没有合适的资源?快使用搜索试试~ 我知道了~
基于M-B-LSTM混合网络的短期交通流预测方法
需积分: 0 5 下载量 26 浏览量
2022-10-16
17:23:43
上传
评论
收藏 5.26MB PDF 举报
温馨提示
【基于M-B-LSTM混合网络的短期交通流预测方法】是一种深度学习技术在智能交通系统中的应用,旨在解决短期交通流量预测中的不确定性和过拟合问题。近年来,深度学习已经在短期交通流量预测方面取得了显著成果,但交通流量的随机性和分布不均衡特性使得预测过程中存在不确定性,容易导致模型过拟合。 M-B-LSTM模型是一种端到端的深度学习网络,由数据映射层、随机性减少层和预测层组成。在这个模型中,首先构建了一个在线自学习网络作为数据映射层,它的主要任务是学习和均衡交通流量的统计分布,以减轻由于分布不均衡引发的过拟合问题。这种自我学习机制能够动态调整,适应不断变化的交通流量数据。 随后,在随机性减少层中,引入了深度双向长短时记忆网络(DBLSTM)。DBLSTM通过前向和反向上下文的近似处理,减少了因交通流量的随机性而带来的不确定性。这种双向结构能够捕捉时间序列数据的前后关联,提供更全面的信息。 预测层采用长短时记忆网络(LSTM)进行实际的流量状态预测。LSTM因其对长期依赖性的有效处理,特别适合于处理具有时间序列特性的交通流量数据。 为了验证M-B-LSTM模型的有效性,进行了充分的对比实验。实验结果表明,该模型在解决不确定性和过拟合问题上,相比现有的先进方法具有更好的性能。 本文的工作不仅强调了深度学习在交通流量预测中的潜力,还提出了针对交通数据特性的创新解决方案。M-B-LSTM模型的提出,对于提升交通流量预测的准确性和稳定性具有重要意义,有助于实现更智能的交通管理和决策支持,进而缓解交通拥堵,优化城市交通运行效率。 关键词:短期交通预测,深度学习,混合网络,不平衡数据处理,数据随机性。
资源推荐
资源详情
资源评论
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![pdf](https://img-home.csdnimg.cn/images/20210720083512.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![zip](https://img-home.csdnimg.cn/images/20210720083736.png)
![](https://csdnimg.cn/release/download_crawler_static/86769341/bg1.jpg)
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 23, NO. 1, JANUARY 2022 225
Short-Term Traffic Flow Forecasting Method With
M-B-LSTM Hybrid Network
Qu Zhaowei , Li Haitao, Li Zhihui , and Zhong Tao
Abstract— Deep learning has achieved good performance in
short-term traffic forecasting recently. However, the stochasticity
and distribution imbalance are main characteristics to traffic
flow, and these will bring the uncertainty and induce the
network overfitting problem during deep learning. To deal with
the problems, a new end-to-end hybrid deep learning network
model, named M-B-LSTM, is proposed for short-term traffic flow
forecasting in this paper. In the M-B-LSTM model, an online self-
learning network is constructed as a data mapping layer to learn
and equalize the traffic flow statistic distribution for reducing the
effect of distribution imbalance and overfitting problem during
network learning. Besides, the deep bidirectional long short-term
memory network (DBLSTM) is introduced to reduce the uncer-
tainty problem by forward and reverse contexts approximation
process in the stochasticity reducing layer, and then the long
short-term memory network (LSTM) is used to forecast the next
traffic flow state in the forecasting layer. Furthermore, sufficient
comparative experiments have been conducted and the results
show the proposed model has better ability on solving uncertainty
and overfitting problems than the state-of-art methods.
Index Terms— Short-term traffic forecasting, deep learning,
hybrid network, unbalanced data processing, data stochasticity.
I. INTRODUCTION
T
RAFFIC flow forecasting is an essential component
of intelligent transportation system. Real-time accurate
short-time traffic flow forecasting can give traffic guidance for
traffic participants to select an appropriate travel route, and
help the traffic managers with a reasonable control strategy to
relieve traffic congestion. Traffic flow is a time series data with
strong regularity and cyclicity which are the basis for accurate
forecasting, but the randomness and uncertainty of it increase
forecast difficulty [1]. Thus, short-time traffic flow forecasting
is essential also challenging in transportation management and
research fields.
In the past f ew decades, a lot o f methods have been carried
out to forecast short-term traffic flow, such as auto-regressive
integrated moving average ( ARIMA) [2], Kalman filter [3],
[4], fuzzy theory [5], [6], artificial neural network (ANN) [7].
These methods are proven to be useful in extracting the traffic
flow temporal tendency and predicting the future traffic flow.
Manuscript received September 19, 2018; revised March 16, 2019,
May 29, 2019, October 10, 2019, and March 9, 2020; accepted July 6, 2020.
Date of publication July 29, 2020; date of current version December 28,
2021. This work was supported in part by the National Key Research
and Development Program of China under Grant 2019YFB1600500. The
Associate Editor for this article was M. Zhou. (Corresponding author:
Li Zhihui.)
