In modern discrete flexible manufacturing systems, dynamic disturbances frequently occur in real time and each job may contain several special operations in partial-no-wait constraint due to technological requirements. In this regard, a hierarchical multiagent deep reinforcement learning (DRL)-based real-time scheduling method named hierarchical multi-agent proximal policy optimization (HMAPPO) is developed to address the dynamic partial-no-wait multiobjective flexible job shop scheduling problem (DMOFJSP-PNW) with new job insertions and machine breakdowns. The proposed HMAPPO contains three proximal policy optimization (PPO)-based agents operating in different spatiotemporal scales, namely, objective agent, job agent, and machine agent. The objective agent acts as a higher controller periodically determining the temporary objectives to be optimized. The job agent and machine agent are lower actuators, respectively, choosing a job selection rule and machine assignment rule to achieve the temporary objective at each rescheduling point. Five job selection rules and six machine assignment rules are designed to select an uncompleted job and assign the next operation of which together with its successors in no-wait constraint on the corresponding processing machines. A hierarchical PPO-based training algorithm is developed. Extensive numerical experiments have confirmed the effectiveness and superiority of the proposed HMAPPO compared with other well-known dynamic scheduling methods. Note to Practitioners—The motivation of this article stems from the need to develop real-time scheduling methods for modern discrete flexible manufacturing factories, such as aerospace product manufacturing and steel manufacturing, where dynamic events frequently occur, and each job may contain several operations subjected to the no-wait constraint. Traditional dynamic scheduling methods, such as metaheuristics or dispatching rules, either suffer from poor time efficiency or fail to ensure good solution quality for multiple objectives in the long-term run. Meanwhile, few of the previous studies have considered the partial-no-wait constraint among several operations from the same job, which widely exists in many industries. In this article, we propose a hierarchical multiagent deep reinforcement learning (DRL)-based real-time scheduling method named HMAPPO to address the dynamic partial-no-wait multiobjective flexible job shop scheduling problem (DMOFJSP-PNW) with new job insertions and machine breakdowns. The proposed HMAPPO uses three DRL-based agents to adaptively select the temporary objectives and choose the most feasible dispatching rules to achieve them at different rescheduling points, through which the rescheduling can be made in real time and a good compromise among different objectives can be obtained in the long-term schedule. Extensive experimental results have demonstrated the effectiveness and superiority of the proposed HMAPPO. For industrial applications, this method can be extended to many other production scheduling problems, such as hybrid flow shops and open shop with different uncertainties and objectives.

This paper proposes an enhanced YOLO v7-based method for detecting insulator defects in transmission lines, addressing the challenges of low accuracy and high leakage rates caused by complex backgrounds and electric poles alongside varying sizes of insulator targets in the image. Firstly, to address the issue of background interference and improve the importance of insulator features, a lightweight attention mechanism named Efficient Channel Attention (ECA) was introduced. With the incorporation of ECA, this model could effectively suppress background noise and provide more focus to insulator regions, thus enhancing its ability to detect insulator defects accurately. Secondly, a partial convolution (PConv) approach was employed in the backbone network instead of conventional convolution, which learned some important channels. This substitution improved both the network model’s accuracy and the training speed. Finally, the Normalized Wasserstein Distance (NWD) prevented insulator features from being lost during pre-feature extraction, which reduced the leakage rate and improved the detection accuracy of small target insulators and defective insulators. The experimental results demonstrate that the improved YOLO v7 network model achieved an average detection accuracy (mAP) of 98.1%, recall of 93.7%, and precision of 96.8% on the TISLTR dataset. On the FISLTR dataset, the average detection accuracy (mAP) for flashover insulators was 93%, with a recall of 92.3% and precision of 87.1%. The average detection accuracy (mAP) for broken insulators was 92.2%, with a recall of 90.3% and a precision of 95.2%. These metrics demonstrate significant improvements in both datasets, highlighting the proposed algorithms’ strong generalization capability and practicable potential to detect insulator targets.

In prognostics and health management, multiple sensors have been widely used to monitor the health condition of complex machines. As each sensor provides partial and dependent information, data-level fusion techniques have been developed and aim at increasing the reliability and safety of machines by providing the current health assessment and remaining useful life (RUL) prediction. While most existing data-level fusion techniques have shown a promise for prognostics, they are mainly limited by only focusing on the improvement of point prediction, which makes it difficult to provide adequate available information for decision-making in predictive maintenance. In this work, prediction interval (PI) is adopted to model the prediction uncertainty of RUL for its nature of high variability. An improved echo state Gaussian process (IESGP), which is able to provide an estimation of prediction uncertainty, is developed as a novel data-driven approach for RUL prediction and PI construction. The proposed IESGP is a novel Bayesian approach which combines the merits of echo state networks and Gaussian processes to enhance the prediction accuracy. Based on this, a comprehensive cost function is constructed to characterize the accuracy and uncertainty simultaneously, then a multi-objective genetic algorithm (MOGA) is applied to optimize the point prediction and prediction interval. The validity of the proposed approach is verified on the widely used turbofan benchmark datasets. Experimental results show that the proposed approach can not only achieve superior predictive accuracy in comparison with several state-of-the-art approaches, but also obtain PIs with high quality.

Fault prognosis under time-varying operating conditions is critical to predictive maintenance of complex industrial machines. However, prognostic methods proposed in the literature toward varying operating conditions typically focus on limited discrete states with clear boundaries, which are not always encountered in practical scenarios. This article aims to propose a more general method that can handle complex operating conditions with more nonlinearity and dynamics. In this work, a novel deep model named dual attention-based temporal convolutional network (DATCN), is developed to integrate multisensor data with time-varying operating settings for health assessment of industrial machines. An efficient deep temporal convolutional network is introduced to extract degradation features for sequence modeling. A novel dual attention model is developed to adaptively capture the temporal dynamics of sensor data and operating data, further enhancing the feature representations and highlighting the degradation information under time-varying operating settings. Besides, domain adaptation layer with multikernel maximum mean discrepancies (MK-MMD) is embedded to mitigate the domain-shift issue caused by varying operating conditions. The experimental results on the popular turbofan engine data set and a real-world Ion mill etching process validate the effectiveness of the proposed method and show its superiority in yielding significant improvements of prognostic accuracy relative to existing state-of-the-art methods.

Fault prognosis under multiple fault modes is critical to predictive maintenance of complex tools in semiconductor manufacturing. However, the inherent data discrepancy among different tools and data imbalance with limited fault data coexist in real industrial scenario, making the task quite challenging. Therefore, this article proposes a novel two-stage deep transfer learning-based framework for prognosis under multiple fault modes, which aims at accurately predicting the time-to-failure of an Ion mill etching process. In the first stage, a base fault mode is selected and data alignment on condition monitoring data from multiple tools is performed via domain adversarial learning, wherein the temporal convolutional network is embedded to learn temporal representations from time-series sensor data. The second stage handles the prognostic tasks with remaining fault modes, the well-trained deep model from the first stage is employed as a pre-trained model, which will be fine-tuned with a relatively small amount of data from other fault modes, further accelerating the training process and enhancing the prediction performance. Comprehensive experiments are carried out on a real-world IME dataset, and the results show that the proposed model not only achieves better prediction accuracy but also saves much time for training compared with other existing methods.