Power system operation and control are based on power flow calculations. In order to solve the uncertainty of the increasing penetration of renewable energy, the voltage fluctuation at the load point increases in the distribution network, and the inaccuracy of the power flow calculation due to the insufficient power flow data collection capability of the traditional power system. In this paper, a data‐driven power flow analysis model is proposed, a back propagation neural network combined with genetic algorithm (GA) and adaptive moment estimation (ADAM) optimization algorithm model is constructed to analyze the power flow calculation method of distribution networks under stochasticity. Firstly, the power flow initial value information, topology characteristics, and power factor index are introduced to construct a training set, and the mapping relationship between bus voltage and power is fully explored by training the regression model. Second, the GA‐ADAM algorithm is used to optimize the initial values and weight parameters of the model. Finally, it is verified based on IEEE‐33 bus distribution model, and the model is used for power flow calculation, and compared with other methods through each relevant error evaluation indicators. The results show that the model constructed in this paper has small error indicators and high accuracy, which improves the efficiency and accuracy of power flow calculation.

In order to improve the speed and accuracy of power system power flow calculation, this paper proposes a data driven power flow analysis model, and constructs a power flow calculation method based on back propagation neural network combined with genetic algorithm and Adam optimization algorithm model to analyze distribution networks under randomness.image

Protection of an islanded inverter‐based microgrid is challenging because of variable and small fault current contribution of inverter‐based resources (IBR) and the absence of sequence currents. This paper proposes a fast and robust nonpilot directional protection scheme to address this challenge. This scheme relies on support vector machines (SVM) and the harmonic current injection capability of IBRs. Examining the harmonic currents measured by a relay during a fault shows that harmonic currents have similar magnitudes but different orientation under forward and reverse faults. Additionally, harmonic currents have similar orientation but different magnitudes under forward faults at different locations along the protected line. Using this, six SVMs are trained for each relay, given that there are three main types of faults (three‐phase‐to‐ground, line‐to‐line, and line‐to‐ground): three as directional elements and three as zone detection elements. A fault is detected and classified by the undervoltage element of a relay. Then, the measured harmonic currents are routed to the appropriate directionality and zone detection SVMs to facilitate proper relay coordination. The performance of the proposed method is evaluated on the CIGRE North American MV distribution benchmark system under various types of contingency scenarios using PSCAD/EMTDC software.

Protection of an inverter‐based microgrid is challenging due to the lower available fault current and the absence of sequence components. This paper proposes a nonpilot directional protection scheme for inverter‐based microgrids that uses support vector machines to implement fast and reliable protection coordination in case of microgrid faults.image

This novel approach in animal biology could revolutionize identifying endangered species, addressing the issue of misclassifying potentially harmful animals based solely on visual characteristics. Particularly impactful for farmers in agricultural fields, it aims to reduce the heightened risk of venomous animal attacks, ultimately improving safety. Due to a lack of accessible education, illiterate farmers are more susceptible to adopting superstitious beliefs, which tragically leads to fatal snakebites even when medical treatment is readily available. Furthermore, environmental factors can unexpectedly hold typically non‐threatening animals responsible for a large number of human deaths each year. However, the complexity of human recognition of these hazards has prompted the development of a novel design approach aimed at simplifying the process. Integration of the ResNet learning algorithm in conjunction with You Only Look Once (YOLOv5) within the framework is recommended to facilitate real‐time processing and improve accuracy. This combined approach not only speeds up animal recognition but also takes advantage of ResNet's deep learning capabilities. The first phase entails deploying YOLOv5 to detect the presence of snakes in the proposed study, achieving a remarkable 87% precision in snake detection thanks to the synergistic fusion of ResNet and YOLOv5.

To propose a framework that utilizes the real‐time object detection capabilities of the You Only Look Once (YOLO) algorithm to enable rapid and efficient detection of health‐related cellular features or dangerous reptiles in cellular images, reducing the processing time compared to traditional methods. To leverage the high accuracy of the YOLO algorithm in object detection tasks to achieve reliable and precise identification of health‐related cellular features or dangerous reptiles, reducing the chances of false positives or false negatives.image

The extraction effect and feature matching are poor in the three‐dimensional reconstruction of metal cultural relics, which leads to a poor reconstruction effect. Therefore, an optimization method of three‐dimensional reconstruction of metal cultural relics based on 3D laser scanning data reduction is proposed. The overall technical design route of the method is shown as follows. Based on this model, the 3D laser scanning method is used to collect 3D images of metal artefacts, combined with colour space and 2D entropy detection methods to pre‐process 3D images, and feature matching of point clouds is carried out to extract and optimize the significant value of superpixels, and a 3D reconstructed visual model is constructed. Affine transformation is used to obtain the affine invariant moment of the structural light parameters of the visual feature lines of metal cultural relics. The light stability adjustment of the three‐dimensional reconstruction of metal cultural relics is realized by using the linear structural light adjustment method in the HSV colour space. The rigid and non‐rigid registration methods are introduced to match the point cloud. The product quantization algorithm is used to linearize the error function, and the block feature detection and matching model of spatial image is obtained. The noise number is judged by the threshold value, and the three‐dimensional reconstruction technology is combined to realize the three‐dimensional reconstruction of metal relics. The simulation results show that this method has good visual expression ability, high feature recognition rate, and improves the three‐dimensional reconstruction ability of metal relics.

the main directions of the feature points are corresponding to the coordinate axes according to the line structured light block model of the metal cultural relic image shown in Figure 2, and four spherical neighborhoods with different radii are constructed by taking the measured feature points as the origin.image

This paper considers a Non‐Orthogonal Multiple Access (NOMA) network, in which base station transmits a data packet by multi‐user superposition to a destination. In this system, Successive Interference Cancellation (SIC) receiver is used to eliminate the co‐channel interference. Some physical layer parameters, such as modulation order and precoding matrix indicator (PMI), are required for the SIC receiver to separate the overlapped signals. However, an abundance of parameters would bring a large number of signalling overhead. To reduce the signalling overhead, some of these parameters can be blindly detected instead of signalling notification. To detect these kinds of parameters, a novel blind detection algorithm is proposed in this paper. Firstly, feature extraction based on wavelet cluster is introduced to obtain feature information from received data. Then a filter is designed to reduce the interference among these features. Theoretical analysis and simulation results show that the proposed algorithm achieves high detection performance under the computation complexity of approximate the max‐log likelihood algorithm.

This paper proposes an AIT‐WC‐based blind detection algorithm for successive interference cancellation in Non‐Orthogonal Multiple Access system. For the authors’ application scenario, theoretical analysis and simulation results show that the proposed algorithm achieves higher detection performance but with an approximate computation complexity, compared with the traditional max‐log likelihood algorithm. The proposed AIT‐WC algorithm is effective for the joint blind detection of modulation order and precoding matrix indicator (PMI). When Signal‐to‐Noise Ratio (SNR) is the same, compared with the max‐log likelihood algorithm, the proposed algorithm has an advantage before the blind detection accuracy reaches 100%. Meanwhile, the proposed AIT‐WC algorithm can reach 100% blind detection accuracy and the peak throughput in advance than the max‐log likelihood algorithm, and it also demonstrates that the proposed algorithm is more stable than the max‐log likelihood algorithm when the PMI is different.image