Search

Identifying safety risks in distribution networks is of great significance for ensuring the safety of personnel and the stable operation of the distribution system. Existing research on safety risk identification in distribution network operations mainly focuses on personnel irregular dress detection and dynamic unsafe behavior identification. However, the actual operation scenario of the distribution network involves a complex process of multi-element interaction and integration of personnel, tools, and equipment machinery, where the risk of violations is often hidden within the intricate web of interactions. For this reason, this paper focuses on the problem of violation identification of human-object interaction relations in distribution network operation scenarios and proposes a violation risk identification method based on multiple interaction relations. The method firstly extracts the features of the distribution network operation image by convolutional neural network resnet101, then introduces the coding-decoding structure to re-encode the feature vectors to get the feature vectors with different interactions, and at the same time, utilizes the conditional filtering module to improve the convergence speed of the structure, and utilizes the Residual Information Exchange Module and the multi-layer mlp structure to discriminate the interaction pairs of multiple relationships, and finally takes the ladder climbing operation scenario as an example for the experimental validation. The experimental results showed that the proposed method can realize the accurate identification of human-object interaction relationships and violation risk, and has strong practical application value.

Introduction: Qi-Xian Decoction (QXD), a traditional Chinese medicine (TCM) formula consisting of eight herbs, has been clinically used to treat asthma. However, the underlying mechanisms have not been completely elucidated. This study aimed to combine metabolomics and network pharmacology to reveal the mechanism of action of QXD in asthma treatment.Methods: An ovalbumin (OVA)-induced asthma mouse model was constructed to evaluate the therapeutic effects of QXD. Serum metabolomics and network pharmacology were combined to study the mechanism of anti-asthma action as well as the potential target, and related biological functions were validated.Results: The QXD treatment has demonstrated significant protective effects in OVA-induced asthmatic mice, as evidenced by its ability to inhibit inflammation, IgE, mucus overproduction, and airway hyperreactivity (AHR). Metabolomic analysis has revealed a total of 140 differential metabolites associated with QXD treatment. In addition, network pharmacology has identified 126 genes that are linked to the effects of QXD, including TNF, IL-6, IL1β, STAT3, MMP9, EGFR, JUN, CCL2, TLR4, MAPK3 and MAPK8. Through comprehensive gene-metabolite interaction network analysis, seven key metabolites have been identified and associated with the potential anti-asthmatic effect of QXD, with palmitic acid (PA) being the most notable among them. In vitro validation studies have confirmed the gene-metabolite interaction involving PA, IL-6, and MAPK8. Furthermore, our research has demonstrated that QXD treatment can effectively inhibit PA-promoted IL-6 expression in MH-S cells and reduce PA concentration in OVA-induced asthmatic mice.Conclusion: The regulation of metabolic pathways by QXD was found to be associated with its anti-asthmatic action, which provides insight into the mechanism of QXD in treating asthma.

Oil-immersed transformers play a pivotal role owing to their environmentally friendly characteristics, compact footprint, and cost-effectiveness. Ensuring the online monitoring of oil-immersed transformers is a fundamental measure to ensure the secure and stable operation of modern power systems. In this paper, metal particle cluster-doped SnS is firstly used in the adsorption and sensing of decomposition components (CO, C2H2) under fault conditions in oil-immersed transformers. The study comprehensively analyzed band structure, differential charge density, density of states, and molecular orbital theory to unveil the adsorption and sensing mechanisms of target gases. The findings suggest that the modification of metal particle clusters can enhance the surface electronic properties of single-layer SnS. In the regions of metal particle clusters and the gas–surface reaction area, electronic activity is significantly heightened, primarily attributed to the contribution of d-orbital electrons of the metal cluster structures. The modified SnS exhibits adsorption capacity in the following order: Ru3-SnS > Mo3-SnS > Au3-SnS. Additionally, the modified material demonstrates increased competitiveness for C2H2, with adsorption types falling under physical chemistry adsorption. Different metal elements exert diverse effects on the electronic distribution of the entire system, providing a theoretical foundation for the preparation of corresponding sensors. The findings in this work offer numerical insights for the further preparation and development of SnS nanosensors, concurrently shedding light on the online monitoring of faults in oil-immersed transformers.

