Aiming at the current insufficiency of search engines alone in adequately addressing information overload across various domains, a graph neural network recommendation system incorporating review text representations was designed. This work includes designing an attention mechanism-based review feature extraction module, which inputs text segments within a sliding window into an attention network to compute the attention weights of words in context. Simultaneously, the traditional random negative sampling method was improved by replacing the traversal step unfavorable for parallel computation in conventional random negative sampling algorithms with matrix computations, thereby enhancing the algorithm's computational efficiency on parallel platforms. In addition, a hybrid negative sampling strategy was designed for this model, which employs the improved random negative sampling algorithm in the early training phase to rapidly iterate model parameters and adopts the positive and pruned negative sampling algorithm in the later training phase to sample high-quality negative samples, providing reliable support for further improving model accuracy. During the experimental phase, the effectiveness of the model and the hybrid negative sampling strategy was validated on three real-world open-source recommendation datasets.

Meta-learning methods are an effective approach to tackling the challenges of few-shot learning. However, current meta-learning approaches based on artificial neural networks do not introduce brain-inspired learning mechanisms, thereby hindering these models from achieving few-shot learning efficiencies and flexibilities comparable to human performance. Furthermore, existing brain-inspired spiking neural network approaches for meta-learning fail to meet Dale’s law constraints on excitation–inhibition balance during synaptic plasticity-driven weight regulation. This limitation compromises both the biological plausibility and the learning capability of spiking meta-learning. This work introduces a Dale’s law-constrained neural excitation–inhibition balance mechanism for meta-learning the initialization of spiking neural networks, constructing internal representations broadly applicable to many few-shot tasks. Specifically, to ensure strict compliance with Dale’s law (neurotransmitter type invariance) during synaptic weight updates, we integrate a modified projected gradient descent method with synaptic plasticity, thereby maintaining the invariance of neural excitatory and inhibitory properties. Building upon this, we incorporate the Dale’s law-constrained excitation–inhibition balance mechanism into spiking meta-learning and propose the excitation–inhibition balance-based spiking meta-initialization (EI-SMI) algorithm. Next, to enhance the stability and generalization capability of spiking meta-initialization, we employ a multistep weighted loss strategy to guide the network’s meta-learning process. Extensive experiments on five benchmark datasets (Omniglot, Mini-ImageNet, CUB, CIFAR-FS, and CWRU) demonstrate that the proposed EI-SMI outperforms early artificial neural network-based meta-learning approaches on certain specialized tasks and surpasses Spiking MAML across all tasks. The experimental results indicate that our method serves as a solid empirical foundation for developing meta-learning algorithms that are both biologically plausible and mechanistically credible.

This article investigates the optimal control problem of a stochastic nonlinear strict-feedback system subjected to both Wiener and Poisson noises. Since the strict-feedback system does not satisfy the matching condition and involves unknown nonlinear terms and unmeasurable stochastic noises in modeling, an optimized backstepping (OB) technique based on the reinforcement learning (RL) strategy is adopted to design the controller within the identifier–critic–actor architecture. However, the OB technique needs to construct all the virtual and actual controllers as the optimal solutions of their respective subsystems, which inevitably increases the complexity of the algorithm. To alleviate this situation, a novel RL method is proposed, so that the optimized control is unaffected by the disturbances induced by both Wiener and Poisson noises. In addition, an adaptive neural network identifier is incorporated into the RL framework to ensure that the proposed control scheme can be smoothly applied to the unknown nonlinear dynamic system. Finally, a vehicle tracking control example is presented to demonstrate the effectiveness of the proposed method.

Large group decision-making (LGDM) involves multiple decision-makers (DMs) and criteria, frequently resulting in conflicts and inconsistencies. This study proposes a novel conflict resolution model based on dynamic trust relationships to effectively identify and address potential disputes in LGDM. First, hesitant fuzzy 2-tuple linguistic sets (HF2TLSs) are utilized to accurately capture DMs’ preferences in uncertain environments. Then, a Markov trust state transition model is developed to capture the dynamic evolution of trust relationships. Next, an enhanced PageRank algorithm, built on trust networks, is employed to determine the weights of DMs and subgroups, thereby improving decision quality. Additionally, multidimensional conflict detection indicators are introduced to quantify cognitive and interest conflicts among subgroups. Finally, an adaptive conflict resolution mechanism is presented to balance heterogeneous interests and achieve agreement. An illustrative example validates the model, with comparative analyses demonstrating its rationality and superiority.

In this article, we propose an event-triggered interval observer fault diagnosis scheme for cyber–physical dc microgrids. First, a distributed interval observer is designed for each distributed generation unit interconnected by a power line. Then, an adaptive periodic event-triggered mechanism is put forward for saving communication cost. With the bounded fault signal and disturbances, the observer gain and event-triggered parameter can be determined by involving disturbance robustness, fault sensitivity, nonnegativity conditions, and regional pole placement for fast fault detection simultaneously. Based on it, the convergence rate of the state error dynamics can be enhanced to implement fast fault detection. It is shown that the proposed interval observer does not need to develop the residual evaluation function and threshold generator since zero is a natural threshold. The effectiveness and superiority of the proposed scheme are verified through simulations.

In the field of music generation using algorithms, processing time-series data has consistently been a complex task. To improve music generation with long sequences, insightful-unit-conditional variational autoencoder is proposed, which can enhance unit-conditional variational autoencoders with an improved attention mechanism. This model integrates TransformerXLs recurrent mechanism and relative positional encoding with measure-level granularity. For practical applications, a scheme is addressed that uses optical flow to extract motion features from video frames, quantifying motion rate and intensity. Furthermore, a dynamic correlation method is proposed to align video motion features with musical rhythm, guiding the model to generate melodies that match the videos rhythm.

