Accurate prediction of remaining useful life (RUL) lithium-ion battery plays an increasingly crucial role in the intelligent health management systems. The advances deep learning introduce new data-driven approaches to this problem. This paper proposes integrated approach for RUL by integrating autoencoder with neural network (DNN). First, we present a multi-dimensional feature extraction method model represent degradation. Then, model-based DNN is trained multi-battery cycle estimation....

In modern society, many infrastructures are interdependent owing to functional and logical relations among components in different systems. These networked can be modeled as networks. the real world, networks carry traffic loads whose values dynamic stem from load redistribution same network disturbance network. Interdependency makes so fragile that even a slight initial may lead cascading failure of entire this paper, interdependencies cascade process is studied considering their effects on...

Rumors mislead judgments of people, affect economic development, and the stability social order. The research on rule spreading rumors is significant meaningful. This paper improves traditional Barabási–Albert scale-free network proposes a topology model that conforms to characteristics sharing networks based complex theory actual networks. In addition, credulous spider rational taciturn rumor propagation proposed by improving model, which solves overspread problem model. further studies...

Many real networks, such as infrastructure are interdependent and their structure is influenced by spatial constraints. However, the existence of constraints dependency links makes system more vulnerable to an initial random failure. In this paper, we model network with strength embedding length ?. We propose effective recovery strategy which recovers boundary failed nodes during cascading failures probability . find that without changing transition type, both range duration failure can be...

With the wide application of lithium-ion battery in various fields, security and reliability have attracted great attention. Under mode continuous development Internet vehicles technology, will be connected with each other future, hackers attack energy system vehicle. However, health assessment can timely grasp running state power system, so as to realize active defense against hacker attacks. This paper proposes a method for batteries using incremental capacity analysis weighted Kalman...

In this paper, we model epidemic spreading by considering the mobility of nodes in complex dynamical network based on mean field theory using differential equations. Moreover, a resistance factor which can characterise impact individual's difference propagation dynamics is proposed influence total number connections and continuous time to remain contact. The effect heterogeneity evolution process explored simulation. Extensive simulations are conducted study key parameters them dynamics,...

Complex networks have been widely studied recent years, but most researches focus on the single, non-interacting networks. With development of modern systems, many infrastructure are coupled together and therefore should be modeled as interdependent For networks, failure nodes in one network may lead to dependent other This happen recursively a cascade. In real world, different carry traffic loads. Overload load redistribution more nodes' failure. Considering dependency between load, small...

Due to the rapid development of information technology, a huge and complex traffic network has been established across various sectors including aviation, aerospace, vehicles, ships, electric power, industry. However, because complexity diversity its structure, is vulnerable be attacked faces serious security challenges. Therefore, this paper innovatively proposes resilience recovery method based on trend forecasting. In paper, risk value introduced into analysis fault propagation process,...

In the modern society, most networked critical infrastructures are coupled together and can be modeled as interconnected networks. Cascading failures caused by overload in networks extremely complicated still lack of corresponding researches. It is crucial to explore a systematic design load management method for complex network. this paper, probabilistic heterogeneous capacity allocation model preventive redistribution firstly proposed. Based on novel models failure cascades studied detail....

In this brief, the robust trajectory tracking control problem is addressed for uncertain quadrotor systems with time delays. The dynamical model an under‐actuated system subject to non‐linearity, unmodelled uncertainties, parametric perturbations, external disturbances, and state input A cascade controller developed by backstepping method compensation theory. Theoretical analysis given prove that errors can converge into neighbourhood of origin ultimately. An on‐line unidirectional tuning...

This paper proposes a novel performance degradation assessment method for bearing based on ensemble empirical mode decomposition (EEMD), and Gaussian mixture model (GMM). EEMD is applied to preprocess the nonstationary vibration signals get feature space. GMM utilized approximate density distribution of lower-dimensional space processed by principal component analysis (PCA). The confidence value (CV) calculated overlap between baseline that testing indicate bearing. experiment results...

Remaining useful life (RUL) prognosis of bearing is an enabling step for efficient implementation condition based maintenance. Through intelligent tracking features, status can be monitored and a rough RUL calculated. This paper presents application important Bayesian machine learning method named Gaussian Process Regression (GPR) features tracking. The process model provide variance around its mean prediction to describe associated uncertainty in the evaluation prediction. In this case, GPR...

In this article, a novel method of deep learning based on wavelet transform and perceptron neural networks (DPNNs) is proposed to predict the remaining useful life (RUL) bearings. The approach first extracts from recorded signals energy features by packet transform. After training these data, DPNN model can be used RUL bearing. To verify model, compared with least-squares support vector machine (LS-SVM) long short-term memory (LSTM). experimental results illustrate that effectively bearing...

The article focuses on the design and application of an active reconfigurable controller that mitigates effects gust load actuator faults a flexible aircraft. A novel integrated adaptive output feedback scheme is investigated to address faults. real-time fault values provided by estimation module are considered in control law improve fault-tolerant capability. estimate gains calculated analyzing stability overall system. proposed simulated using aircraft model with discrete '1-cosine' gust,...

The paper regards the complex dynamical network (CDN) as a static with temporal characteristics so to consider its dynamic behavior. influence factor and dynamics laws in CDN are explored by using methods of simulation statistical physics. It is found that both behavior nodes number initial disturbed can accelerate failure propagation enlarge scope. There phase transformation point repair ratio increase time τ for homogeneous (the value all same). Beyond threshold, entire will collapse;...

Active fault-tolerant control systems perform fault diagnosis and reconfigurable control. There is a bidirectional uncertainty between them, an integrated scheme proposed here to account for that. The system considers both actuator sensor faults, as well the external disturbance. diagnostic module designed using unknown input observer, controller constructed on basis of adaptive method. strategy presented, stability overall analyzed. Moreover, different kinds anti-windup techniques are...