The normal data are sampled from the industrial process far more than fault data. Due to the existence of unbalanced data, the traditional learning method are difficult to classify the fault correctly. This paper designs a generation network to effectively augment the fault data and obtain the balanced industrial dataset. Conditional Generative Adversarial Network (CGAN) is constructed to generate a large amount of new fault data which has similar distribution with the original real fault data. The time series data are first transform into matrix format in order to adapt the input of CGAN. Then convolutional neural network (CNN) is used for fault diagnosis. The proposed method is applied in an actual gas-solid fluidized bed equipment to verify its effectiveness.

The idea of stacking layers is adopted to construct a deep multivariate statistical model, hierarchical canonical correlation analysis (HCCA). Its hierarchical structure is motivated form the deep network. The proposed HCCA model has the features of low computational complexity, strong correlative feature extraction, and causal interpretability. Its correlation advantages are theoretically demonstrated, then the evaluation metrics about accuracy and complexity are presented. The HCCA-based fault monitoring method is proposed for industrial processes, and the variable contributions are analyzed based on the residual statistic. The experiment results on Tennessee Eastman and real industry wastewater treatment processes show an average fault detection rate of 87.91$\%$and 99.49$\%$. It also decreases an average false alarm rate to 0.92$\%$and 0.32$\%$, respectively.

This article provides an efficient edge-end implementation solution for deep learning-based surface defect detection to improve the accuracy and efficiency when applied on edge devices with limited resource. An efficient you only look once (YOLO) network YOLOv5s-GhostNet is proposed, which highlights at the lightweight backbone/neck network, efficient feature extraction modules, and a fast learning scheme based on knowledge distillation. The parameter compression ratio is theoretically analyzed to show the decrease of computation complexity. The jointed loss is designed to enhance the generalization ability for new defects. An industrial testing platform with real-time edge-terminal-cloud detection system is developed with Raspberry Pi as edge. The experimental results show that the proposed method gets performances at complexity (floating-point operations per second (FLOPS) 8.2G, pt 7.9M), detection accuracy (precision 97.91$\%$, mean average precision (mAP) 96.66$\%$), efficiency [frames per second (FPS) 294 for single defect], and fast learning convergence (50 epochs). Compared to the existing methods, it reduces model size by 50$\%$on overage, increases the detection efficiency by 4 times and maintains the higher accuracy.

This article presents a novel physics-constrained zonotopic Kalman filter method for online capacity prediction of lithium-ion batteries. To describe capacity degradation, a state-space formulation is devised using the autoregressive model and an indirect representation of capacity. The approach consists of three steps: First, a zonotopic Kalman filter is proposed to estimate model parameters and parameter intervals. Subsequently, considering the capacity regeneration phenomenon, a physics-based constraint term is presented to optimize parameters, which updates the estimated model parameters obtained by the zonotopic Kalman filter. Finally, parameters and interval estimation are utilized to predict the future short-term capacity. The case study demonstrates the validity of our approach. Moreover, comparisons with the ellipsoid-based extended Kalman filter and predictive maintenance toolbox suggest that our approach can obtain more precise capacity prediction and tighter capacity interval results.

In the field of autonomous driving and specific environments, target detection is a critical task module. Currently, the mainstream approach to target detection is to use deep learning to train specific network models, enabling them to recognize targets. Although many effective network models have been proposed by scholars, there are few target detection networks tailored for low-light environments. In practical applications, complex lighting changes can lead to decreased accuracy in target detection. This paper combines a low-light enhancement network with a LiDAR-camera fusion target detection network to achieve target detection in low-light environments and validates the algorithm using the Nuscenes benchmark dataset. The experimental results demonstrate that the improved network exhibits greater robustness in target detection under complex lighting conditions.

This paper proposes an improved semantic segmentation model based on Fully Convolutional Network(FCN). Firstly, this paper integrates the channel attention module into the backbone of FCN to identify the channels that are more important for semantic segmentation. Secondly, this paper adds a multi-scale feature fusion module to the network to integrate multi-scale feature information. Thirdly, this paper replaces some of standard convolutions in the backbone network with dilated convolutions to increase the receptive field. Experimental results demonstrate that after adding the two modules, the pixel accuracy can reach 75.8%, and the mIOU value has increased by 2.8% compared to the original FCN model. Moreover, utilizing pre-trained weights during the training process can greatly enhance the performance of the model.

This paper proposes a flexible job-shop scheduling problem optimization method, which focuses on providing solutions for industrial production. First, in terms of model construction, the method further considers the cost of automated guided vehicles and the priority of workpiece production based on previous methods. Then, this method solves the model by non-dominated sorting genetic algorithm with self-cross and delete-mutation. It reduces the production time and energy by an average of 6.4% and 19.4%, which are 5.4% and 15.1% with the priority. Finally, the simulation verifies that the method effectively reduces the production cost while realizing the adjustment of the automated guided vehicle number and the workpiece production sequence.