Existing self-supervised multivariate time series anomaly detection methods struggle with interference among variables during reconstruction. They also tend to miss capturing critical anomaly information, resulting in unsatisfactory performance, especially in scenarios with strong mechanistic contexts. To this end, we propose a targeted anomaly detection algorithm called inference stacked recurrent autoencoder (ISRAE). Its key contribution lies in the design of a specific inference kernel, derived from specialist knowledge, which captures the strong mechanistic relationships among variables. This kernel is then fused with the multidimensional anomalies predicted by the SRAE, which mitigates interference among variables through the stacking technique. Furthermore, a novel differential constraint is introduced into the loss function, which not only highlights anomaly reconstruction errors, but also smooths the reconstructions, enhancing overall detection performance. Comprehensive comparison experiments and ablation studies show that ISRAE achieves superior anomaly detection performance under strong mechanistic contexts and highlight the importance of each key module in ISRAE.

Multiagent formation obstacle avoidance is a crucial research topic in the field of multiagent cooperative control, and deep reinforcement learning has shown remarkable potential in this domain. However, most existing studies are not fully distributed and often involve relatively simple scenarios. In this article, we propose an advanced method based on multiagent deep reinforcement learning to address formation and obstacle avoidance in dynamic obstacles environments. For handling complex environments with an unknown number of obstacles, we use long short-term memory (LSTM) networks to encode dynamic obstacles, thereby improving the efficiency of obstacle avoidance. Our method achieves formation and obstacle avoidance in scenarios with both dynamic and static obstacles, where agents coordinate through fully independent and autonomous decision-making. We utilize the multiagent proximal policy optimization (MAPPO) algorithm for centralized training and distributed execution, enhancing the agents' formation and obstacle avoidance capabilities in complex settings. Through simulation and real-world experiments, and by comparing with benchmark methods, we demonstrate significant improvements in formation effectiveness and obstacle avoidance success rates, showcasing the superiority and practicality of our proposed approach.

Rapid developments of offshore wind industry offer a strong demand opportunity for offshore wind turbine remote diagnosis. As offshore wind turbines are often located in harsh and communication-constrained environments, the collection and transmission of data is severely restricted, which poses a serious challenge to the conventional centralized diagnostic paradigm that relies on data aggregation. To address this challenge, we propose a novel event-triggered federated learning framework for decentralized fault diagnosis of offshore wind turbines. Specifically, federated learning is first employed to learn decentralized local knowledge from geographically distributed offshore wind turbines, so that the communication objects are transformed from massive raw data into learned parameters, thereby relieving the communication burden. Then, we design an event-triggered communication mechanism and incorporate it into federated learning, the core of which is to modify the communication requirement from uploading all trained parameters periodically to communicating only when necessary. The proposed framework is verified by a real-world offshore wind turbine dataset from six large wind farms in China. An ablation study shows that the proposed framework can maintain high diagnostic performance while reducing communication costs. A comprehensive comparison based on three benchmark models demonstrates that the proposed framework can reduce the communication burden by up to 63% while obtaining better diagnostic performance.Note to Practitioners—This study was motivated by the problem of collaborative diagnosis of distributed offshore wind turbines under the constraints of data privacy and communication overhead. The method employs a federated learning-based fault diagnosis framework, which permits to obtain global fault diagnosis knowledge without aggregating raw data scattered in each end device, thus avoids the risk of data leakage. Moreover, a strategy integrating parameter variation and accuracy gain is designed to avoid communication redundancy for collaborative training. The practicability and superiority of our proposed framework is demonstrated using extensive experiments against actual industrial data collected from six offshore wind farms.

This article investigates an event-triggered interval observer (ETIO) fault detection and isolation method for multiagent systems. First, an event-triggered mechanism is developed to reduce unnecessary communication transmission. Then, a distributed ETIO is designed by combining an interval observer and the proposed event-triggered mechanism. Furthermore, for achieving the desired tradeoff between the robustness to disturbances and the sensitivity to faults, the ETIO is formulated as a multiobjective optimization with$ l_{1}$$/$$ H_{\infty}$performance. Second, a bank of ETIOs are interpreted to isolate the faulty agent on a local agent using only the output information from itself and its neighbors. Comparison result with the existing method is given to highlight the superiority of our methodology. Finally, the multiunmanned aerial vehicles system is utilized as the case research, and specific simulation results are presented.

Quadrotor UAVs, renowned for their agility and versatility, are extensively utilized in a range areas. However, their inherent underactuated dynamic characteristics render them particularly vulnerable to external disturbances and systemic failures. To address this issue, our study introduces a hybrid control method tailored to combat the most prevalent types of drone failures—actuator faults. This innovative approach leverages reinforcement learning to enhance fault tolerance. Specifically, we employ reinforcement learning techniques to output compensatory control signals that bolster the core functionalities of the base controller. This integration aims to preserve the stability and continuity of mission-critical tasks even in the face of operational faults, thereby ensuring robust safety controls. We utilized the proximal policy optimization algorithm for the strategic training of our control systems. We test in both simulated environments and real-world scenarios was conducted to evaluate the efficacy of our control strategy under conditions of actuator failure. The results affirm that our method significantly enhances the safety and stability of drone operations, maintaining control integrity during rotor failures.

Despite the widespread use of Fourier transform (FT) networks and generative adversarial networks (GANs) in audio signal processing, their practical effectiveness in unsupervised offline systems has not yet reached a fully satisfying level. Accumulating substantial experience in recent years, this article showcases how to construct an optimized, efficient music generation system. In the proposed system, the short-time Fourier transform is employed to divide a long music signal into equally sized short melodic segments. Each short melodic segment undergoes FT, and a nonautoregressive parallel WaveGAN system is trained by jointly optimizing multiresolution spectrograms and adversarial loss functions. This approach effectively captures the time–frequency distribution of real music waveforms. In essence, the proposed music generation system is a self-feedback unsupervised model relying on specific melody and note model pruning techniques. To further refine the music evaluation mechanism, in addition to conducting data analysis on the output melodies, subjective evaluation mechanisms are also incorporated.

