1,256 publications from this institution
Graph convolutional networks (GCNs) as the emerging neural networks have shown great success in Prognostics and Health Management because they can not only extract node features but can also mine relationship between nodes in the graph data. However, the most existing GCNs-based methods are still limited by graph quality, variable working conditions, and limited data, making them difficult to obtain remarkable performance. Therefore, it is proposed in this paper a two stage importance-aware subgraph convolutional network based on multi-source sensors named I
In this work, we perform a large-scale investigation of teacher connections in online social media. To this end, we first construct a large dataset of teachers on Pinterest, an image-based popular online social media. Our dataset includes 540 teachers across 5 states and 48 districts, as well as thousands of connections they have established. Then, considering some crucial teacher-related attributes (e.g., their states and grade levels), we characterize direct and indirect teacher connections. Through this characterization, we discover that teachers are predominately connected to their peers in the same district or at least within the same state, and seldom there exist links between teachers outside their districts and states. This hinders the proper diffusion of information and many other advantages that a teacher-teacher connection in an online social media can bring about, e.g., getting advice from their peers. To alleviate this problem, we utilize advances in machine learning and propose a link recommendation system suggesting teachers connect with their similar peers on Pinterest. Our system's evaluation reveals that many new teacher-teacher connections are suggested, which leads to a more cohesive network among teachers rather than the existing localized ego networks.
Object Oriented Programming (OOP) is a programming paradigm which be highlighted using the 800xa system. This work will investigate the use of OOP programming for designing industrial application control based on a soft controller. The different types of objects which can be defined in different classes, declaration and implementation using the 800xa will be explored, thus highlighting the use of static, dynamic and embedded objects. In the next phase, an illustration on how to map and use these objects in ABB 800xA for efficient control, hence, providing a systematic approach for accessing any of the plant aspects using a unique plant object.
No abstract is provided for this article.
Abstract — This paper considers the problem of stability analysis for a class of production networks of autonomous work systems with delays in the capacity changes. The system under consideration does not share information between work systems and the work systems adjust capacity with the objective of maintaining a desired amount of local work in progress (WIP). Attention is focused to derive explicit sufficient delay-dependent stability conditions for the network using properties of matrix norm. Finally, numerical results are provided to demonstrate the proposed approach. I.
This paper presents the design and implementation of a robust drive train torque control for rotor inertia emulation in a dynamometer test bench application. The controller is part of a hardware-in-the-loop (HiL) framework for a 50 kW small test rig that enables a real-time interaction between test rig drive train and a rotor model. The control objective is to compensate undesired dynamics of the test rig, in order to guarantee emulation of model dynamics through HiL simulation within a desired bandwidth. Therefore, a robust H <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</inf> controller has been implemented taking advantage of the design in frequency domain. Furthermore, a Kalman filter has been incorporated providing the unmeasured variables in the presence of measurement noise. For the purpose of control performance evaluation, simulation results as well as experimental tests on the hardware are presented.
In this paper, the problem of exponential stability is studied for a class of Markovian jump neutral nonlinear systems with mixed neutral and discrete time delays. By Lyapunov-Krasovskii function approach, a novel mean-square exponential stability criterion is derived for the situation that the system's transition rates are partially or completely accessible. Finally, some numerical examples are provided to illustrate the effectiveness of the proposed methods.
This paper deals with the problem of observer-based controller design for a class of nonlinear systems subject to unknown inputs. A novel method is presented to design a controller using estimated state variables which guarantees all the state variables of the closed-loop system converge to the vicinity of the origin and stay there forever. This is done via satisfying several sufficient conditions in terms of nonlinear matrix inequalities. In light of linear algebra, particularly matrix decompositions, the achieved conditions will be converted to a Linear Matrix Inequality (LMI) problem to facilitate the procedure of computing the observer and controller gains. Finally, the effectiveness of the proposed method is illustrated by implementing on a highly nonlinear chaotic system.
An input–output approach to the stability and stabilization of uncertain Takagi–Sugeno (T–S) fuzzy systems with time-varying delay is proposed in this paper. The time-varying parameter uncertainties are assumed to be norm-bounded, and the delay is intervally time varying. A novel method is employed to approximate the time-varying delay, based on which the considered system is transformed into a feedback interconnection form. The new formulation of the system is comprised of a forward subsystem with constant time delay and a feedback subsystem embedding the uncertainties. By applying the scaled small-gain theorem to the converted system, less conservative stability and stabilization criteria are obtained. Moreover, the applicability of the proposed approach to the robust case is simpler since both delay and parameter uncertainties are processed in a unified framework. Numerical experiments are performed to illustrate the advantage of the proposed techniques.
An exponential H8 synchronization method is addressed for a class of uncertain master and slave neural networks with mixed time-delays, where the mixed delays comprise different neutral, discrete and distributed time-delays. An appropriate discretized Lyapunov-Krasovskii functional and some free weighting matrices are utilized to establish some delay-dependent sufficient conditions for designing a delayed state-feedback control as a synchronization law in terms of linear matrix inequalities under less restrictive conditions. The controller guarantees the exponential H8 synchronization of the two coupled master and slave neural networks regardless of their initial states. Numerical simulations are provided to demonstrate the effectiveness of the established synchronization laws.Request access from your librarian to read this chapter's full text.
No abstract is provided for this article.
The problem of sliding mode control design for nonlinear switching semi-Markovian jump delayed systems is investigated in this paper. By choosing a linear switching surface function, we first derive reduced-order sliding mode dynamics; Then, the property of sliding mode dynamics with generally uncertain transition rates is analyzed by checking a set of new conditions; Further, an adaptive sliding mode control law is constructed to ensure the finite-time reaching condition. Finally, a practical example is provided to disseminate the effectiveness of the proposed method numerically.
SUMMARY This paper investigates the problem of delay‐dependent dynamic output feedback control for a class of discrete‐time Markovian jump linear systems (MJLSs). The systems under consideration are subject to time‐varying delay and defective mode information. The defective transition probabilities comprise of three types: exactly known, uncertain, and unknown. By employing a two‐term approximation for the time‐varying delay, the original MJLSs can be equivalently converted into a feedback interconnection form, which contains a forward subsystem with constant time‐delays and a feedback one with norm‐bounded uncertainties. Then, based on the scaled small‐gain theorem, the problem is therefore recast as an control problem in the face of uncertainties via an input–output framework. It is shown that the explicit expressions of the desired controller gains can be characterized in terms of strict linear matrix inequalities via some linearization techniques. Simulation studies are performed to illustrate the effectiveness and less conservatism of the proposed methods. Copyright © 2013 John Wiley & Sons, Ltd.