Many machine-learning-based defect detection methods, especially deep learning-based approaches, have high requirements on computing power and network. They lead to time delay, high cost, and energy consumption when applied to deal with the massive data in an autonomous manufacturing enterprise. So efficient detection in the end-edge-cloud architecture is a good solution to overcome the above challenges. A branchy deep learning detection model with early exit ability of inference is proposed, in which the main branch is deployed on the cloud server and the side branches are on edge equipment. The proposed method quickly and effectively detects the category and location of the defect in printed circuit boards since partial computing task is offloaded to the edge nodes. A prototype system is implemented based on a computer as the cloud server and a Raspberry Pi as an edge node in order to verify the feasibility of the proposed method. The experiment result manifests high detection accuracy and fast computing speed.

In this article, a model predictive control (MPC)-based cooperative target enclosing control approach is investigated for multiple nonholonomic mobile agents with input constraints and unknown disturbances. The agents are required to move along a desired circular orbit centered at a stationary target and maintain an even distribution on the orbit. Based on a dual-mode MPC strategy, a cooperative target enclosing control law is designed by only using the local sensing information. When the agents are inside a terminal region, a locally cooperative stabilizing control law is designed with a signal function defined componentwise part compensating for the unknown disturbances. A robust MPC algorithm is designed for the agents to enter the terminal region in finite time. Global asymptotic stability is guaranteed for multiple nonholonomic mobile agents with input constraints and unknown disturbances. Simulation results illustrate the effectiveness of the proposed approach.

A hybrid filtering strategy, including the gain-scheduled filter (GSF) and probability density function (PDF) compensation, is proposed for a discrete system with missing measurements and Gaussian mixture noise. Commonly, the estimate error is strictly controlled to 0 is impossible in a complex environment. The aim of the proposed filter is to stabilize the filtering dynamics in the mean square sense and to ensure the estimate error PDF close to the desired PDF as much as possible. The GSF is designed via appropriate linear matrix inequalities (LMIs), and the PDF compensator is proposed based on the real-time Wasserstein distance between the past data and the desired distribution to force the estimate error distribution to track the desired distribution. Two comparison experiments are performed to verify the effectiveness of the proposed method.

This note studies an enclosing control problem for a multiagent system with a moving target of unknown bounded velocity. The objectives are to let each agent move along a circular orbit with a prescribed radius centered at the target and maintain desired spacing from neighboring agents. A distributed controller composed of three parts is designed by only using the relative position information from each agent to the target and its neighbors. The first two parts are designed to achieve target circling and spacing adjustment, respectively. The last part is designed discontinuously to compensate for the unknown bounded velocity of the target. Due to the discontinuously distributed controller, sufficient conditions are given by a nonsmooth analysis. Furthermore, the agents are shown to have order preservation and collision avoidance properties when the target is stationary. The effectiveness of theoretical results is illustrated by simulations.

This article investigates a class of multiobjective optimization fault detection observer design problems for linear parameter varying (LPV) systems considering the unknown but bounded disturbance with an adaptive event-triggered scheme. In this study, the actuator faults are considered in the low-frequency domain. First, to save the communication bandwidth and improve communication efficiency, an adaptively adjusted event-triggered (AAET) mechanism is proposed. Then, in order to make the designed observer gain satisfy both fault sensitivity and disturbance robust conditions, an$H_{-}/L_{\infty }$multiobjective optimization problem is proposed and solved by appropriate linear matrix inequalities. Next, the upper and lower bounds of the generated residual are calculated by the zonotope method when considering the estimation uncertainty. Fault detection can be achieved by judging whether the zero value belongs to the generated range of the residual signal. Finally, a simulation case is used to verify the effectiveness of the proposed method.

The distributed monitoring framework is undoubtedly more suitable for large-scale complex industrial systems. However, most existing distributed monitoring methods ignored the information interaction between the local system and its neighbors. In this article, an improved distributed fault detection framework that considering the communication between subsystems is present. The system decomposition is optimized based on the monitoring performance with mechanism knowledge as constraints. The integration of mechanism and data is helpful to find the appropriate common variables between subsystems. The distributed partial least squares (DPLSs) algorithm is proposed to address the local monitoring challenges caused by the propagation of a common variable. The local monitoring model takes full advantage of the information from neighbors to reduce the uncertainty of the local system. Bayesian fusion performance metrics strategy is implemented to detect system status. The simulation results of the Tennessee Eastman process verify the effectiveness of the proposed scheme.Note to Practitioners—This article attempted to tackle an issue derived from distributed process monitoring of industrial processes. Even in an era of big industrial data, the fusion idea of process data and mechanism knowledge also provides a solution to the process decomposition monitoring strategy. It reduces the computational complexity, corrects the misdirection caused by the false information hidden in the measurements, and further increases the monitoring accuracy. Considering the information flowing and spreading along with the process equipment, common variables are used to describe the interaction between different subsystems. Then, the pretrained monitoring model and the online monitoring strategy are given to promote automatic implementation. The operability and monitoring accuracy of the proposed method is verified. It is suitable for process monitoring of large-scale complex industrial systems.

