Search

Abstract Rotary machines are constructed and operated in diverse industrial environments and operate according to various specifications and characteristics. When rotary machinery constructed under dynamic real world environments is in operation, various types of vibrations are generated depending on the normal or defective state of the machinery. In this way, Numerous studies have been conducted on vibration analysis for diagnosing the state of rotary machinery. However, Without performing robust data cleansing and comprehensive labeling of the internal and external state of complex machinery, the analysis process of the condition monitoring system faces difficulties in accurately identifying the various and complex states of rotary machines and making decisions in the dynamic real world. To overcome these limitations, this paper proposes Multi Combination Pattern Labeling (MCPL) method. By simultaneously considering the complex internal and external states of rotary machines, MCPL generates detailed vibration frequency pattern criteria and labels for each state. Based on these complex pattern classifications, it is able to classify various types of abnormal states. The MCPL generates FFT patterns and spectrogram patterns by considering the simultaneous internal and external states of the rotary machine. Extracting internal and external patterns, each pattern is combined for identifying convergence patterns, named MCP. Each MCP proceeds labeling process, named MCPL, then MCPL dataset is structured. MCPL dataset is verified based on Deep Neural Network (DNN) and Convolutional Neural Network (CNN). By utilizing the DNN and CNN techniques to derive MCPL from MCP, it becomes possible to perform unbiased state diagnosis across a variety of patterns, based on the complex patterns of the internal and external states of the rotating machinery. Presenting high accuracy and stable results, MCPL are able to classify rotary machine states and detect anomalies under the convergence environment. Our source code and utilized data are available on https://github.com/JEJESBSB/Journal-of-Big-Data .

The Korean government funded 12.8 billion USD to 652 research and development (R&D) projects supported by 20 ministries in 2019. Every year, various organizations are supported to conduct R&D projects focusing on selected core technologies by evaluating emerging technologies which industries are planning to develop. To manage the whole cycle of national R&D projects, information sharing on national R&D projects is very essential. The blockchain technology is considered as a core solution to share information reliably and prevent forgery in various fields. For efficient management of national R&D projects, we enhance and analyse the Perfect Sharing Project (PSP)-Platform based on a new blockchain-based platform for information sharing and forgery prevention. It is a shared platform for national ICT R&D projects management with excellent performance in preventing counterfeiting. As a consensus algorithm is very important to prevent forgery in blockchain, we survey not only architectural aspects and examples of the platform but also the consensus algorithms. Considering characteristics of the PSP-Platform, we adopt an atomic proof (POA) consensus algorithm as a new consensus algorithm in this paper. To prove the validity of the POA consensus algorithm, we have conducted experiments. The experiment results show the outstanding performance of the POA consensus algorithm used in the PSP-Platform in terms of block generation delay and block propagation time.

With the development of the Internet of Things (IoT), the amount of data is growing and becoming more diverse. There are several problems when transferring data to the cloud, such as limitations on network bandwidth and latency. That has generated considerable interest in the study of edge computing, which processes and analyzes data near the network terminals where data is causing. The edge computing can extract insight data from a large number of data and provide fast essential services through simple analysis. The edge computing has a real-time advantage, but also has disadvantages, such as limited edge node capacity. The edge node for edge computing causes overload and delays in completing the task. In this paper, we proposes an efficient offloading model through collaboration between edge nodes for the prevention of overload and response to potential danger quickly in emergencies. In the proposed offloading model, the functions of edge computing are divided into data-centric and task-centric offloading. The offloading model can reduce the edge node overload based on a centralized, inefficient distribution and trade-off occurring in the edge node. That is the leading cause of edge node overload. So, this paper shows a collaborative offloading model in edge computing that guarantees real-time and prevention overload prevention based on data-centric offloading and task-centric offloading. Also, we present an intelligent offloading model based on several scenarios of forest fire ignition.

CCTV-based behavior recognition systems have gained considerable attention in recent years in the transportation surveillance domain for identifying unusual patterns, such as traffic jams, accidents, dangerous driving and other abnormal behaviors. In this paper, a novel approach for traffic behavior modeling is presented for video-based road surveillance. The proposed system combines the pachinko allocation model (PAM) and support vector machine (SVM) for a hierarchical representation and identification of traffic behavior. A background subtraction technique using Gaussian mixture models (GMMs) and an object tracking mechanism based on Kalman filters are utilized to firstly construct the object trajectories. Then, the sparse features comprising the locations and directions of the moving objects are modeled by PAMinto traffic topics, namely activities and behaviors. As a key innovation, PAM captures not only the correlation among the activities, but also among the behaviors based on the arbitrary directed acyclic graph (DAG). The SVM classifier is then utilized on top to train and recognize the traffic activity and behavior. The proposed model shows more flexibility and greater expressive power than the commonly-used latent Dirichlet allocation (LDA) approach, leading to a higher recognition accuracy in the behavior classification.

As the vast amount of data in social Internet of Things (IoT) environments considering interactions between IoT and people is accumulated and processed through cloud and big data technologies, the services that utilize them are applied in various fields. The trust between IoT devices and their data is recognized as the core of IoT ecosystem creation and growth. Connection with suspicious IoT devices may pose a risk to services and system operation. Therefore, it is essential to analyze and manage trust information for devices, services, and people, as well as to provide the trust information to the other devices or users that need it. This paper presents a trust information management framework which contains a generic IoT reference model with trust capabilities to achieve the goal of converged trust information management. Additionally, a trust information management platform (TIMP) consisting of trust agents, trust information brokers, and trust information management systems has been proposed, which aims to provide trustworthy and safe interactions among people, virtual objects, and physical things. Implementing and deploying a TIMP enables a trustworthy ecosystem to be built while activating social IoT businesses by reducing transaction costs, as well as by eliminating the uncertainties in the use of social IoT services and data transactions.