Search

Long-term failures in flexible converter valve equipment pose significant risks, potentially compromising operational efficiency or leading to complete malfunction. Accurately identifying equipment faults is essential to improve overall reliability and minimize downtime. This study introduces an innovative fault diagnosis method utilizing an attention mechanism. The method integrates a lightweight model incorporating one-dimension depthwise convolutional layers for spatial feature extraction and bidirectional long short-term memory for capturing temporal dynamics. A pioneering time-channel joint attention module enhances the extraction of fault-related data from time series and channel maps. Experimental results underscore the method’s efficacy in fault diagnosis under varying Gaussian noise conditions. Notably, the approach demonstrates remarkable consistency in accuracy across various experimental setups, underscoring its robust performance and potential applicability in real-world scenarios where reliability is critical. In addition, the proposed method has a moderate number of parameters and training time, indicating that the model can be embedded in front-end equipment.

Reinforcement learning (RL) can obtain the supervisory controller for discrete-event systems modeled by finite automata and temporal logic. The published methods often have two limitations. First, a large number of training data are required to learn the RL controller. Second, the RL algorithms do not consider uncontrollable events, which are essential for supervisory control theory (SCT). To address the limitations, we first apply SCT to find the supervisors for the specifications modeled by automata. These supervisors remove illegal training data violating these specifications and hence reduce the exploration space of the RL algorithm. For the remaining specifications modeled by temporal logic, the RL algorithm is applied to search for the optimal control decision within the confined exploration space. Uncontrollable events are considered by the RL algorithm as uncertainties in the plant model. The proposed method can obtain a nonblocking supervisor for all specifications with less learning time than the published methods.

Model abstraction for finite state automata is beneficial to reduce the complexity of discrete-event systems (DES), enhance the readability and facilitate the control synthesis and verification of DES. Supremal quasi-congruence computation is an effective way for reducing the state space of DES. Effective algorithms on the supremal quasi-congruence relation have been developed based on the graph theory. This paper proposes a new approach to translate the supremal quasi-congruence computation into a satisfiability (SAT) problem that determines whether there exists an assignment for Boolean variables in the state-to-coset allocation matrix. If the result is satisfied, then the supremal quasi-congruence relation exists and the minimum equivalence classes is obtained. Otherwise, it indicates that there is no such supremal quasi-congruence relation, and a new set of observable events needs to be modified or reselected for the original system model. The satisfiability problem on the computation of supremal quasi-congruence relation is solved by different methods, which are respectively implemented by mixed integer linear programming (MILP) in MATLAB, binary linear programming (BLP) in CPLEX, and a SAT-based solver (Z3Py). Compared with the MILP and BLP methods, the SAT method is more efficient and stable. The computation time of model abstraction for large-scale systems by Z3Py solver is significantly reduced.

To cope with the limited radio and power resources, designing energy- and cost-efficient resource allocation strategy with RAN slicing and scheduling is important in ensuring the extreme QoS of differentiated Internet of things (IoT) services. In this regard, we focus on guaranteeing the latency and reliability of sporadic uRLLC uplink traffic while improving the quality of continuous eMBB services (e.g., quality of the video) together in this paper. Firstly, a dynamic optimization model considering power consumption and service quality is used to construct the cost function in both time domain and frequency bandwidth for heterogeneous services, subject to the latency constraint. Secondly, given its complexity, a novel two-timescale algorithm with employing Lyapunov optimization is designed, including two sub-algorithms: long-timescale bandwidth allocation and short-timescale service control. In further, the theoretical optimality is analyzed according to the relationships between control parameters and service performances. The utility of our approach and its hard latency guarantee are also illustrated through simulation results under tolerable power consumption.

By leveraging the 5G enabled V2X networks, the vehicles connected by cellular base-stations can support a wide variety of computation-intensive services. In order to solve the arisen challenges in end-to-end low-latency transmission and backhaul resources, mobile edge computing (MEC) is now regarded as a promising paradigm for 5G-V2X communications. Considering the importance of both reliability and delay in vehicle communication, this article innovatively envisions a joint computation and URLLC resource allocation strategy for collaborative MEC assisted cellular-V2X networks and formulate a jointly power consumption optimization problem while guaranteeing the network stability. To solve this NP hard problem, we decouple it into two sub-problems: URLLC resource allocation for multi-cells to multi-vehicles and computation resource decisions among local vehicle, serving MEC server and collaborative MEC server. Secondly, non-cooperative game and bipartite graph are introduced to reduce the inter-cell interference and decide the channel allocation, which aims to maximize the throughput with a guarantee of reliability in URLLC V2X communication. Then, an online Lyapunov optimization method is proposed to solve computation resource allocation to get a trade-off between the average weighted power consumption and delay where CPU frequency are calculated using Gauss-Seidel method. Finally, the simulation results demonstrate that our proposed strategy can get better trade-off performance among power consumption, overflow probability and execution delay than the one based on centralized MEC assisted V2X.

