Balancing objectives and constraints is challenging in addressing constrained multiobjective optimization problems (CMOPs). Existing methods may have limitations in handling various CMOPs due to the complex geometries of the Pareto front (PF). And the complexity arises from the constraints that narrow the feasible region. Categorizing problems based on their geometric characteristics facilitates facing this challenge. For this purpose, this article proposes a novel constrained multiobjective optimization framework with detection and supervision phases, called COEA-DAS. The framework categorizes the problems into four types based on the overlap between the obtained approximate unconstrained PF (UPF) and constrained PF (CPF) to guide the coevolution of the two populations. In the detection phase, the detection population approaches the UPF ignoring the constraints. The main population is guided by the detection population to cross infeasible barriers and approximate the CPF. In the supervision phase, specialized evolutionary mechanisms are designed for each possible problem type. The detection population maintains evolution to assist the main population in spreading along the CPF. Meanwhile, the supervision strategy is conducted to reevaluate the problem types based on the evolutionary state of the populations. This idea of balancing constraints and objectives based on the type of problem provides a novel approach for more effectively addressing the CMOPs. Experimental results indicate that the proposed algorithm performs better or more competitively on 57 benchmark problems and 12 real-world CMOPs compared with eight state-of-the-art algorithms.

Intelligent fault diagnosis has attracted much attention in industrial processes. The difficulty of collecting fault samples and high price of labeling data, has led to a relative scarcity of labeled data for deep learning tasks in the field. To address this gap, we propose a domain knowledge-guided contrastive learning framework based on complementary data views for fault diagnosis with limited data. Seven data views of either time- or frequency-domains are introduced and designed first. Then, the framework extracts task-specific features by 1) considering complementary information provided by multiple data views to each other, and 2) embedding a domain knowledge-involved space as the guide for the learning process. The results on two bearing datasets show the proposed framework can produce diagnosis accuracies of 96.60% and 94.24% when just 5% of samples have labels. This study determines two pairs of complementary data views that can boost the performance of the proposed framework.

This article considers the time-varying formation (TVF) tracking issue of heterogeneous multiagent systems (HMASs) with the dynamic event-triggered control. The HMASs contain heterogeneous multiple leaders, all of which have the input signals to generate flexible reference, and only the output information can be measured. All leaders do not have access to the same followers, that is, the well-informed follower assumption is removed in this article. In this setting, the adaptive multileader state compensator is designed for each follower to estimate the integrated state information of all leaders, which can equip with two kinds of dynamic event-triggered mechanisms, that is, node-based event-triggered mechanism and edge-based event-triggered mechanism, to save communication bandwidth. Then, the TVF controllers are built by some estimation values to regulate the followers to achieve and maintain the geometric shape while tracking the reference which is the convex combination of outputs of leaders. The event-triggered compensator and TVF controller constitute the control protocol of HMASs, which are independent of global information with the fully distributed manner. The stability analysis and numerical simulations are given to verify the presented control protocol.

In this article, the time-varying output formation issue of heterogeneous multiagent systems is investigated by the event-triggered control scheme. Only the outputs of all agents, including leader agent and follower agents, are measurable. The leader agent contains an unknown input signal to generate flexible reference trajectory. Also, only a subset of follower agents have the direct access to the leader agent. First, for each follower, the leader-state compensator is designed to estimate the state of leader. Two kinds of dynamic event-triggered (DET) mechanisms, i.e., node- and edge-based event-triggered schemes, can be equipped on the compensator to save the communication bandwidth of leader-follower and follower-follower interactions, respectively. Then, the distributed formation controller is built to drive each follower achieving formation tracking. The presented control protocol consisting of the DET state compensator and formation controller is fully distributed, which is independent of the global information of communication topology, such as the eigenvalues of Laplacian matrix of communication topology and amount of whole agents. Finally, the numerical experiments and comparison experiments are exhibited to verify the effectiveness of the presented control protocol.