This paper investigates a finite-frequency H-/H∞fault detection method for discrete-time T-S fuzzy systems with unmeasurable premise variables. To minimize the effect of uncertainties on system performance and maximize that of actuator faults on the generated residual, both the H∞disturbance attenuation index and finite-frequency H-fault sensitivity index are utilized. Since the premised variables are unmeasurable, the existing generalized Kalman-Yakubovich-Popov lemma cannot be directly extended to these nonlinear systems. In this paper, the conditions of allowing one to design the proposed H-/H∞fault detection observer are established and transformed into linear matrix inequalities. Some scalars and slack matrices are introduced to bring extra degrees of freedom in observer design. Finally, a single-link robotic manipulator model is utilized to illustrate that the proposed technique can detect faults with smaller amplitude than that required by a normal H∞observer technique.

This paper presents a novel interval fault estimation approach by using zonotope technique for discrete-time linear parameter-varying systems in the presence of bounded parametric uncertainties, measured perturbation, and system disturbance. First, an augmented descriptor system is generated by using augmentation technique. Thus, the problem of interval fault estimation is transformed into the interval augmented state estimation. Then, an outer approximation of the new augmented state estimation domain is computed by using zonotope method. A zonotope should be consistent with the given outputs, perturbation, disturbance, and parametric uncertainties. Besides, it is minimized at each sampled time via an analytic formulation. Finally, a vehicle lateral dynamic nonlinear model is utilized to show the feasibility and effectiveness of the proposed zonotopic fault estimation technique.

Microbial fuel cells (MFC) is a new technique for the environmental protection and new energy. Microbial desalination cells (MDC) is a kind of MFC which has the function of desalination while producing electricity. The design of control strategies is a scarcity part in the field of microbial fuel cells. This paper presents to design an explicit model predictive controller for a kind of microbial fuel cell, which uses the machine learning method for easy to implement. Also, a systematic data-driven control method is presented for the design of explicit model predictive controller for time-varying output tracking in nonlinear model systems. The design consists of (1) sampling the admissible state space by the mathematical model to make the controller better suited to the model; (2) solving for optimal model predictive control actions at each sampled data point and determining feasible region of the nonlinear programming problem; and (3) constructing the control surface of explicit model predictive controller using artificial neural network. In particular, the designed control algorithm is performed on a seven-dimensional mathematical model of microbial desalination cells to test the good control performance.

Semisupervised scheme has emerged as a popular strategy in the machine learning community due to the expensiveness of getting enough labeled data. In this paper, a semisupervised incremental support vector machine (SE-INC-SVM) algorithm based on neighborhood kernel estimation is proposed. First, kernel regression is constructed to estimate the unlabeled data from the labeled neighbors and its estimation accuracy is discussed from the analogy with tradition RBF neural network. The incremental scheme is derived to improve the learning efficiency and reduce the computing time. Simulations for manual data set and industrial benchmark-penicillin fermentation process demonstrate the effectiveness of the proposed SE-INC-SVM method.

Incipient faults are difficult to be detected due to the intrinsic fault tolerance of traditional controller, but it should be eliminated as soon as possible before it deteriorates with time into something more serious. As a consequence of an intrinsic inability to assess whether a fault occurs based on output residual, the existing detection methods are failure for incipient fault. So the aim of active fault detection (AFD) is to make the system be unstable when incipient fault has occurred, which drives rapid fault detection. The fault-tolerant control (FTC) is designed to maintain the system stable and eliminate the fault impact without shutting the process down even if faults occur. In this paper, the Youla-Jabr-Bongiorno-Kucera (YJBK) parameter is employed to build the AFD and the FTC based on the relationship analysis between the fault and the dual YJBK parameter. A new structure of the tolerant controller parameter for FTC is designed, named as partial active FTC (PAFTC). PAFTC is dependent upon the fault detection information but not the fault size considering the parameter fault with unknown size and known form. A unified operation architecture for AFD and PAFTC with different YJBK parameters for incipient faults is proposed. Some illustrative examples are given to indicate the effectiveness of the proposed unified operation architecture.

Although second‐order cone programming (SOCP) has been applied to optimise camera parameters in computer vision, it is occasionally been used to refine fisheye camera external parameters as well. This study presents a fisheye camera external parameter estimation based on SOCP in convex optimisation. The homography constraint between two spherical images are first exploited to derive an equation with respect to a given error threshold. Then, the fisheye camera external estimation is transformed into an SOCP optimisation problem through reformulating the parameter estimation equation. The SOCP method has been implemented in Matlab and the optimisation toolbox has been made publicly available. The fisheye camera external parameter optimisation method has been validated by some experiments with synthetic and real data. Comparison experiments between the proposed method and other methods in the literature are also carried out, and the results show that the SOCP method is better for the corrected images.