The authors a re with the School of Transportation, Jilin University,
Changchun 130022, China (e-mail: quzw@jlu.edu.cn; liht16@jlu.end.cn;
lizhih@jlu.edu.cn; 1047706274@qq.com).
Digital Object Identifier 10.1109/TITS.2020.3009725
For the shallow architecture or short-step training strategy
of these methods, the computed forecasting results are still
unsatisfying and need to be more accurate in application. With
the development of deep learning and the increase of traffic
big data, many deep learning models have been carried out
to predict short-term traffic flow. Deep belief network (DBN)
[8], [9], stacked autoencoders (SAE) [10], [11], recurrent
neural networks (RNN) [12], long short-term memory network
(LSTM) [13] and hybrid deep learning network [14], [15], etc.
used multiple-layer architecture to capture deep feature in his-
torical data from multiple levels, a nd achieved more reasonable
forecast performance than superficial pattern learning models.
As usual, the stochasticity and distribution imbalance are
common characteristics to traffic flow. The data stochasticity
will bring the uncertainty to deep network for learning and
forecasting, and the distribution imbalance is the main feature
to traffic flow for that the data statistic distribution is over-
concentration in some frequency bands as shown in Fig. 2.
Data in narrow over-concentration bands will be learned too
many times, which will induce the overfitting problem in learn-
ing stage and decrease the deep learning network discrimina-
tion. In general, oversampling and undersampling are common
solutions to deal with distribution imbalance problem in the
training phase for classification problems [16], and the sam-
pling process is used as an independent preprocessing before
deep network learning, this will induce the deep learning net-
work performance varied with hand-crafted sampling strategy
and network parameters are not overall optimized. Further-
more, the sampling process is an assumption that all traffic
states are sampled in advance for learning and all traffic states
distributions are obtained beforehand. In fact, it is impossible
to get all traffic states and its distributions before learning for
traffic flow’s time-varying and stochasticity. Up to now, the
problems on traffic flow’s stochasticity and distribution imbal-
ance are still confused to the online traffic flow forecasting.
To deal with the problems, a new end to end hybrid deep
learning network model, named M-B-LSTM, is constructed by
integral parameters optimization and self-learning for online
short-term traffic flow forecasting. The model integrates the
data mapping layer, stochasticity reducing layer and forecast
layer together to reduce the ‘hand-crafted’ network implement
effect. In the data mapping layer, an online self-learning proba-
bility distribution mapping network is construc ted to generate
and equalize the traffic flow distribution for improving the
data discrimination and reducing the overfitting problem in
network learning. DBLSTM structure is adopted by forward
and backward contexts approximation process to reduce the
1558-0016 © 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See https://www .ieee.org/publications/rights/index.html for more information.
Authorized licensed use limited to: DALIAN MARITIME UNIVERSITY. Downloaded on September 08,2022 at 12:48:55 UTC from IEEE Xplore. Restrictions apply.
![](https://csdnimg.cn/release/download_crawler_static/86769341/bg2.jpg)
226 IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 23, NO. 1, JANUARY 2022
uncertainty problem for deep learning network in the stochas-
ticity reducing layer. To forecast the next traffic flow state,
LSTM is resorted in the forecast layer and then the M-B-
LSTM network is implemented. In experiments, the compared
experiment results show the M-B-LSTM network has better
ability to deal with th e uncertainty and overfitting problems
than the other methods. The main contributions of this paper
can be summarized as follows:
1). An end-to-end deep network model, M-B-LSTM, is con-
structed to deal with the uncertain ty and overfitting prob-
lems during deep network learning and f orecasting for traffic
flow stochasticity and distribution imbalance. Furthermore, the
M-B-LSTM can be used as a new network model for all time
series data forecasting to d eal with the data’s stochasticity and
distribution imbalance.
2). An o nline self-learning probability distribution mapping
network is implemented to generate and equalize traffic flow
historical distribution automatically for improving the data
discrimina tion and reducing the overfitting problem in network
learning. Also, the network can be used to deal with the hard
problem on generating probability distribution about online
time serials data up to now.
3) DBLSTM is in tegrated into M-B-LSTM network to
reduce the uncertainty on deep network learning and forecast-
ing for traffic flow stochasticity according to the network’s
forward and backward contexts approximation.
4) Comparison experiments on the state-of-art traffic flow
forecasting algorithms have been carried out in the paper and
the comparison results provide a comprehensive reference for
research.
The rest of this paper is organized as follows. In Section
II, we introduce the traffic flow forecasting related works.
In Section III, the M-B-LSTM hybrid network architecture
and calculation process will be illustrated in detail. Experi-
ments and comparison results with several state-of-art forecast
methods are shown in Section IV. The last section V makes
a conclusion and discusses the future work.
II. R
ELATED WORK
Short-term traffic flow forecasting has a long history in
transportation literature. With the development of several
decades, there are quite many methods for traffic flow forecast.
