is used to obtain the potential time relationship of the three modules.Extensive experiments were conducted on two traffic flow datasets,PeMS04 and PeMS08 in the Caltrans Performance Measurement System to prove the validity of the proposed model.

MCA-TFP Model:A Short-Term Traffic Flow Prediction Model Based on Multi-characteristic Analysis

With the urbanization,urban transportation has become a key factor restricting the development of a city.In a big city,it is important to improve the efficiency of urban transportation.The key to realize short-term traffic flow prediction is to learn its complex spatial correlation,temporal correlation and randomness of traffic flow.In this paper,the convolution neural network(CNN)is proposed to deal with spatial correlation among different regions,considering that the large urban areas leads to a relatively deep Network layer.First three gated recurrent unit(GRU)were used to deal with recent time dependence,daily period dependence and weekly period dependence.Considering that each historical period data to forecast the influence degree of the time period is different,three attention mechanism was taken into GRU.Second,a twolayer full connection network was applied to deal with the randomness of short-term flow combined with additional information such as weather data.Besides,the prediction model was established by combining these three modules.Furthermore,in order to verify the influence of spatial correlation on prediction model,an urban functional area identification model was introduced to identify different functional regions.Finally,the proposed model was validated based on the history of New York City taxi order data and reptiles for weather data.The experimental results show that the prediction precision of our model is obviously superior to the mainstream of the existing prediction methods.

THREE-DIMENSIONAL COUPLED IMPELLER-VOLUTE SIMULATION OF FLOW IN CENTRIFUGAL PUMP AND PERFORMANCE PREDICTION

A three-dimensional turbulent flow through an entire centrifugal pump is simulated using k-ε turbulence model modified by rotation and curvature, SIMPLEC method and body-fitted coordinate. The velocity and pressure fields are obtained for the pump under various working conditions, which is used to predict the head and hydraulic efficiency of the pump, and the results correspond well with the measured values. The calculation results indicate that the pressure is higher on the pressure side than that on the suction side of the blade; The relative velocity on the suction side gradually decreases from the impeller inlet to the outlet, while increases on the pressure side, it finally results in the lower relative velocity on the suction side and the higher one on the pressure side at the impeller outlet; The impeller flow field is asymmetric, i.e. the velocity and pressure fields arc totally different among all channels in the impeller; In the volute, the static pressure gradually increases with the flow route, and a large pressure gratitude occurs in the tongue; Secondary flow exists in the rear part of the spiral.