Multivariate time-series (MTS) collected from multiple sensors on industrial pumps often exhibit concept drift and noise contamination due to variable working conditions and complex environments. To detect anomalies in such MTS, we propose a novel model called spectral residual with self-attention variational autoencoder (SR-SAVAE). Specifically, the spectral residual operation is used to mitigate concept drift, while the variational inference combined with a total variation regularization is used to address the issue of noise contamination. Experimental results on three public datasets indicate that the SR-SAVAE model achieves good anomaly detection results for general MTS. More importantly, compared to other state-of-the-art models on a private dataset about pumps, the results illustrate the superiority of the SR-SAVAE model in anomaly detection for MTS with concept drift and noise contamination. Finally, ablation studies on the SR-SAVAE model detail the efficacy of each component.

Precise timing over timestamped packet-exchange communication is an enabling technology in the mission-critical industrial Internet of Things (IIoT), particularly when satellite-based timing is unavailable. The main challenge is to ensure timing accuracy when the clock synchronization system is subject to disturbances caused by the drifting frequency, time-varying delay, jitter, and timestamping uncertainty. In this work, a robust packet-coupled oscillators (R-PkCOs) protocol is proposed to reduce the effects of perturbations manifested in the drifting clock, timestamping uncertainty, and delays. First, in the spanning-tree clock topology, time synchronization between an arbitrary pair of clocks is modeled as a state-space model, where clock states are coupled with each other by one-way timestamped packet exchange (referred to as packet coupling), and the impacts of both drifting frequency and delays are modeled as disturbances. A static output controller is adopted to adjust the drifting clock. The$H_{\infty }$robust control design solution is proposed to guarantee that the ratio between the modulus of synchronization precision and the magnitude of the disturbances are always less than a given value. Therefore, the proposed time synchronization protocol is robust against the disturbances, which means that the impacts of drifting frequency and delays on the synchronization accuracy are limited. The one-hour experimental results demonstrate that the proposed R-PkCO’s protocol can realize time synchronization with the precision of 6$\mu \text{s}$in a 21-node IEEE 802.15.4 network. This work has widespread impacts in the process automation of automotive, mining, oil, and gas industries.

The deployment of renewable energy in power systems may raise serious voltage instabilities. Electric vehicles (EVs), owing to their mobility and flexibility characteristics, can provide various ancillary services including active and reactive power. However, the distributed control of EVs under such scenarios is a complex decision-making problem with enormous dynamics and uncertainties. Most existing literature employs model-based approaches to formulate active and reactive power control problems, which require full models and are time-consuming. This article proposes a multiagent reinforcement learning algorithm featuring a deep deterministic policy gradient (DDPG) method and a parameter sharing framework to solve the EVs’ coordinated active and reactive power control problem toward both demand-side response and voltage regulations. The proposed algorithm can further enhance the learning stability and scalability with privacy perseverance via the location marginal prices. Simulation results based on a modified IEEE 15-bus network are developed to validate its effectiveness in providing system charging and voltage regulation services. The proposed location marginal price (LMP) PSDDPG algorithm is evaluated to achieve 38%, 16%, and 25% speedup, and 1.58, 0.69, and 0.27 times higher reward over the benchmarks DDPG, TD3, and LMP-DDPG, respectively.

This article proposes a novel data-driven gap metric fault detection and isolation (FDI) approach for small multiplicative fault. First, the scheme of model-based fault classification and gradation is developed by means of the gap metric. Subsequently, the data-driven gap metric is utilized to detect a small fault via the mechanism model. Furthermore, fault detectability criterion is derived with the help of the developed fault detectability indicator. The relationship between fault detectability indicator and fault detection index is then investigated to analyze fault detection performance. To enhance fault isolability, a solution of appropriate fault cluster center model and radius is provided under the condition of fault isolation. Third, a gap metric fault-tolerant control strategy is exploited to guarantee system stability when a large fault is diagnosed by the developed FDI approach. The speed regulation of dc-motor and dc–dc converter are used for simulation and experiment verifications. Moreover, the comparison results and Monte Carlo simulation demonstrate the superiority and reliability of the proposed method.

Time-sensitive wireless applications have strict requirements on real-time data transmission and control operation. Even though time synchronization has been extensively studied for providing a common timing among distributed wireless nodes, there still exists a lack of research for low-accuracy and large-drifting clocks, such as internal Resistor-Capacitor (RC) oscillator clocks with around$4\times 10^{5}$parts per million (ppm) frequency drift, which are widely used in wearable sensor systems. This paper proposes a Proportional Packet-Coupled Oscillators (P-PkCOs) protocol for synchronizing poor-performing internal RC oscillator clocks with high disturbances in the single-cluster wireless network. The behaviour of such a drifting clock is described by a non-identical and time-varying model. To achieve time synchronization on low-accuracy internal RC oscillator clocks, a packet-coupled synchronization scheme is proposed for adjusting drifting clocks via the proportional control-based correction scheme. The RC oscillator frequency in an embedded system cannot be corrected, and this work utilizes the clock threshold adjustment as a substitute for frequency correction. The stability region of controller parameters is given to guarantee that the clock threshold approaches a value, which is jointly determined by the nominal threshold and the corresponding clock frequency. We also propose a linear matrix inequality condition to prove that the P-PkCOs performance is robust against the large clock disturbances. We demonstrate the implementation of P-PkCOs. The experimental results show that P-PkCOs can achieve and maintain robust time synchronization on the internal RC oscillator clocks.