This article presents some new improvements to the relevant constrained predictive fault-tolerant tracking control (FTTC) methods through embedding optimal preview regulation and reference governance. The main novelty of such a strategy is that some valuable information of finite future references can be adequately scheduled to optimize the robust tracking performance and significantly enlarge the fault-tolerant admissible region. In order to better describe the wide applicability of the proposed strategy, the robust FTTC problem for a class of LPV systems with state/input constraints is considered. Overall, the involved key designs consist of three parts. First, an unconstrained FTTC component is constructed by combining tracking error feedback, reference input regulation, and fault signal compensation. It is used to guarantee the robust tracking stability of closed-loop systems when the constraints are not activated. Second, a tube-based predictive FTTC policy with an embedded optimal preview regulator is designed to achieve the robust constraint satisfaction and transient response improvement. Third, an embedded reference governor is additionally integrated to significantly enlarge the size of the fault-tolerant admissible region. This design further reinforces the feasibility of constrained FTTC optimization. The effectiveness of these results is finally validated by a case study of a single transistor dc/dc Forward converter.

The main task of noninvasive load monitoring is to disaggregate the power consumption of a single household appliance from an electricity meter that detects the power consumption of all household appliances. The deep neural network method has achieved leading results in this field. In this article, a multichannel spatio-temporal feature fusion method is proposed, where the spatial features extracted by convolution neural network and the temporal features extracted by the recurrent neural network are fused. And the attention module is introduced to further improve the performance of the model. Finally, the effectiveness and superiority of the proposed method are verified on three public datasets.

In the study, a MIMO model-free-adaptive-iterative-learning-control-based (MFAILC-based) consensus tracking scheme for multiagent systems (MASs) has been proposed. The dynamics of agents are heterogeneous and unknown. And the compact form dynamic linearization (CFDL) technique is utilized to describe the unknown nonlinear dynamics of agents along iteration axis. Then, the MIMO MFAILC-based consensus tracking algorithm is proposed for MASs under fixed topology. From the proof, we can obtain that the consensus tracking error can converge to zero along iteration axis asymptotically. Next, the MFAILC-based consensus tracking algorithm is extended to controlling the MASs under iteration-switching topologies and the MASs with external disturbances, respectively. Compared with prior work, the main features of this article are the MFAILC-based consensus strategy can be utilized for MIMO MASs, and the framework of robust MFAILC is built for MIMO MASs with external disturbances. Finally, three simulations are given to verify the effectiveness of the consensus strategy for MASs under fixed and switching topology and MASs with external disturbances, respectively.

This article presents a novel robust estimator-based dual-mode predictive fault-tolerant control (FTC) design scheme for linear parameter varying systems with state/input constraints. The overall FTC is composed of an unconstrained robust fixed FTC component and a constrained robust predictive FTC component. The former is determined by solving an offline integrated design of generalized unknown input observer and fault compensation control law. It is mainly used to compensate the influence of faults and stabilize the tracking error system. The robust predictive FTC is formulated based on the tightened invariant set constraint and quadratic programming. It is used to guarantee the recursive feasibility of the overall FTC under constraints, so its optimal value should be determined by solving the programming problem in real time. An algorithm is also provided to summarize all the involved steps of offline design and online implementation. Finally, the effectiveness of the proposed results is verified in a practical dc-dc buck converter.

This article focuses on addressing a fault-tolerant tracking control (FTTC) problem for a class of constrained polytopic LPV systems. The dual-mode predictive control method is introduced to design an efficient FTTC strategy such that the performance output of LPV systems can be regulated to track a given reference within state/input constraints. On the whole, the FTTC policy is constructed by the reduced-order estimator, tracking error feedback, double signal compensation, and optimal predictive regulation (PR). In specific, a novel reduced-order estimator is first designed to provide the state and fault information. Then, an unconstrained robust tracking error feedback control is designed to stabilize the LPV systems. Following that, a double signal compensation mechanism is proposed to remove the influences of fault, fault estimation error and other mismatched disturbances. Afterward, an optimal PR policy is further designed to optimize the tracking performance and handle system constraints. Finally, the effectiveness of these proposed results is verified by three case studies.