As an important index of risk protection, the safety distance is crucial to ensure the safe and stable operation of the power system and the safety of personnel’s life. Traditional monitoring methods are difficult to balance recognition accuracy and convenience. Therefore, this paper presents a power safety distance sensing method based on monocular visual images to achieve the recognition of the safety distance of external damage in complex scenes of transmission corridors, and proposed a power density depth distance model. In this model, a codec network with skip-connection to extract features and aggregate shallow and deep features for input power system images. Then, the regularization method, migration learning strategy, cosine annealing learning strategy, and data enhancement strategy are used to further optimize the model, so as to obtain a model with good accuracy and generalization in complex conditions. The effectiveness and superiority of the proposed method are verified in comparison to other external damage monitoring methods. The experimental results showed that the proposed method has high accuracy for the distance of external damage in the actual scenario. Moreover, the method has good generalizability, which can be easily deployed in video monitoring systems on different transmission corridors.

The power operation site typically faces high voltages, strong currents, and other environmental hazards. Safety fence as a boundary for power operation is of great importance to ensure the safety of operators. The preparation of operation fence is temporary and random, and the risk identification is a complex problem that involves a dynamic interaction between individuals and fence. The existing methods of determining the safety area based on prior knowledge are insufficient. Therefore, this paper proposes a risk identification method based on neighborhood correlation networks and vector calculation. In this method, firstly, the key targets of fence operation are detected by the target detection method. Then, considering that the fence operation target has the characteristics of flat shape and small size; the region proposal network by guided anchoring (GA-RPN) and neighborhood-related network module is introduced to improve the detection accuracy. Finally, the vector calculation criterion is used to judge whether the operator is in the safety fence area, to realize the risk identification of the operation fence overrun. Experiments are conducted on available power fence operation datasets, and the obtained results showed that the proposed method presents better performance than the classical target detection methods.

Electrical power communication networks undertake complicated power services. They interact with physical power grids all the time, which will result in a high risk to power grids when their electric power communication network is attacked. Therefore, to improve the reliability of power systems, it is of great significance to identify the key links of communication networks. Existing research studies focus more on the risk of power communication services undertaken by communication links, which ignores the frequent interaction between power grids and communications networks. In this paper, from a multi-layer network perspective, an assessment method of communication equipment is proposed in view of the operation characteristics of nodes in different systems as well as the dynamic fault propagation between homogeneous communication nodes. Firstly, we build a model of cyber-physical power interdependent network. Then, for communication nodes, a dynamic model is adopted for cascading failure among the nodes based on the load-capacity model. Afterward, we define an index to evaluate the node importance considering characteristics of the operation of both communication networks and power grids. Finally, a local power grid is taken as an example for verification by comparing the effects of different indexes. The simulation results demonstrate the validity of the evaluation method proposed.

The Unmanned Aerial Vehicle (UAV) inspection mode has been gradually implemented in the power system. The UAV inspection image is checked by the target detection technology, but there is no high-precision target detection algorithm as the technical support. In this regard, this paper proposes a target detection algorithm based on the improved RetinaNet which is suitable for transmission lines defect detection. In this algorithm, the shortcomings of the RetinaNet anchor frame extraction mechanism based on Apriori are corrected. At the same time, the number and size of anchor frames are redesigned by using the improved K-means + + algorithm, so that the anchor frame of the improved algorithm gets the highest average IoU (Intersection over Union) value, which matches the actual size of the transmission line defects. Then, in RetinaNet, the feature pyramid network based on DenseNet is built as the backbone network to improve the model accuracy and make the model lighter. The improved model is trained and tested by using the data set of transmission line defects for validation. The results show that the proposed method has advantages and effectiveness in the detection of transmission line defects, and meets the requirements of intelligent inspection in terms of accuracy.

In complex power systems, when power equipment fails, multiple concurrent failures usually occur instead of a single failure. Concurrent failures are so common and hidden in complex systems that diagnosis requires not only analysis of failure characteristics, but also correlation between failures. Therefore, in this paper, a concurrent fault diagnosis method is proposed for power equipment based on graph neural networks and knowledge graphs. First, an electrical equipment failure knowledge map is created based on operational and maintenance records to emphasize the relevance of the failed equipment or component. Next, a lightweight graph neural network model is built to detect concurrent faults in the graph data. Finally, a city’s transformer concurrent fault is taken as an example for simulation and validation. Simulation results show that the accuracy and acquisition rate of graph neural network mining in Knowledge Graph is superior to traditional algorithms such as convolutional neural networks, which can achieve the effectiveness and robustness of concurrent fault mining.

