In smart city, all kinds of users' data are stored in electronic devices to make everything intelligent. A smartphone is the most widely used device and it pivot systems. However, current smartphones not competent manage sensitive data, they facing privacy leakage caused by over-collection. Data over-collection, which means apps collect more than its original function while within permission scope, rapidly becoming one serious potential security hazards city. this paper, we study state...

Nowadays, the industrial Internet of Things (IIoT) has been successfully utilized in smart manufacturing. The massive amount data IIoT promote development deep learning-based health monitoring for equipment. Since mechanical fault diagnosis collected on different working conditions or equipment have domain mismatch, models trained with training may not work practical applications. Therefore, it is essential to study methods adaptation ability. In this article, we propose an intelligent...

Data-driven fault diagnosis is essential for the reliability and safety of industry equipment. However, lack real labeled data make machine learning-based methods difficult to carry out. To solve this problem, article proposes a new framework called multilabel one-dimensional (1-D) generation adversarial network (ML1-D-GAN). In our method, Auxiliary Classifier GAN utilized first damage generation. Then generated are both used train classifier. Experimental results reveal that applicable,...

Tower cranes are complex multi-input multioutput underactuated mechatronics systems. The anti-swing control issue of tower crane with varying suspension cable length and double spherical pendulum effect is still open. Furthermore, the system parameters uncertainty makes it more challenging to implement control. In this study, we present an adaptive sliding mode approach based on time-delay estimation for effect. First, employ Lagrange's method develop a seven-degree-of-freedom (7-DOF)...

The dramatically growing demand of Cyber Physical and Social Computing (CPSC) has enabled a variety novel channels to reach services in the financial industry. Combining cloud systems with multimedia big data is approach for Financial Service Institutions (FSIs) diversify service offerings an efficient manner. However, security issue still great which availability often conflicts constraints when media are varied. This paper focuses on this problem proposes using Semantic-Based Access...

Massive industrial data collected from the Industrial Internet-of-Things (IIoT) assets improve data-driven methods for prognostics and health management (PHM) systems. As an important role in PHM, remaining useful life (RUL) prediction is essential to maintain reliability safety of manufacture. However, recent approaches bearing RUL do not weight contributions different sensors time steps, which decreases efficiency big era. In this context, we present a deep learning-based method with...

The smartphone is a typical cyberphysical system (CPS). It must be low energy consuming and highly reliable to deal with the simple but frequent interactions cloud, which constitutes cloud-integrated CPS. Dynamic voltage scaling (DVS) has emerged as critical technique leverage power management by lowering supply frequency of processors. In this paper, based on DVS technique, we propose novel Energy-aware Task Scheduling (EDTS) algorithm minimize total consumption for smartphones, while...

Being an essential component of smart education, we propose a novel recommendationsystem for course selection in the specialty information management inChinese Universities.To implement this system, firstly collect enrollment data-set specific group students. The sparse linear method (SLIM) is introduced our framework to generate top-N recommendations courses appropriate Meanwhile, aL0 regularization term isexploited as optimization strategywhich established on observation items current...

Deep learning based intelligent fault detection for mechanical equipment has become an important research fields. However, due to various and working conditions, it is difficult collect sufficient samples by monitoring sensors, which restricts the accuracy of existing diagnostic approaches. In this paper, we propose a novel generative adversarial networks (GAN) on local weights-shared multi-generator generation data building diagnosis model. Unlike GAN augmentation methods, proposed model...

The remaining useful life (RUL) prediction has always been the key technology to realize predictive maintenance. An accurate can give decision-makers a reliable reference develop maintenance schedules and adjust production planning. When dealing with spatiotemporal data of multisensor system, recent deep learning (DL) methods, however, still remain unexplored weigh contributions from both spatial temporal dimensions. In this article, we propose novel DL-based approach dual channel feature...

Bearing faults are among the most common causes of machine failures. Therefore, bearing fault diagnosis should be performed reliably and rapidly. Currently, many types modal data for monitoring running state bearings available. They include from different kinds sensors various domains (such as time frequency domains). However, obtaining features with high quality single-modal is difficult because complex working conditions. extracting integrating complementary multimodal problems that remain...

Currently, with the development of Internet Things (IoTs) and artificial intelligence, a new IoT structure known as Intelligence (AIoTs) comes into play. With AIoT, large amount unlabeled industrial big data has been accumulated. The analysis amounts is labor-intensive time-consuming for diagnostic personnel. To improve this situation, novel two-stage unsupervised fault recognition algorithm, namely, deep adaptive fuzzy clustering algorithm (DAFC) proposed in article. DAFC amalgamates...

Bearing fault diagnosis is vital for saving invaluable time and cost since it the most critical components in rotary machines. The feature fusion method has been a effective way to enhance performance of diagnosis. However, how extract fuse complementary features from multi-sensor data still problem be solved. This study proposes hybrid network (MSHFFN) fully mine information signals bearing In network, empirical calculated by statistical methods deep are fused classification, more...

Traditional assembly tasks often require robots to transfer the acquired skills new tasks. However, previous reinforcement learning methods typically ignore inherent relationship between source and target domain This requires a substantial amount of interaction data compensate for this deficiency, generally results in poor effects. To address issue, strategy method that establishes shared feature space is proposed enhance efficiency on peg-in-hole assembly. Initially, by calculating distance...

The bearing is the key component of rotating mechanical equipment, so fault diagnosis bearings important to improve reliability and safety equipment operation. In recent years, feature fusion method has been extensively explored in health monitoring bearings. However, almost all existing feature-fusion-based methods extract features from different signals independently concatenating them simply. It will lead failure achieving expected diagnostic accuracy because complementary information not...