Winter jujubes get bruised easily during harvest and transportation. In order to detect subtle bruises on winter jujubes in a more efficient way, a rapid and accurate technique, hyperspectral imaging was used. Near-infrared reflectance (NIR) and visible/near-infrared reflectance (Vis-NIR) hyperspectral imaging at the spectral region of 874-1734 nm and 380-1030 nm, respectively, were applied in this study. The hyperspectral images of winter jujubes from four geographical origins were acquired. Pixel-wise spectra were extracted and preprocessed; pixel-wise principal component analysis (PCA) was used to conduct a qualitative analysis. Accuracy, true positive rate (TPR) and false positive rate (FPR) were utilized to compare the efficiency of the models. Support vector machine (SVM), logistic regression (LR) and a deep learning method, and convolutional neural network (CNN) were used to build pixel-wise classification models based on single or all geographical origins for quantitative analyses. All the models using NIR spectra obtained decent results with accuracies in the range of 90-100%, and TPRs and FPRs close to 1 and 0, respectively. Compared with the other two methods using Vis-NIR spectra, the CNN model based on all geographical origins got the best performance with most of the accuracies surpassing 85%. For Vis-NIR spectra and NIR spectra, the overall time efficiency for modeling and prediction of CNN was at an intermediate level among the three models. The short prediction time of CNN indicated that CNN had the potential for real-time detection. The prediction maps obtained by the CNN models indicated that the color information and geographical origins could affect the detection performance. The overall results demonstrated the promising potential for detecting subtle bruises on winter jujubes using pixel-wise spectra extracted from the hyperspectral images at the two spectral ranges with the deep learning method. The results in this study would help to develop an online winter jujube bruises detection system in the future.

Cell outage compensation is a crucial way for public safety network (PSN) to recover network communication independently during disasters. As a significant technology for 5G, a cloud-radio access network (C-RAN) plays an important role in a PSN. C-RAN can better achieve multi-cell cooperative outage compensation under dense network coverage because of the characteristics of densification and centralization. Therefore, we propose an efficient multi-cell cooperative outage compensation convergence for the scene where more than one radio remote units (RRUs) are destructed in C-RAN-based PSN. This scheme compensates the network using both cooperative transmission and power adjustment. First, RRUs are selected to participate in the compensation according to the topology structure. Then, the problem model is built aiming at optimizing system outage probability. At the same time, the constraints of resources and service quality are considered. In addition, an enhanced immune-genetic algorithm is proposed innovatively to solve this NP-hard problem and to get the results of RRU transmit power parameters. Finally, simulation results demonstrate that the proposed algorithm convergences rapidly and our scheme reduces the PSN system outage probability effectively with the guarantee of users received power and the restriction of interference effects. The outage probability of the PSN system using the proposed method is 17.8% of the only cooperative transmit method and 3.9% lower than the only adjust power method.

With the ever-increasing demands for popular content sharing among humans, device-to-device (D2D) multicast communication, as a promising technology to support wireless services within a local area, is introduced in a 5G cellular network. However, the existing resource allocation approaches for D2D multicast communication usually consider only physical domain constraints but neglect social domain factors, which would result in ineffective D2D links between users unwilling to share interests. Different from existing works, the D2D multicast scheme proposed in this paper will produce effective D2D multicast links by sufficiently utilizing both the physical and social properties of mobile users, with the goal to maximize the throughput of the overall social-aware network and guarantee fairly allocation of the channel between different D2D multicast clusters. The scheme mainly consists of two parts, the formation of D2D multicast clusters and joint optimization of power and channel allocation. In the formation of D2D multicast clusters, members and head in each cluster are selected by taking into account both social attributes and physical factors, such as community, ties, and geographical closeness. In the joint optimization, a two-step scheme is designed to first calculate the optimal power allocation by geometric proximity and then select suitable cellular channels for each D2D multicast cluster utilizing an extended one-to-many bipartite graphs matching algorithm. Simulation results demonstrate that, compared with heuristic algorithm and stochastic algorithm, the proposed scheme can increase the throughput of the overall social-aware network by about 5% and 50%, respectively.