With the deep coupling between the cyber side and the physical side of power systems, the failure of any link of both sides may lead to power outages, so it is necessary to analyze their vulnerability and vulnerable links for targeted improvement of systems. By dynamically attacking the coupled network nodes, this paper proposes a multilevel model and node-to-edge cyber-physical power system and the corresponding indexes system to analyze the vulnerability of the coupled power grid and its key components. The results showed that in the order of the indexes proposed in this paper, attacking surviving power nodes and cyber nodes results in a network crash rate of 25.0% and 66.7% faster than that in the order of “betweenness” and that attacking surviving cyber nodes results in a network crash rate of 89.4% faster than that in the order of “degree.” In terms of attacking power nodes, the index proposed in this paper has the same rate as “degree.” Therefore, the proposed model can better describe the vulnerability of the power grid to withstand attacks.

In view of the current situation that the new generation of smart grids with “double high” characteristics is in urgent need of effective state evaluation methods due to the characteristics of strong volatility and diverse demands, a method of operation state evaluation of smart distribution networks based on free probability theory is proposed, which is combined with high-order moment indexes to describe the operation trajectory of distribution networks from a data-driven perspective. First, the state assessment problem of smart distribution networks is modeled as a binary hypothesis testing problem, and the asymptotic free equation is established based on free probability theory to provide a framework for state assessment of distribution networks. Then, a high-order moment evaluation index is proposed, combined with the sliding time window processing, and the high-order moment sequence was obtained based on the high-dimensional data of the distribution network, which is used to describe the state evolution of the distribution network. Finally, this method is applied to a certain 110-kV distribution network. The analysis of an example shows that the proposed evaluation framework and indicators can effectively reflect the data changes in the distribution network and support the state assessment and evolution analysis of the distribution network.

Insulator is an important equipment of power transmission line. Insulator icing can seriously affect the stable operation of power transmission line. So insulator icing condition monitoring has great significance of the safety and stability of power system. Therefore, this paper proposes a lightweight intelligent recognition method of insulator icing thickness for front-end ice monitoring device. In this method, the residual network (ResNet) and feature pyramid network (FPN) are fused to construct a multi-scale feature extraction network framework, so that the shallow features and deep features are fused to reduce the information loss and improve the target detection accuracy. Then, the full convolution neural network (FCN) is used to classify and regress the iced insulator, so as to realize the high-precision identification of icing thickness. Finally, the proposed method is compressed by model quantization to reduce the size and parameters of the model for adapting the icing monitoring terminal with limited computing resources, and the performance of the method is verified and compared with other classical method on the edge intelligent chip.

Polluted insulators seriously threaten the safe and stable operation of power grids, which attaches great significance to insulator contamination perception. Among the present methods, the non-contact approaches based on infrared images have gradually been widely used, as they are much more safe and are of low cost. However, the thermal effect of insulators is largely affected by meteorological conditions, which makes the infrared image-based methods less accurate. To solve the above problem, we take infrared image and meteorological parameters including humidity and temperature as input, and propose a feature fusion model to perceive insulator contamination in different weather conditions. Firstly, different feature extraction networks are used to perform feature extraction on the two types of data; secondly, the two features are concatenated to fuse together; thirdly, further feature extraction is performed and contamination is classified according to the pollution severity. Case studies show that the proposed method can better explore the relationship between humidity, temperature and pollution level of the insulators, thus can better separate the contamination grades and outperform the conventional infrared image based methods.

The contradiction between the demand of massive intelligent scene caused by the interconnection of things in power system and the bottleneck of the chip itself is becoming more and more serious. How to realize the self-controllable and domestic substitution of power chips has been widely concerned in the power field. Therefore, this paper analyzes the demand for power chips in power system and discusses the scene specificity of power chips. Then, it combs the design architecture of power chips and discusses the application scenarios and research focuses of power chips. Finally, the key scientific problems and major technical bottlenecks of power chips are prospected.