Non-rechargeable batteries remain as the main source of energy for small systems, owing to their unique advantages in energy density, safety, reliability and sustainability. Accurate prediction of the remaining useful life of the battery is not only beneficial to maintenance and production safety, but also can be regarded as a starting point for possible secondary life applications. In this study, an interactive attention sequence-to-sequence network is proposed for the remaining useful life prediction of the non-rechargeable batteries. The proposed approach can effectively extract the degenerate information of each variable-length sequence and dynamically weight the sequence features of different dimensions. For illustration, a case of primary battery dataset collected from the power supply system of 139 vibration sensors is utilized. The extensive experiments verify the effectiveness of the proposed approach.

Aiming to overcome the difficulty to obtain the fault data of practical system, a fault detection approach using health data only is proposed based on the whole space of the system being divided into the fault status and the fault-free status. Firstly the time series of observation window is generated by a deconvolutional network with an input of initial data obtained by Monte Carlo method. The probability distribution of generated data approximates to the actual sample data by discriminator of generative adversarial network. Through continuous iteration, the health probability distribution is finally obtained in the whole space. Concurrently the discriminator is evolved into a fault detector which realizes the detection of new data. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.

In this article, an adaptive sliding-mode control scheme is developed for a class of uncertain quarter vehicle active suspension systems with time-varying vertical displacement and speed constraints, in which the input saturation is considered. The integral terminal SMC is adopted to improve convergence accuracy and avoid singular problems. In addition, neural networks are used to model unknown terms in the system and the backstepping technique is taken into account to design the actual controller. To guarantee that the time-varying state constraints are not violated, the corresponding Barrier Lyapunov functions are constructed. At the same time, a continuous differentiable asymmetric saturation model is developed to improve the stability of the system. Then, the Lyapunov stability theory is used to verify that all signals of the resulting system are semi globally uniformly ultimately bounded, time-varying state constraints are not violated, and error variables can converge to the small neighborhood of 0. Finally, results of the simulation of the designed control strategy are given to further prove the effectiveness.

Impulsive control is widely applied to achieve the consensus of multiagent networks (MANs). It is noticed that malicious agents may have adverse effects on the global behaviors, which, however, are not taken into account in the literature. In this study, a novel delayed impulsive control strategy based on sampled data is proposed to achieve the resilient consensus of MANs subject to malicious agents. It is worth pointing out that the proposed control strategy does not require any information on the number of malicious agents, which is usually required in the existing works on resilient consensus. Under appropriate control gains and sampling period, a necessary and sufficient graphic condition is derived to achieve the resilient consensus of the considered MAN. Finally, the effectiveness of the resilient delayed impulsive control is well demonstrated via simulation studies.

This is the editorial of the SS entitled ‘`Digital Twinning: Integrating AI-ML and Big Data Analytics for Virtual Representation’'.

This article concentrates on the event-driven controller design problem for a class of nonlinear single input single output parametric systems with full state constraints. A varying threshold for the triggering mechanism is exploited, which makes the communication more flexible. Moreover, from the viewpoint of energy conservation and consumption reduction, the system capability becomes better owing to the contribution of the proposed event-triggered mechanism. In the meantime, the developed control strategy can avoid the Zeno behavior since the lower bound of the sample time is provided. The considered plant is in a lower triangular form, in which the match condition is not satisfied. To ensure that all the states retain in a predefined region, a barrier Lyapunov function (BLF) based adaptive control law is developed. Due to the existence of the parametric uncertainties, an adaptive algorithm is presented as an estimated tool. All the signals appearing in the closed-loop systems are then proven to be bounded. Meanwhile, the output of the system can track a given signal as far as possible. In the end, the effectiveness of the proposed approach is validated by an aircraft wing rock motion system.

Privacy protection as a major concern of the industrial big data enabling entities makes the massive safety-critical operation data of a wind turbine unable to exert its great value because of the threat of privacy leakage. How to improve the diagnostic accuracy of decentralized machines without data transfer remains an open issue; especially these machines are almost accompanied by skewed class distribution in the real industries. In this study, a class-imbalanced privacy-preserving federated learning framework for the fault diagnosis of a decentralized wind turbine is proposed. Specifically, a biometric authentication technique is first employed to ensure that only legitimate entities can access private data and defend against malicious attacks. Then, the federated learning with two privacy-enhancing techniques enables high potential privacy and security in low-trust systems. Then, a solely gradient-based self-monitor scheme is integrated to acknowledge the global imbalance information for class-imbalanced fault diagnosis. We leverage a real-world industrial wind turbine dataset to verify the effectiveness of the proposed framework. By comparison with five state-of-the-art approaches and two nonparametric tests, the superiority of the proposed framework in imbalanced classification is ascertained. An ablation study indicates that the proposed framework can maintain high diagnostic performance while enhancing privacy protection.

Owing to its unique, concealment and easy customization by combining different wrist and hand gestures, high-density surface electromyogram (HD-sEMG) is recognized as a potential solution to the next generation biometric authentication, which usually adopts a wireless body sensor network (BSN) to acquire the multi-channel HD-sEMG biosignals from distributed electrode arrays. For more accurate and reliable classification, biometric authentication requires the distributed biosignals to be sampled simultaneously and be well aligned, which means that the sampling jitters among the arrays need to be tiny. To synchronize data sampling clocks of a cluster of BSN nodes for biometric authentication, this article modifies the packet-coupled oscillators protocol by using a dynamic controller (D-PkCOs). This protocol only involves one-way single packet exchange, which reduces the communication overhead significantly. For the purpose of maintaining precise sampling of these BSN nodes subject to drifting clock frequency and varying delays, the dynamic controller is designed via the$H_\infty$robust method, and it is proved that all the BSN nodes’ sampling jitters are bounded. The experimental results demonstrate that the D-PkCOs protocol can keep the sampling jitters less than a microsecond in a 10-node IEEE 802.15.4 network. The application of D-PkCOs to the BSN shows that the HD-sEMG signal with a high signal-to-noise ratio is obtained, which leads to better gesture classification performance.