From the perspective of traffic flow data learning, traffic
flow forecasting models can be divided into two categories:
superficial pattern learning models and deep pattern learning
models.
A. Superficial Pattern Learning Models
Superficial pattern learning models usually used short time
window or data statistics tendency to estimate the next traffic
flow state. The historical average method [17], auto-regressive
integrated moving average (ARIMA) [2], multivariate time-
series model [18], and Kalman filter [3], [4] are introduced
into traffic forecast in earlier times. The forecast performance
of those methods is not quite satisfactory if the traffic shows
irregular variations. To solve this problem, scholars focus
research on the acquisition of nonlinear laws in historical data.
Artificial neural network [7], radial basis function (RBF) [19],
[20], K-nearest neighbor (KNN) [21], support vector machines
(SVM) [22], [23], fuzzy theory [5], [6], functional data analy-
sis (FDA) [24], and genetic algorithm [25], [26], are typical
nonlinear forecasting models, and many combinations of these
algorithms [27]–[30] have also been proposed accordingly. For
shallower architectures of these models, the superficial traffic
flow r ules are extracted and massive traffic data can’t be used
effectively.
B. Deep Pattern Learning Models
For deep learning gained prominence in recent years with
the performance improvements in many fields, some meth-
ods have been successfully applied in traffic flow forecast.
Nicholas G. Polson [31] et.al proposed a deep learning archi-
tecture which combined a sparse linear layer and a sequence
of tanh layers to forecast traffic congestion and duration.
Arief Koesdwiady [32] et.al employed a deep belief network
approach to predict traffic flow by fusing the big data with the
history and weather information. Yang H F [11] et.al built a
stacked autoencoder Levenberg-Marquardt architecture (SAE)
to learn traffic flow features layer-by-layer for traffic flow fore-
casting. Besides, SAE model was improved by combing empir-
ical mode decomposition (EMD) algorithm [33] to decompose
a complex time series into a collection of simple ones in the
network forecasting. To better capture the temporal evolution
of traffic flow, the recurrent neural network (RNN) [12]
was introduced into traffic flow forecasting by self-loops and
chain-like structure. To overcome the vanishing gradient and
exploding gradient of RNN during back propagation process,
long short-term memory network (LSTM) was adopted as a
solution of traffic flow forecasting. Fu Rui [14] et.al used
multilayer LSTM network to predict traffic flow and com pared
experiment results showed LSTM is more appropriate alter-
native for traffic flow than other deep neural networks. Long
Short-Term Memory Recurrent Neural Network was employed
by Tian [15] et.al to automatically determine the optimal time
lags and enhance LSTM’s advantage for traffic flow prediction.
Du S [34] et.al proposed a hybrid deep learning framework
by combining LSTM and convolutional neural network (CNN)
for capturing local and long temporal dependencies by CNN
and LSTM units respectively in traffic flow forecasting.
Up to now, different deep learning structures have been
adopted to improve the traffic flow forecasting accuracy. Sto-
chasticity and distribution imbalance as main characteristics o f
traffic flow, these will bring the uncertainty and induce the net-
work overfitting problem during deep learning. Unfortunately,
there are seldom researches on reducing the effect of traffic
flow’s stochasticity and distr ibution imbalance, especially by
an end-to-end deep learning model.
III. M-B-LSTM H
YBRID NETWORK ARCHITECTURE
To deal with th e uncertainty and overfitting problem s in
deep network learning forecasting, the M-B-LSTM hybrid
network model is constructed and described as Fig. 1. From
Fig. 1, it can be showed that the M-B-LSTM network archi-
tecture includes data mapping layer, stochasticity reducing
Authorized licensed use limited to: DALIAN MARITIME UNIVERSITY. Downloaded on September 08,2022 at 12:48:55 UTC from IEEE Xplore. Restrictions apply.
剩余10页未读,继续阅读
资源评论
![avatar-default](https://csdnimg.cn/release/downloadcmsfe/public/img/lazyLogo2.1882d7f4.png)
![avatar](https://profile-avatar.csdnimg.cn/a6fa99e351774aa799bb5014584c2d83_qq_43537420.jpg!1)
想太多!
- 粉丝: 2773
- 资源: 38
上传资源 快速赚钱
我的内容管理 展开
我的资源 快来上传第一个资源
我的收益
登录查看自己的收益我的积分 登录查看自己的积分
我的C币 登录后查看C币余额
我的收藏
我的下载
下载帮助
![voice](https://csdnimg.cn/release/downloadcmsfe/public/img/voice.245cc511.png)
![center-task](https://csdnimg.cn/release/downloadcmsfe/public/img/center-task.c2eda91a.png)
最新资源
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
![feedback](https://img-home.csdnimg.cn/images/20220527035711.png)
![feedback](https://img-home.csdnimg.cn/images/20220527035711.png)
![feedback-tip](https://img-home.csdnimg.cn/images/20220527035111.png)
安全验证
文档复制为VIP权益,开通VIP直接复制
![dialog-icon](https://csdnimg.cn/release/downloadcmsfe/public/img/green-success.6a4acb44.png)