In most current nondestructive testing systems, a magnetic flux leakage (MFL) method is widely used in various industry fields, where the structural integrity of specimens is of vital importance. The estimation of defect size in specimen from the MFL measurements is a key and difficult problem. The traditional methods have low precision because feature extraction procedure relies on prior knowledge and the ability of designer. Inspired by the idea of convolutional neural network (CNN), a novel visual transformation CNN (VT-CNN) is proposed in this paper to overcome the limitation of traditional method in a feature extraction procedure. By adding a visual transformation layer according to the characteristics of the MFL measurements, the VT-CNN can distinguish the defect feature with different sizes more accurately. Moreover, since the VT-CNN method is designed based on the deep learning theory, more industrial big data with accurate label should be used to train the network. Due to the difficulty of making real industrial big data, a novel mesher magnetic dipole model is designed to simulate this industrial process. A large simulated MFL measurement of irregular defects produced by this model can increase the number of training samples and improve the robustness of the network. Experiments to estimate natural corrosion defects on real industrial pipelines are performed to validate the proposed framework. The experimental results are illustrated in detail, which highlights the superiority of the proposed method in industrial applications.

Though Fisher discriminant analysis (FDA) is an outstanding method of fault diagnosis, it is usually difficult to extract the discriminant information in a complex industrial environment. One of the reasons is that, in such an environment, the discriminant information can not been extracted entirely due to the disturbances, non-Gaussianity and nonlinearity. In this paper, a method named Joint Fisher discriminant analysis (JFDA) is proposed to address the issues. First, JFDA removes outliers caused by disturbances according to the energy density of each datum. Then, for the non-Gaussianity and weakly nonlinearity, the novel scatter matrices are defined to extract both of the local and global discriminant information based on the manifold learning. Finally, the kernel JFDA (KJFDA) is investigated to hold the manifold assumption because the strongly nonlinearity may weaken the assumption and cause overlapping. The proposed method is applied to the Tennessee Eastman process (TEP). The results demonstrate that KJFDA shows a better performance of fault diagnosis than other improved versions of FDA.

This paper studies the robust stability problem for a class of Markovian jumping stochastic neural networks (MJSNNs) with interval time-varying delay. Based on Lyapunov-Krasovskii functional and stochastic analysis approach, a new delay-dependent sufficient condition is obtained in the linear matrix inequalities (LMIs) format such that for all admissible uncertainties delayed MJSNNs is globally asymptotically stable in the mean-square sense. The effectiveness of the proposed method is demonstrated by a numerical example.

In this paper, the problems of quadratic stabilization conditions for stochastic Takagi-Sugeno fuzzy systems have been studied. A new quadratic stability condition, which takes into account the knowledge of the membership function's shape by considering bounds on their cross products, has been proposed. And then a new sufficient condition in terms of linear matrix inequalities is developed to synthesize the state feedback controller that stabilizes the stochastic fuzzy systems. An numerical example is provided to illustrate the effectiveness of the proposed results.

Monitoring oil transporting pipelines is an important task for economical and safe operation, loss prevention, and environmental protection from crude oil emission. The leak detection of oil pipelines, therefore, plays a key role in the overall integrity, monitoring a pipeline system. This paper proposes a fuzzy decision-making approach to oil pipeline leak localization. The two main methods, pressure gradient localization and negative pressure wave localization, are combined with fuzzy logical decision-making to form a novel fault diagnosis scheme. The combination scheme can improve the precision of localization. An application example, a 14 km long oil pipeline leak detection and localization is illustrated, and the effectiveness of the proposed approach is demonstrated using the practical results.