TDFA-band (2-µm waveband) has been considered as a promising optical window for the next generation of optical communication and computing.Absorption modulation, one of the fundamental reconfigurable manipulations, is essential for large scale photonic integrated circuits.However, few efforts have been involved in exploring absorption modulation at TDFA-band.In this work, variable optical attenuators (VOAs) for TDFA-band wavelengths were designed and fabricated based on a silicon-on-insulator (SOI) platform.By embedding a short PIN junction length of 200 µm into the waveguide, the fabricated VOA exhibits a high modulation depth of 40.49dB at 2.2 V and has a fast response time (10 ns) induced by the plasma dispersion effect.Combining the Fabry-Perot cavity effect and plasma dispersion effect of silicon, the attenuator could achieve a maximum attenuation of more than 50 dB.These results promote the 2-µm waveband silicon photonic integration and are expected to the future use of photonic attenuators in crosstalk suppression, optical modulation, and optical channel equalization.
In modern industrial systems, components have complex interactions with each other, which makes it become a challenging task to identify the operational conditions of industrial systems. Considering that an industrial system, the embedded components and their interactions can be expressed as nodes and edges in a graph, respectively. Therefore, graph representation algorithms are powerful tools for fault diagnosis of industrial systems. As one of the most commonly used graph representation algorithms, graph neural networks (GNN) mainly follow the law of "learning to attend." GNN extract training data features learn the statistical correlations between features and labels, resulting in the attended graph favoring for accessing noncausal features as a shortcut for prediction. This shortcut feature is unstable and depends on the data distribution characteristics in the training dataset, which reduces the generalization ability of the classifier. By performing the causal analysis of GNN modeling for graph representation, the results show that shortcut features act as confounding factors between causal features and predictions, causing classifiers to learn wrong correlations. Therefore, to discover patterns of causality and weaken the confounding effects of shortcut features, a causal-trivial attention graph neural network strategy is proposed. First, node and edge representations are given by estimating soft masks. Second, through disentanglement, both causal features and shortcut features are obtained from the graph. Third, the backdoor adjustment of the causal theory is parameterized to combine each causal feature with a variety of shortcut features. Finally, comparative experiments on the three-phase flow facility dataset illustrate the effectiveness of the proposed method.
Abstract Monolithic integration of novel materials for unprecedented device functions without modifying the existing photonic component library is the key to advancing heterogeneous silicon photonic integrated circuits. To achieve this, the introduction of a silicon nitride etching stop layer at selective area, coupled with low-loss oxide trench to waveguide surface, enables the incorporation of various functional materials without disrupting the reliability of foundry-verified devices. As an illustration, two distinct chalcogenide phase change materials (PCM) with remarkable nonvolatile modulation capabilities, namely Sb 2 Se 3 and Ge 2 Sb 2 Se 4 Te 1 , were monolithic back-end-of-line integrated into silicon photonics. The PCM enables compact phase and intensity tuning units with zero-static power consumption. Taking advantage of these building blocks, the phase error of a push-pull Mach-Zehnder interferometer optical switch could be trimmed by a nonvolatile phase shifter with a 48% peak power consumption reduction. Mirco-ring filters with a rejection ratio >25dB could be applied for >5-bit wavelength selective intensity modulation, and waveguide-based >7-bit intensity-modulation photonic attenuators could achieve >39dB broadband attenuation. The advanced “Zero change” back-end-of-line integration platform could not only facilitate the integration of PCMs for integrated reconfigurable photonics but also open up the possibilities for integrating other excellent optoelectronic materials in the future silicon photonic process design kits.
Fault diagnosis and the prediction of remaining useful life play a key role in the Prognostics and Health management (PHM). One of the most important challenges in modern PHM is how to diagnose the fault of mechanical equipment accurately. Aiming to improve the fault diagnosis precision of the rotating machinery, a novel method of fault diagnosis combines mutual information models and second generation wavelet packet decomposition is presented in this paper. The traditional approaches to fault diagnosis always focus on the signals of a certain time. This method is different from traditional models because fault can be diagnosed more accurately by comparing the conditions of two different periods. Firstly, vibration signal of different times is extracted from the working bearings. Secondly, each frequency band's energy is calculated through the second generation decomposition and the energy of joint probability distribution of two different periods of time as well. Finally, the mutual information of two different periods of time is gained by using their joint probability distribution. A life test of a bearing is used to validate the proposed methodology and the results demonstrate that the proposed methodology is an effective tool to improve the accuracy of fault diagnosis of bearings running status.
<abstract><p>In this paper, the asymptotic behavior of solutions to a fractional stochastic nonlocal reaction-diffusion equation with polynomial drift terms of arbitrary order in an unbounded domain was analysed. First, the stochastic equation was transformed into a random one by using a stationary change of variable. Then, we proved the existence and uniqueness of solutions for the random problem based on pathwise uniform estimates as well as the energy method. Finally, the existence of a unique pullback attractor for the random dynamical system generated by the transformed equation is shown.</p></abstract>
ABSTRACT This paper investigates the finite‐time adaptive tracking control problem for non‐linear networked control systems with prescribed performance under unknown deception attacks. To mitigate the effects caused by unknown deception attacks, a series of auxiliary signals and attack compensators are reasonably constructed to overcome the unavailability problem of the compromised state variables. Besides, the neural network approximation technique is utilized to address unknown non‐linear terms and actuator deception attacks. Further, an equivalent system model is acquired by introducing the intermediate transformations of the tracking error and the prescribed performance function. Then, a finite‐time adaptive tracking controller is designed based on the neural network and the backstepping techniques. Moreover, it is mathematically rigorously proved that all the signals of the closed‐loop system are bounded and the tracking error converges within the predefined boundary in a finite‐time. Finally, an example application of a single‐link robotic arm system is applied to verify the effectiveness of the designed control algorithm.
Brownian motion in a washboard potential has practical significance in investigating a lot of physical problems such as the electrical conductivity of super-ionic conductor, the fluctuation of super-current in Josephson junction, and the ad-atom motion on crystal surface. In this paper, we study the overdamped motion of a Brownian particle in a washboard potential driven jointly by a periodic signal and an additive Gaussian white noise. Since the direct simulation about stochastic system is always time-consuming, the purpose of this paper is to introduce a simple and useful technique to study the linear and nonlinear responses of overdamped washboard potential systems. In the limit of a weak periodic signal, combining the linear response theory and the perturbation expansion method, we propose the method of moments to calculate the linear response of the system. On this basis, by the Floquet theory and the non-perturbation expansion method, the method of moments is extended to calculating the nonlinear response of the system. The long time ensemble average and the spectral amplification factor of the first harmonic calculated from direct numerical simulation and from the method of moments demonstrate that they are in good agreement, which shows the validity of the method we proposed. Furthermore, the dependence of the spectral amplification factor at the first three harmonics on the noise intensity is investigated. It is observed that for appropriate parameters, the curve of the spectral amplification factor versus the noise intensity exhibits a peaking behavior which is a signature of stochastic resonance. Then we discuss the influences of the bias parameter and the amplitude of the periodic signal on the stochastic resonance. The results show that with the increase of the bias parameter in a certain range, the peak value of the resonance curve increases and the noise intensity corresponding to the resonance peak decreases. With the increase of the driven amplitude, comparing the changes of the resonance curves, we can conclude that the effect of stochastic resonance becomes more prominent. At the same time, by using the mean square error as the quantitative indicator to compare the difference between the results obtained from the method of moments and from the stochastic simulation under different signal amplitudes, we find that the method of moments is applicable when the amplitude of the periodic signal is lesser than 0.25.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)