A novel compensation control method for heterogeneous multiagent systems under Denial-of-Service (DoS) attacks and transmission delays is investigated in this article. This control method has all the advantages of the cloud-based computation strategy, the adaptive event-triggered strategy, and the predictive control scheme. The adaptive event-triggering mechanism can adjust the event numbers adaptively, the predictive control can reduce or eliminate the negative effects brought out by both DoS attacks and transmission delays actively, while the cloud-based computation strategy can eliminate the negative effects completely as the same as there are no DoS attacks and transmission delays. Through the interval decomposition skill and the augmented system modeling method, the compensated geschlossenes system model is established. Moreover, the joint design for the feedback gain matrices and the event-triggered parameters is implemented. In the simulation part, five VTOL aircraft are used to demonstrate the theoretical results.

Data-driven fault diagnosis and classification for wind turbine systems have received much attention due to a large amount of data available recorded by supervisory control and data acquisition (SCADA) systems and smart meters. It is of interest but challenging to diagnose and classify multiple faults occurring simultaneously in a system monitored. In this study, a data-driven and supervised machine learning-based fault diagnosis and classification algorithm is addressed by the combination and consolidation among Hilbert-Huang Transformation (HHT), Multi-Linear Principal Component Analysis (MPCA), and Support Vector Machine (SVM) to enhance the feasibility and capability of fault diagnosis and classification for systems subjected to multiple faults. The algorithm proposed is applied to the 4.8 MW wind turbine benchmark model, where multiple actuator faults are taken into considerations. The effectiveness of the methodology is demonstrated by using intensive simulations and comparison studies.

The great success of deep neural network (DNN) in image field stimulates its application in fault detection and diagnose. However due to the limitation of system security, it is impossible to obtain complete fault data as the training database for neural network, so that it is challenging to identify a fault that never occurred before. In this paper, an ensemble approach is proposed to adapt to a new fault by adding output branches of the neural network. Firstly, the time series are transferred to numerous imaging matrixes. The intrinsic characteristics of the matrixes are then extracted using deep neural network which are used to judge whether it is a new fault according to the distance criterion. For a new fault, the DNN will retrain by transferring learning in order to reduce the computation and training time. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.

The problem of fault diagnosis (FD) and fault tolerant control for a class of Takagi–Sugeno (T–S) fuzzy stochastic distribution control systems subject to sensor and actuator faults is discussed in this article. First, fuzzy logic models are used to approximate the output probability density function (PDF). Next, an adaptive augmented state/FD observer is proposed to estimate the system state, sensor and the actuator faults simultaneously. New expected weights based on the sensor fault estimation information and a PI-type fuzzy feedback fault tolerant controller are designed to compensate the effect of sensor fault and actuator fault simultaneously. When the sensor fault occurs, the expected objective is redesigned to compensate the sensor fault. Meanwhile, the PI controller can compensate the effect of actuator fault, and the output PDF of the system can still track the desired PDF after the fault occurs. Finally, an example of quality distribution control in chemical reaction process is given to confirm the effectiveness of the algorithm.

This article concerns observer-based consensus compensation control for heterogeneous networked multi-agent systems under both networked Denial of Service (DoS) attacks and transmission delays. We propose a control method that combines the observer-based adaptive event-triggered control and the observer-based cloud predictive control, which can not only reduce the network transmission burden, but also can compensate for the negative effects caused by DoS attacks and transmission delays completely. The consensus conditions, the observer and controller gain matrices and the event-triggering parameter matrices are all simultaneously derived by using the linear matrix inequality method.

Data-driven fault diagnosis and classification for wind turbine systems have received much attention due to a large amount of data available recorded by supervisory control and data acquisition (SCADA) system and smart meters. It is challenging to diagnose and classify multiple faults occurring simultaneously in a system monitored. In this study, a data-driven fault diagnosis and classification algorithm is addressed by integrating fast Fourier transform (FFT) and multi-linear principal component analysis (MPCA) in order to enhance the capability of fault diagnosis and classification for systems subjected to multiple faults. The algorithm proposed is applied to a 4.8-MW wind turbine benchmark system, where multiple actuator faults are taken into accounts. The effectiveness of the algorithm is demonstrated by intensive simulations and comparison studies.

In this study, dimensionality reduction and feature extraction techniques are applied to fault classification for wind turbine systems using principal component analysis (PCA). Experimentations are performed in terms of two topologies of faulty datasets which are respectively actuator effectiveness loss and sinusoidal fault, occurring in a 4.8-MW wind turbine benchmark system. In order to evaluate the PCA technique, additive white Gaussian noise (AWGN) signals are introduced to the aforementioned two faulty conditions respectively. The experimental results demonstrate that PCA not only can extract the significant features but also distinguish different types of fault effectively in this wind turbine benchmark system.

In this paper, a deep learning fault detection approach is proposed based on the convolutional neural network in order to cope with one class of faults in wind turbine systems. Fault detection is very vital in nowadays industries due to the fact that instantly detection can prevent waste of cost and time. Deep learning as one of the powerful approaches in machine learning is a promising method to identify and classify the intrigued problems, which are hard to solve by classical methods. In this case, less than 5% performance reduction in generator torque along with sensor noise, which is challenging to identify by an operator or classical diagnosis methods is studied. The proposed algorithm, which is evolved from convolutional neural network idea, is evaluated in simulation based on a 4.8 MW wind turbine benchmark and the accuracy of the results confirms the persuasive performance of the suggested approach.

In this paper, a concise two-state clock model is proposed to describe the behaviour of drifting internal oscillator clocks with non-identical and time-varying frequencies, which provides a simple analytical solution of different coupling schemes' effects on synchronisation performance. The improved Pulse-Coupled Oscillators (PCO) scheme presented in this paper guarantees the synchronisation on non-identical and time-varying clocks. Meanwhile, it differs from classical PCO by scheduling the transmission of Syncs (i.e., the Pulses of clock) at different time slots. This collision-free pulse coupling scheme makes the concept of PCO applicable to wireless sensor networks, as it is practically impossible in realistic wireless networks for multiple oscillators to transmit Pulses at the same time. Furthermore, a proportional controller is adopted to correct the clock, and convergence and stability of the proposed algorithm are theoretically analysed, proved and evaluated by hardware experiments. The experimental results show that the improved PCO scheme can achieve synchronisation with the precision of around 400 μs on internal oscillator clocks of about 350, 000 ppm.

