Day-ahead traffic flow forecasting based on a deep belief network optimized by the multi-objective particle swarm algorithm

Li, Linchao; Qin, Lingqiao; Xu, Qing; Zhang, Jian; Yong-gang, Wang; Ran, Bin

doi:10.1016/j.knosys.2019.01.015

Cited by 177 publications

(70 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Xiang et al [28], [38] developed stacked denoising auto encoder (SDAE) to predict the traffic flow with missing data. Xie et al [39] and other scholars analyzed the transport problem by a deep belief network (DBN) [40]- [43]. Sun et al [44] and other scholars applied convolutional neural network (CNN) to process time series data for traffic forecast [45]- [47].…”

Section: Literature Reviewmentioning

confidence: 99%

Traffic Flow Forecast Through Time Series Analysis Based on Deep Learning

Jiao

Huang

2020

IEEE Access

View full text Add to dashboard Cite

Traffic congestion is a thorny issue to many large and medium-sized cities, posing a serious threat to sustainable urban development. Recently, intelligent traffic system (ITS) has emerged as an effective tool to mitigate urban congestion. The key to the ITS lies in the accurate forecast of traffic flow. However, the existing forecast methods of traffic flow cannot adapt to the stochasticity and sheer length of traffic flow time series. To solve the problem, this paper relies on deep learning (DL) to forecast traffic flow through time series analysis. The authors developed a traffic flow forecast model based on the long shortterm memory (LSTM) network. The proposed model was compared with two classic forecast models, namely, the autoregressive integrated moving average (ARIMA) model and the backpropagation neural network (BPNN) model, through long-term traffic flow forecast experiments, using an actual traffic flow time series from OpenITS. The experimental results show that the proposed LSTM network outperformed the classic models in prediction accuracy. Our research discloses the dynamic evolution law of traffic flow, and facilitates the decision-making of traffic management. INDEX TERMS Traffic flow forecast, time series analysis, deep learning (DL), long short-term memory (LSTM).

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Traffic Flow Forecast Through Time Series Analysis Based on Deep Learning

Jiao

Huang

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Vlahogianni [44] proposed a surrogate model considering fusing three different models to forecast the short-term speed on the freeway. Li et al [45] put forward a deep belief network optimized by the multiobjective particle swarm algorithm to realize multi-timestep forecasting. Wu et al [14] established a hybrid deep neural network, which employs a convolutional neural network to mine the spatial features and uses the recurrent neural network to mine the temporal features of traffic flow, to predict traffic flow in a long-term horizon.…”

Section: Hybrid Modelsmentioning

confidence: 99%

A Hybrid Model for Lane-Level Traffic Flow Forecasting Based on Complete Ensemble Empirical Mode Decomposition and Extreme Gradient Boosting

Rui

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Accurate and efficient lane-level traffic flow prediction is a challenging issue in the framework of the connected automated vehicle highway system. However, most existing traffic flow forecasting methods concentrate on mining the spatio-temporal characteristics of the traffic flow rather than increasing predictability of traffic flow. In this paper, we propose a novel hybrid model (CEEMDAN-XGBoost) for lane-level traffic flow prediction based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and extreme gradient boosting (XGBoost). The CEEMDAN method is introduced to decompose the raw traffic flow data into several intrinsic mode function components and one residual component. Then, the XGBoost methods are trained and make predictions on the decomposed components respectively. The final prediction results are obtained by integrating the prediction outputs of the XGBoost methods. For illustrative purposes, the ground-truth lane-level traffic flow data captured by remote traffic microwave sensors installed on the 3 rd Ring Road of Beijing are utilized to evaluate the effectiveness of the CEEMDAN-XGBoost model. The experimental results confirm that the CEEMDAN-XGBoost model is capable of fitting the complex volatility of traffic flow efficiently at different types of lane sections. Moreover, the proposed model outperforms the state-of-the-art models (e.g., artificial neural networks and long shortterm memory neural network) and other XGBoost-based models in terms of prediction accuracy and stability. INDEX TERMS Data mining, lane-level traffic flow, short-term prediction, hybrid model, extreme gradient boosting, complete ensemble empirical mode decomposition, urban expressways.

show abstract

“…Compared with other heuristic algorithms, PSO algorithm is suitable for solving large-scale multiobjective problems. It converges fast to optimal solution and encodes simply because it only uses a few parameters for tuning [37][38][39]. Based on the mentioned characteristics, the PSO algorithm is applied to solve the double-objective model that considers the utilizing rate and the walking distance.…”

Section: Particle Swarm Optimizationmentioning

confidence: 99%

A Distribution Model for Shared Parking in Residential Zones that Considers the Utilization Rate and the Walking Distance

Zhang

Gao

ShuRui

et al. 2020

Journal of Advanced Transportation

View full text Add to dashboard Cite

Efficient parking tends to be challenging in most large cities in China. Drivers often spend substantial amounts of time looking for parking lots while driving at low speeds, thereby resulting in interference with road traffic. This paper focuses on efficiently allocating parking spaces to the demanders. A double-objective model is proposed that considers both the utilizing rate and the walking distance. First, managers want to utilize parking resources fully. Therefore, they tend to prioritize the efficient distribution of parking spaces in response to parking demands. However, demanders typically choose parking spaces according to convenience. The second objective is the acceptable walking distance from the parking space to the destination. The particle swarm optimization (PSO) algorithm is used to solve this model. We collected parking demand and supply data in a central business district (CBD) of Harbin in China and evaluated the feasibility of the model. The results demonstrate that the proposed model increases the occupying rates of parking lots in residential zones while decreasing the walking distance. The shared use of parking spaces maximizes the utility and alleviates the shortage of parking spaces in downtown.

show abstract

Day-ahead traffic flow forecasting based on a deep belief network optimized by the multi-objective particle swarm algorithm

Cited by 177 publications

References 49 publications

Traffic Flow Forecast Through Time Series Analysis Based on Deep Learning

Traffic Flow Forecast Through Time Series Analysis Based on Deep Learning

A Hybrid Model for Lane-Level Traffic Flow Forecasting Based on Complete Ensemble Empirical Mode Decomposition and Extreme Gradient Boosting

A Distribution Model for Shared Parking in Residential Zones that Considers the Utilization Rate and the Walking Distance

Contact Info

Product

Resources

About