Under the complicated environment of large wind turbines, the vibration signal of a wind turbine has the characteristics of coupling and nonlinearity. The traditional feature extraction method for the signal is hard to accurately extract fault information, and there is a serious problem of information redundancy in fault diagnosis. Therefore, this paper proposed a multidomain feature fault diagnosis method based on complex empirical mode decomposition (CEMD) and random forest theory (RF). Firstly, this paper proposes a novel method of complex empirical mode decomposition by using the correlation information between two-dimensional signals and utilizing the idea of ensemble empirical mode decomposition (EEMD) by adding white noise to suppress the problem mode mixing in empirical mode decomposition (EMD). Secondly, the collected vibration signals are decomposed into IMFs by CEMD. Then, calculate 11 time domain characteristic parameters and 13 frequency domain characteristic parameters of the vibration signal, and calculate the energy and energy entropy of each IMF components. Make all the characteristic parameters as the multidomain feature vectors of wind turbines. Finally, the redundant feature vectors are eliminated by the importance of each feature vector which has been calculated, and the feature vectors selected are input to the random forest classifier to achieve the fault diagnosis of large wind turbines. Simulation and experimental results show that this method can effectively extract the fault feature of the signal and achieve the fault diagnosis of wind turbines, which has a higher accuracy of fault diagnosis than the traditional classification methods.

The wake effect reduces the wind speed at downstream wind turbines (WTs), making it necessary for the central controller to collect wind power generation data from each WT. However, wind farms (WFs) face a more complex problem in maintaining the voltage stability at the WT terminal while following the transmission system operator (TSO) instructions due to the information collection as well as the possible data loss during transmission. Therefore, this study proposes a coordinated control method for WTs and multiple power sources based on model predictive control under wake disturbance conditions, aiming to reduce the average voltage deviation in WT terminals and go close to the rated voltage and ensure effective compliance with TSO commands in large-scale WFs. Accordingly, the Jensen wake model was utilized to accurately calculate the available active and reactive power limits for each WT. Energy storage systems and static Var generators were modeled to coordinate and maintain the voltage in all WT terminals within the feasible range, providing peak shaving and valley filling support to reduce wind energy waste and shortfall, thereby enhancing the economic and operational reliability of WF. Further, the effectiveness of the proposed method was validated in MATLAB/Simulink. [ABSTRACT FROM AUTHOR]

Accurate traffic state prediction plays an important role in traffic guidance, travel planning, etc. Due to the existence of complex spatio-temporal relationships, there are some challenges in forecasting. Firstly, in terms of spatial correlation, some models only consider the road network structure information, and ignore the relative location relationships between nodes. Secondly, some models ignore the different impacts of nodes in the global road network on traffic. To solve these problems, we propose a new traffic state-forecasting model, namely, spatio-temporal attention-gated recurrent neural network (ST-AGRNN). In the proposed model, structure-based and location-based localized spatial features are obtained simultaneously by Graph Convolutional Networks (GCNs) and DeepWalk. The localized temporal features are obtained by gated recurrent unit (GRU). The attention-based approach is used to obtain global spatio-temporal features. Experimental validation is performed with two real-world public datasets, and the results show that the ST-AGRNN model outperforms the state-of-the-art methods.

As a typical spatiotemporal problem, there are three main challenges in traffic forecasting. First, the road network is a nonregular topology, and it is difficult to extract complex spatial dependence accurately. Second, there are short- and long-term dependencies between traffic dates. Third, there are many other factors besides the influence of spatiotemporal dependence, such as semantic characteristics. To address these issues, we propose a spatiotemporal DeepWalk gated recurrent unit model (ST-DWGRU), a deep learning framework that fuses spatial, temporal, and semantic features for traffic speed forecasting. In the framework, the spatial dependency between nodes of an entire road network is extracted by graph convolutional network (GCN), whereas the temporal dependency between speeds is captured by a gated recurrent unit network (GRU). DeepWalk is used to extract semantic information from road networks. Three publicly available datasets with different time granularities of 15, 30, and 60 min are used to validate the short- and long-time prediction effect of this model. The results show that the ST-DWGRU model significantly outperforms the state-of-the-art baselines.

Abstract With the rapid progress of virtual reality technology, 360 videos have become increasingly popular. Given that the resolution of a 360 video is ultra-high (generally 4K to 8K), the encoding time for this type of video is considerably high. To reduce encoding complexity, this study proposed a fast intra algorithm that is based on image texture characteristics. On the one hand, the proposed algorithm determines whether to terminate the coding unit partition early on the basis of texture complexity. On the other hand, the proposed algorithm reduces the number of candidate modes in mode decision according to texture directivity. Experimental results showed that the proposed algorithm can obtain an average time reduction rate of 53% and a Bjontegaard delta rate increase of only 1.3%, which is acceptable for rate distortion performance.