In this paper, a robust fault estimation approach is proposed for multi-input and multioutput nonlinear dynamic systems on the basis of back propagation neural networks. The augmented system approach, input-to-state stability theory, linear matrix inequality optimization, and neural network training/learning are integrated so that a robust simultaneous estimate of system states and actuator faults are achieved. The proposed approaches are finally applied to a 4.8 MW wind turbine benchmark system, and the effectiveness is well demonstrated.

As a key enabling technology in mission-critical Wireless Sensor Networks (WSNs), time synchronization provides a common timescale for distributed sensor nodes in many wireless applications, such as coordinated control and underwater navigation and tactical surveillance. Inspired by the behaviour of fireflies, along with mathematical model, Pulse-Coupled Oscillators (PCO), has been proposed to enable synchronization in complex networks, where all the PCO's firing signal Pulses are broadcasted simultaneously when synchronization is achieved. The requirement of zero-drift clock oscillators, fully-connected network and concurrent transmission of Pulses are, in reality, impossible to achieve. To avoid transmission collision and enable the PCO extension in the multi-hop WSNs, the desynchronization mechanism is adopted to enable the Pulse packets to be transmitted to the wireless channel in a uniformly distributed fashion. Due to the contention-free period's feature of low-latency, thereby avoiding the need to wait for a random and potentially long period until the channel is available, the PCO's Pulse packets are transmitted in the contention-free period of IEEE 802.15.4-2015 superframe. Thus, a novel state-space model for desynchronization-based pulse-coupled nonidentical oscillators is proposed to model a realistic drifting clock oscillator. Moreover, the timestamped Pulse packets are transmitted to determine the offset of connected sensor nodes, and an attenuated clock correction scheme is adopted to correct the local drifting clocks by using measured offset and skew. The intensive simulations of the three-hop three-cluster wireless network and the seven-hop linear network have been carried out to evaluate performance of timestamped PCO with desyn-chronization method.

In this paper, a model identification method based on artificial neural networks (ANN) for wind turbine dynamics is studied. Due to the fact that wind turbine has a nonlinear dynamics with partially measured states, ANN cannot be applied directly. To cope with this problem, first a Luenberger observer is designed to estimate the states (both measured and unmeasured ones) and then, for the nonlinear part, a multi-input multi-output (MIMO) back propagation neural-network based observer is proposed. By having an ANN model as the reference, a fault detection method is studied based on the residual of the system. This algorithm is evaluated in simulation on a 4.8 MW wind turbine benchmark and the results approve satisfactory performance of the proposed approach.

In this paper, a novel discrete-time sliding mode control (DSMC) method based on event-trigger strategy is proposed. The DSMC method here used is a chattering-free method. With the introduction of event-trigger strategy, the system performance is improved in terms of the control updating times. Hence, less resource is required in control execution. It is shown that the proposed control techniques ensure the reachability of the sliding surface with a small band. Finally, simulations are presented to verify the effectiveness of the proposed methods.

Time synchronization in wireless sensor networks,aiming to provide a common sense of timing among distributed sensor nodes, is a key enabling technology for many applications, such as collaborative condition monitoring, time-of-flight localization and underwater navigation and tactical surveillance. In order to solve the challenges of the manufacturing tolerance and working condition variations in any real-world environments, a novel state-space model for pulse-coupled non-identical oscillators is proposed to model a realistic clock oscillator with nonidentical and time-varying frequency. A state feedback correction, referred to as hybrid coupling mechanism, is also proposed to ensure the system move into steady state, thus achieving time synchronization in wireless sensor networks. Furthermore, the intensive simulations of single-hop wireless sensor networks have been carried out to evaluate the performance of proposed pulsecoupled non-identical oscillators. It is shown that a partially connected wireless network consisting of 50 non-identical pulsecoupled oscillators can achieve the synchronization with the precision of $40us.$

Today large-scale simulation applications are becoming common in research and industry. A significant fraction of them run on multi-core clusters. Current parallel simulation kernels use multi-process and multi-thread to exploit inter-node parallelism and intra-node parallelism on multi-core clusters. We exploit task-base parallelism in parallel discrete event simulation (PDES) kernels, which is more fine-grained than thread-level and process-level parallelism. In our system, every simulation event is wrapped to a task. Work-stealing task scheduling scheme is applied to achieve dynamic load balancing among the multi-cores, and a graph partitioning approach is applied in partitioning simulation entities among the cluster nodes. Experimental results show that our PDES kernel outperforms existing PDES kernels by fully exploiting task parallelism.

In this paper, a reinforcement learning approach is proposed to detect unexpected faults, where the noise-to-signal ratio of the data series is minimized for achieving robustness. The model parameter is taken as a special action of the reinforcement learning, and the policy valuation and policy improvement are utilized to find the parameters, which can make the estimated model consistent to the real-time system process. Compared with the model forecasted parameters and real-time process parameters, fault diagnosis is implemented. The fault degrees are also discussed by analyzing the distance differences between the healthy parameters and faulty parameters. The effectiveness of algorithm is demonstrated by a numerical simulation example.

As one of the most popular methods for image segmentation, fuzzy C-means algorithm suffers two unavoidable initialization difficulties including obtaining initial cluster centroids and deciding cluster number, which affect the algorithm performance. Motivated by the above, an automatic fuzzy clustering algorithm is proposed in this paper, where observation matrix, judgment matrix and set partitioning are used to select appropriate clustering number automatically. Experimental results show that automatic fuzzy clustering algorithm not only can spontaneously estimate the appropriate number of clusters but also can achieve better segmentation quality.

