With the development of urban rail transit, energy conservation has been highly concerned by the worldwide railway industry. The existing energy-saving control methods are hard to meet the ...requirements of large-scale multi-train intelligent cooperation. Furthermore, in order to optimize the overall energy consumption and make effective utilization of regenerative braking energy (RBE), the micro train driving strategies must be taken into account in the macro train schedule regulation. This article aims to establish a systematic optimization model to describe the train traffic environment and design a deep reinforcement learning (DRL) approach using multi-agent cooperative actor-critic (MACAC) to reschedule multiple trains for energy saving. Due to the high dimensional characteristic of control actions, the action representation technique is applied for generalization. The multiple agents are deployed on the Spark cloud platform to effectively improve the learning efficiency through multi-node parallel computing. The proposed MACAC approach on Spark cloud is capable of dealing with the intelligent learning and computing efficiency of large-scale train operation synergy, and saving energy to the maximum extent on the premise of satisfying various time constraints. The evaluation was carried out through the field data of Beijing Yizhuang subway line. The results show that the proposed MACAC approach can effectively learn to improve the energy efficiency, and the parallel computing method on Spark cloud is suitable for practical applications.
Train movement prediction simulation is an effective method of enhancing operation safety and efficiency of railway transportation. Train movement data generated by the trains running in a railway ...network are huge, and on the other hand, the data quantity generated from data processing is also explosively increased for the analysis and decision such as conflict recognition and scheduling optimization, which greatly increases the time cost of prediction simulation. This paper attempts to propose a train movement model driven by the movement authorities (MAs) issued from radio block centers (RBCs) and its parallel simulation algorithm realized on Spark cloud. This paper provides a solution of iterative computing of dynamic process simulation based on cloud. Different from the general big data processing of independent datasets, the resilient distributed datasets are expandable along iterative computing processes. The Dataframe of SparkSQL modules on Apache Spark is employed to handle the problems of usage interdependency of datasets. The parallel simulation is realized by Scala language that is used to build the Spark platform. The simulation results on a high-speed railway network demonstrates that the proposed train movement model and parallel algorithm can achieve theoretical rationality and decrease the time cost to satisfy real-time performance.
Time series data is common in data sets has become one of the focuses of current research. The prediction of time series can be realized through the mining of time series data, so that we can obtain ...the development process and regularity of social economic phenomena reflected by time series, and extrapolate to predict its development trend. More and more attention has been paid to time series prediction in the era of big data. It is the basic application of time series prediction to accurately predict the trend. In this paper, we introduce various time series autoregressive (AR) model, moving average (MA) model, and ARIMA model that is combined by AR and MA. As the time series prediction in general scenarios, the ARIMA is applied to the risk prediction of the National SME Stock Trading (New Third Board) in combination with specific scenarios. The case studies show that the results of our analysis are basically consistent with the actual situation, which has greatly helped the prediction of financial risks.
User-perceived performance continues to be the most important QoS indicator in cloud-based data centers today. Effective allocation of virtual machines (VMs) to handle both CPU intensive and I/O ...intensive workloads is a crucial performance management capability in virtualized clouds. Although a fair amount of researches have dedicated to measuring and scheduling jobs among VMs, there still lacks of in-depth understanding of performance factors that impact the efficiency and effectiveness of resource multiplexing and scheduling among VMs. In this paper, we present the experimental research on performance interference in parallel processing of CPU-intensive and network-intensive workloads on Xen virtual machine monitor (VMM). Based on our study, we conclude with five key findings which are critical for effective performance management and tuning in virtualized clouds. First, colocating network-intensive workloads in isolated VMs incurs high overheads of switches and events in Dom0 and VMM. Second, colocating CPU-intensive workloads in isolated VMs incurs high CPU contention due to fast I/O processing in I/O channel. Third, running CPU-intensive and network-intensive workloads in conjunction incurs the least resource contention, delivering higher aggregate performance. Fourth, performance of network-intensive workload is insensitive to CPU assignment among VMs, whereas adaptive CPU assignment among VMs is critical to CPU-intensive workload. The more CPUs pinned on Dom0 the worse performance is achieved by CPU-intensive workload. Last, due to fast I/O processing in I/O channel, limitation on grant table is a potential bottleneck in Xen. We argue that identifying the factors that impact the total demand of exchanged memory pages is important to the in-depth understanding of interference costs in Dom0 and VMM.
