Power system faults are significant problems in power transmission and distribution. Methods based on relay protection actions electrical component have been put forward recent years. However, they deficiencies dealing with fault. In this paper, a method for data-based line trip fault prediction systems using long short-term memory (LSTM) networks support vector machine (SVM) is proposed. The temporal features of multisourced data captured LSTM networks, which perform well extracting the...

The recent spades of cyber attacks have compromised end-users' data security and privacy in Medical Cyber-Physical Systems (MCPS) the era Health 4.0. Traditional standard encryption algorithms for protection are designed based on a viewpoint system architecture rather than end-users. As such transferring to keys, safety, will be once key is exposed. In this paper, we propose secure storage sharing method consisted selective algorithm combined with fragmentation dispersion protect safety even...

As representation scheme can severely limit the window by which system observes its world, deep learning for fault diagnosis is put forward in this paper. It a real time online that enhance accuracy of detection, classification and prediction, efficient incipient faults cannot be detected traditional statistic technology. A stacked sparse auto encoder used to learn architectures data minimize loss information. Experiment results show proposed method not only improves divisibility between...

This letter aims at developing new memory architecture to overcome the daunting wall and energy issues in multicore embedded systems. We propose a heterogeneous scratch-pad (SPM) that is configured with SRAM, MRAM, Z-RAM. Based on this architecture, we two algorithms: dynamic programming (MDPDA) genetic algorithm (AGADA) allocate data different banks, therefore, reducing access cost terms of power consumption latency. Extensive experiments are performed show merits hybrid SPM effectiveness...

The brief studies the asynchronous observer-based sliding mode control (SMC) for Markov jump systems (MJSs) with actuator failures. Considering phenomena of unmeasurable states and case that controller/observer to be devised have different modes from original systems, a hidden model (HMM) is used construct an observer corresponding surface designed. Then, SMC strategy developed guarantee reachability predetermined in limited time. A sufficient condition established mean-square stability...

Electronic Health Record (EHR) systems have been playing a dramatically important role in tele-health domains. One of the major benefits using EHR is assisting physicians to gain patients' healthcare information and shorten process medical decision making. However, physicians' inputs still great impact on making decisions that cannot be checked by systems. This consequence can influenced human behaviors or knowledge structures. An efficient approach alerting unusual an urgent requirement for...

Short-term load forecasting is an important task for the planning and reliable operation of power grids. High-accuracy individual customers helps to make arrangements generation reduce electricity costs. Artificial intelligent methods have been applied short-term in past research, but most did not consider use characteristics, efficiency, more influential factors. In this paper, a method with multi-source data using gated recurrent unit neural networks proposed. The are preprocessed by...

Detecting behavioral anomalies in human daily life is important to developing smart assisted-living systems for elderly care. Based on data collected from wearable motion sensors and the associated locational context, this paper presents a coherent anomaly detection framework effectively detect different life. Four types of anomalies, including spatial anomaly, timing duration sequence are detected using probabilistic theoretical framework. This based complex activity recognition dynamic...

This paper investigates the cooperative estimation problem to recover parametric flow field through sensor measurements from an autonomous underwater vehicle (AUV) team. We establish model that incorporates concept of incompressibility provide a physical property. Then, considering influence unknown on AUVs' trajectories while submerged, we define: 1) deviation between actual and predicted relative positions each its neighbors as motion-integration error, which is available using local...

The distributed control of DC microgrid is becoming increasingly important in modern power systems. One objective to ensure bus voltage stability and proper current sharing with a reduced communication burden. This paper presents new event-triggered secondary strategy for single-bus microgrid. Through this strategy, both regulation can be guaranteed. Moreover, through the event-triggering mechanism, each converter decide locally when transmit signals its neighbours. In way, burden among...

This article investigates the adaptive fuzzy asynchronous control problem for discrete-time nonhomogeneous Markov jump power systems under hybrid attacks. A process is used to describe phenomenon of transient failures occurring in lines and subsequent switching associated circuit breakers. The corresponding hidden model utilized detect modes systems. Both deception attack denial-of-service are analyzed simultaneously owing vulnerability With detected logic systems, an strategy proposed....

This brief studies exponential H(infinity) synchronization of a class general discrete-time chaotic neural networks with external disturbance. On the basis drive-response concept and control theory, using Lyapunov-Krasovskii (or Lyapunov) functional, state feedback controllers are established to not only guarantee stable between two or without time delays, but also reduce effect disturbance on error minimal norm constraint. The proposed can be obtained by solving convex optimization problems...

This paper deals with application of deep learning neural network for power system fault diagnosis. Deep is a more effective approach than traditional to solve problems including availability data, better local optimum, and diffusion gradients. In the paper, data extracted from dispatching department preprocessed before training in network. Then, processed put into auto-encoders hidden features are observed different dimensions so that we can preliminarily judge about fault. Afterwards,...