To achieve reliable operation of wind energy conversion technology, this ρ aper develops a robust observer-based fault tolerant control technique for wind turbine drive train systems in presence of simultaneous unknown inputs, faults and Brownian perturbations. Integration of several advanced techniques, namely, augmented approach, unknown input observer method, and linear matrix inequaity, is employed to estimate the means of the system states and the considered faults robustly. Based on the estimates, robust fault tolerant control strategy is implemented to drive the system trajectory convergent and eliminate the effects of faults from both actuators and sensors successfully. The control gains are selected to guarantee the convergence of the means of system states and com pens ate for the de grad ation caused by concerned faults. The ob server gain is determined via a linear matrix inequality optimization such that the closed-loop system is stochastically input-to-state stable satisfying required robust performance. The desi gned observer-based fault tolerant control c an make the over all system work in a steady condition and the system outputs c an be compensated to successfully track the healthy outputs in fault-free c ases. Finally, the proposed fault estimation-based fault tolerant control method is applied to a drive train system of the 4.8 MW benchmark wind wind turbine to validate the effectiveness.

Engineering systems are always subjected to faults or malfunctions due to age or unexpected events, which would degrade the operation performance and even lead to the operation failure-Therefore, there is a strong motivation to develop fault-tolerant control strategy so that the system can operate with tolerated perform ance de ggr ad ation-In this p ap er, a novel approach based on reinforcement leaning is proposed to design a fault-tolerant controller without need of the information on faults-T simulation example.

Nowadays, industrial plants are calling for high-performance fault diagnosis techniques to meet stringent requirements on system availability and safety in the event of component failures. This paper deals with robust fault estimation problems for stochastic nonlinear systems subject to faults and unknown inputs relying on Takagi-Sugeno fuzzy models. Augmented approach jointly with unknown input observers for stochastic Takagi-Sugeno models is exploited here, which allows one to estimate both considered faults and full system states robustly. The considered unknown inputs can be either completely decoupled or partially decoupled by observers. For the un-decoupled part of unknown inputs, which still influence error dynamics, stochastic input-to-state stability properties are applied to take nonzero inputs into account and sufficient conditions are achieved to guarantee bounded estimation errors under bounded unknown inputs. Linear matrix inequalities are employed to compute gain matrices of the observer, leading to stochastic input-to-state-stable error dynamics and optimization of the estimation performances against un-decoupled unknown inputs. Finally, simulation on wind turbine benchmark model is applied to validate the performances of the suggested fault reconstruction methodologies.

In this study, a robust fault reconstruction approach is proposed for the 4.8 MW wind turbine benchmark system. Firstly, through weighted combination of a number of locally valid linear systems, the nonlinear wind turbine model is well represented by a Takagi-Sugeno fuzzy model. Then, augmented system approach jointly with unknown input fuzzy observer technique are utilized to estimate faults and system states simultaneously, while decouple a part of unknown inputs which consist both system perturbations and Takagi-Sugeno modelling errors. After that, linear matrix inequality approach is used to ensure convergence of estimation error and attenuate the influences from un-decoupled unknown inputs. Finally, the proposed algorithms are demonstrated to be effective by using the 4.8 MW wind turbine benchmark system.

In this paper, linear parameter varying (LPV) modelling technique is addressed for modelling a wind turbine system, with real-time changing scheduling parameters. Based on the LPV wind turbine model, a LPV observer-based fault detection method is utilized to detect faults under four scenarios. The effectiveness of the proposed modelling and fault detection techniques is demonstrated by using the widely-recognized wind turbine benchmark system.

Wireless Sensor Actuator Networks (WSANs) have shown its advantages of flexibility and cable-free deployment in industrial applications. However, the sensors explore directly to the harsh environment and are the most prone to faults and malfunctioning, which poses a big challenge to the WSAN's applicability in applications with high safety requirements. This paper focuses on highly reliable WSANs equipped with double-sensing system and proposes a model-based on-line sensor fault detection scheme by exploiting information redundancy between sensors and actuators. An optimization method is developed to make the proposed scheme robust to the largely fluctuating end-to-end packet transmission delays due to wireless communications. The performance of the proposed sensor fault detection scheme is demonstrated by the simulation of a pick and place robot.

In this paper, an innovative unknown input observer (UIO) is developed to estimate the faults of the system subjected to faults and process disturbances. By representing the concerned faults as auxiliary states, an augmented system is constructed. By designing an unknown input observer for the augmented system, the simultaneous estimations of the system states and concerned faults can be obtained readily. In this study, the unknown input disturbance is not assumed to be decoupled completely. The integration of the UIO technique and the linear matrix inequality (LMI) optimization technique is proposed to decouple and attenuate the input disturbances. The existence condition of such an observer and the design method of the observer gains by using the LMI are addressed as well. Finally, the proposed algorithm is applied to the simulation study of a three shaft gas turbine system to show the excellent fault estimation performance.

Wind turbine system is a safety-critical system, which has the demand to improve the operating reliability and reducing the cost caused by the shut-down time and component repairing. As a result, condition monitoring and fault diagnosis have received much attention for wind turbine energy systems. Noticing that environmental disturbances are unavoidable, therefore how to improve the robustness of a fault diagnosis scheme against disturbances/noises has been a key issue in fault diagnosis community. In this investigation, a robust fault estimation approach with the aid of eigenstructure assignment and genetic algorithm (GA) optimization is presented so that the estimation error dynamics has a good robustness against disturbances. A simulation study is carried out for a 5MW wind turbine dynamic model, which has demonstrated the effectiveness of the proposed techniques.