With the development of science and technology, the application of big data is becoming more and more widespread, and it has gradually expanded to various fields such as economy and commerce. Since ...the 2008 international financial crisis, the mainstream economics has shown deficiencies to a certain extent. On the one hand, the expressions pursued by mainstream economic theories are too strict, restricting its processing capabilities. On the other hand, the linearization method ignores the diversity, complexity, and variability of changes in the economic system, which may ignore the emergence of some serious crises. Due to the increasing distance between theoretical models and practice, theoretical models cannot guide the practice and sometimes even mislead the latter. In this paper, we propose a method of dynamic feedback early warning based on big data, which uses the LPPL model to fit parameters. Finally, we used this method to analyze the case of the A-share disaster. The research results show that the method makes the early warning coefficients of dynamic and complex systems more scientific and accurate.
Server consolidation and application consolidation through virtualization are key performance optimizations in cloud-based service delivery industry. In this paper, we argue that it is important for ...both cloud consumers and cloud providers to understand the various factors that may have significant impact on the performance of applications running in a virtualized cloud. This paper presents an extensive performance study of network I/O workloads in a virtualized cloud environment. We first show that current implementation of virtual machine monitor (VMM) does not provide sufficient performance isolation to guarantee the effectiveness of resource sharing across multiple virtual machine instances (VMs) running on a single physical host machine, especially when applications running on neighboring VMs are competing for computing and communication resources. Then we study a set of representative workloads in cloud-based data centers, which compete for either CPU or network I/O resources, and present the detailed analysis on different factors that can impact the throughput performance and resource sharing effectiveness. For example, we analyze the cost and the benefit of running idle VM instances on a physical host where some applications are hosted concurrently. We also present an in-depth discussion on the performance impact of colocating applications that compete for either CPU or network I/O resources. Finally, we analyze the impact of different CPU resource scheduling strategies and different workload rates on the performance of applications running on different VMs hosted by the same physical machine.
Citizen complaint classification plays an important role in the construction of the smart city. For text data, the most expressive semantic information is reflected in the keyword of the text. With ...the proposed Transformer structure and further expansion of the model structure, natural language processing has embarked on a path of fine-tuning the pre-trained model based on the multi-headed attention mechanism. Although the above method works well, it further deepens the black box model of the network. To verify whether the multi-headed attention mechanism adds enough attention to the keyword information, this paper proposes a joint attention enhancement network that places the attention mechanism outside the main network model. This paper uses the idea of lexical frequency statistics to obtain keyword information through the macroscopic use of corpus contents and improves the attention through knowledge incorporation based on soft attention. In this paper, a comparison experiment is performed by the current hot open-source network models on Hugging Face. Experiments show that the proposed model improves about 10%-20% in accuracy compared with the different original models, while the network training time only increases about 5%. The joint enhancement network can identify the key region of input data more accurately and converge quickly.
The alleviation of traffic congestion relies on efficient traffic control and traffic guidance, which are based on real-time short-term traffic flow prediction. In this paper, the stacked autoencoder ...(SAE) deep learning model with powerful feature learning capability is selected to predict the traffic flow on road sections. The process of training SAE includes the pre-training phase and the fine-tuning phase, which mainly apply the BP algorithm. However, the process of training SAE is time-consuming and cannot meet the real-time performance of modern application systems. This paper proposes a parallel training strategy for the SAE prediction model based on data parallel mode. The gradient solution process in our algorithm satisfies the conditions of parallel computing, so the training process can be designed in a parallel manner. The original dataset is distributed to some computing nodes, which are work nodes. The work node is responsible for gradient calculation using the local data. The task of the sole master node is to synthesize the gradient calculation results and then broadcast the updated gradient to each work node. The simulation results show that the SAE-based prediction model achieves better results than the traditional model, and the parallel algorithm reduces the running time of training processes.
Abstract
In order to comprehensively evaluate the achievements of the 'Belt and Road' in integrated transportation, researchers need to optimize the method of generating evaluation indices and ...construct the framework structure of the 'Belt and Road' transportation index system. This paper used GDELT database as data source and obtained full text data of English news in 25 countries along ‘the Belt and Road’. The paper also introduced the topic model, combined with the unsupervised method (latent Dirichlet allocation, LDA) and the supervision method (labeled LDA) to mine the topics contained in the news data. It constructed the transportation development model and analyzed the development trend of transportation in various countries. The study found that the development trend of transportation in the countries along the line is unbalanced, which can be divided into four types: rapid development type, stable development type, slow development type and lagging development type. The method of this paper can effectively extract temporal and spatial variation of news events, discover potential risks in various countries, support real-time and dynamic monitoring of the social development situation of the countries along the border and provide auxiliary decision support for implementation of the ‘the Belt and Road’ initiative, which has important application value.