This review paper is to give a full picture of fault detection and diagnosis (FDD) in complex systems from the perspective of data processing. As a matter of fact, an FDD system is a data-processing system on the basis of information redundancy, in which the data and human's understanding of the data are two fundamental elements. Human's understanding may be an explicit input–output model representing the relationship among the system's variables. It may also be represented as knowledge implicitly (e.g., the connection weights of a neural network). Therefore, FDD is done through some kind of modeling, signal processing, and intelligence computation. In this paper, a variety of FDD techniques are reviewed within the unified data-processing framework to give a full picture of FDD and achieve a new level of understanding. According to the types of data and how the data are processed, the FDD methods are classified into three categories: model-based online data-driven methods, signal-based methods, and knowledge-based history data-driven methods. An outlook to the possible evolution of FDD in industrial automation, including the hybrid FDD and the emerging networked FDD, are also presented to reveal the future development direction in this field.

It is our pleasure to present this Special Issue on "Data-Driven Approaches for Complex Industrial Systems" of the IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, which provides a forum for researchers and practitioners to report recent results on data-driven methods with applications to complex industrial systems, and to identify critical issues and challenges for future investigations in this field. Roughly, data-driven methods can be categorized into three sets, i.e., data-driven modeling, data-driven monitoring and fault diagnosis, and data-driven control and optimization (cf. Fig. 1). In this Special Issue, 13 papers are selected with novel contributions in data-driven modeling, data-driven monitoring and diagnosis, data-driven control and their industrial applications, respectively.

A review of black-box models for short-term time-discrete prediction of the silicon content of hot metal produced in blast furnaces is presented. The review is primarily focused on work presented in journal papers, but still includes some early conference papers (published before 1990) which have a clear contribution to the field. Linear and nonlinear models are treated separately, and within each group a rough subdivision according to the model type is made. Within each subsection the models are treated (almost) chronologically, presenting the principle behind the modeling approach, the signals used and the main findings in terms of accuracy and usefulness. Finally, in the final section the approaches are discussed and some potential lines of future research are proposed. In an Appendix , a list of commonly used input and output variables in the models is presented.

MOTIVATION: Hepatitis B virus and hepatitis C virus are the two leading causes resulting in hepatocellular carcinoma (HCC). It is observed that hepatitis C virus (HCV) is relatively difficult to induce HCC compared with hepatitis B virus (HBV). This motivates us to reveal the reasons behind this from the viewpoint of immune genes. RESULTS: To distinguish the immune genes with low-level expression in HBV-induced HCC, but high-level expression in HCV-induced HCC, the concept of distinction immune gene is proposed. A filter is then designed to screen these genes. By using gene positive network with strong correlations between genes, the genes are further filtered to form the set of key distinction immune genes. The 23 key distinction immune genes are screened, which are divided into four clusters, T cells, B cells, immune signalling and major histocompatibility complex. It is evident that the screened genes are important immune genes, which are activated in HCV-induced HCC, but inactivated in HBV-induced HCC. In HCV-induced HCC, the structures of HCV adaptively update, so that they are difficult to be identified by antigens. Therefore, the clinic advice is either to increase the update speed of antigens or reduce the update speed of the viruses during the treatment of HCV-induced HCC. Moreover, it is also advised to add T cells or add the expression levels of T cells to strengthen the ability to kill cancer cells. In contrast, HBV updates slowly, but the immunity system in HBV-induced HCC has been damaged seriously. As a result, the clinic advice is to improve the immune ability of patients subjected to HBV-induced HCC, such as increasing immunoglobulin, T cells and B cells and so forth.

This study is motivated by the onboard fault detection of gas turbine engines (GTEs), where the computation resources are limited and the disturbance is assumed to be band-limited. A fast Fourier transformation (FFT)-based disturbance frequency estimation approach is proposed and performance indexes are improved by integrating such frequency information. Furthermore, in the left eigenvector assignment, both eigenvalues and free parameters are optimized. As illustrated in the application to the actuator fault detection of a GTE, significant improvements are achieved compared to the existing methods. By combining the frequency estimation and eigenvalue optimization, the main contribution of the paper is the reduction of the computation complexity and the avoidance of the local optimal solution due to fixed eigenvalues.

In this study, a novel identification technique, that is high-gain observer-based identification approach, is proposed for systems with bounded process and measurement noises. For system parameters with abnormal changes, an adaptive change detection and parameter identification algorithm is next presented. The presented technique and algorithm are finally applied to the parameter identification of the gas turbine engine by using the recorded input data from the engine test-bed. The identified parameters and the response curves are desirable. The simulations have proved the effectiveness of the proposed procedure compared with the previous identification approach.

For systems with both state and input time delays, a novel state and sensor fault observer is proposed in this paper to estimate system states and sensor faults simultaneously. In this design, a descriptor system approach and a linear matrix inequality technique are adopted, where the considered sensor fault may be in any form, even unbounded. Unbounded sensor faults will make the system fail unavoidably; it is indispensable to derive a reliable control scheme against sensor failures. Using the estimated state and sensor fault, a reliable observer-based controller is proposed, which makes the system work well no matter whether sensor faults occur or not. The present approaches are next extended to the case for systems with multiple time delays. Finally, a simulation example of the network of three cascaded reactors is used to illustrate the design procedure and demonstrate the efficiency of the present techniques.

A novel fuzzy-observer-design approach is presented for Takagi-Sugeno fuzzy models with unknown output disturbances. In order to decouple the unknown output disturbance, an augmented fuzzy descriptor model is constructed by supposing the disturbance to be an auxiliary state vector. A fuzzy state-space observer is next designed for the augmented fuzzy descriptor system, and the simultaneous estimates of the original state and disturbance are thus obtained. The proposed observer technique is further applied to estimate sensor faults. Finally, a numerical example is given to illustrate the design procedure, and the simulation results show the desired tracking performance. The preknowledge of the disturbance and fault is not necessary for our design. Moreover, the considered disturbance and sensor fault can be in any form.