Search

Existing manual detection methods have limitations, particularly for pavement cracks in complex backgrounds, which are manifested in low recognition accuracy, high misjudgment rate, and long time-consuming. To overcome these problems, artificial intelligence technology and charge-coupled devices (CCD) imaging technology are combined to construct an automatic identification method for pavement hidden cracks under complex background interference conditions. First, the classic semantic segmentation model U-net is selected as the basic model, and the MobileNet lightweight network is utilized to replace the encoder part of U-Net with huge parameters, to realize the lightweight of the model and improve the segmentation effect of pavement cracks. On this basis, the Atrous Channel Pyramid Attention (ACPA) mechanism is introduced into the U-net to further improve contextual information capability to focus on selectively relevant features. A pavement crack data set containing different complex and diverse crack types and background noise is used to evaluate the effectiveness and scope of application of the developed model. Quantitative evaluation results show that the developed model achieves an overall performance in the test set with a precision of 88.84%, recall of 89.76%, accuracy of 98.87%, and IoU of 89.95%, respectively. Combined with the analysis of the results of the comparison experiment and the ablation experiment, it can be inferred that the utilization of the MobileNet lightweight network to replace the encoder part of U-net can effectively construct a lightweight model while the ACPA module can effectively perform multi-scale and long-distance cross-channel interaction, help suppress useless features, strengthen useful features, and help the network learn stronger feature representations of hidden areas of pavement cracks.

The ac-dc hybrid distribution network is a credible path for the future evolution of distribution network. State estimation is a paramount foundation for the safe and stable operation of the complex distribution network. The application of the centralized state estimation method in an ac-dc distribution network has some obstacles, such as low computational efficiency, large communication capacity, and privacy protection problem. Based on the three-stage state estimation theory, this paper established a three-stage state estimation model for the ac-dc hybrid distribution network integrating supervisory control and data acquisition system and phasor measurement unit. The proposed three-stage estimation model achieves linearization of the nonlinear state estimation for the ac-dc hybrid distribution network. That is, the first and third stages simply solve the linear state estimation problem, and the second stage is a one-step nonlinear transform. Moreover, the alternating direction method of multipliers (ADMM) is applied to solving the distributed problem. Thus, an accurate and efficient three-stage distributed state estimation method for the ac-dc hybrid distribution network is proposed. In this method, ac subsystems and dc subsystems execute state estimation tasks based on local information and eventually achieve the overall consistency of the system state estimation through the transfer iteration of the boundary information, respectively. The combination of bilinear theory and ADMM ensures the convergence of distributed state estimation while improving computational efficiency. The simulation results verified the advantage of the proposal over the existing methods in many aspects.

With the proliferation of renewable energy, the uncertainty has challenged the continuous operation of microgrids; thus, it is of importance to tackle uncertainties in power system operation. In this paper, a global sensitivity analysis (GSA) method is proposed to evaluate the influence of uncertainties on the power flow of islanded microgrids (IMGs). First, a probabilistic power flow model for IMGs is established considering the droop-controlled distributed generation units and the uncertainties of renewable energy generation output and load demands. Then, the global sensitivity analysis is introduced to identify important variables that affect IMG power flow. In addition to conventional GSA indices, the Shapley value-based GSA index is designed to evaluate the influence of correlated input variables. Moreover, the sparse polynomial chaos expansion is used to establish the surrogate models of IMG power flow, which improves the efficiency of GSA. Finally, the proposed method is tested on the 33-bus and 69-bus IMG systems, and the simulation results are compared with those considering other methods. The rankings of random input variables that affect IMG power flow are given, and the influence of correlation between different variables is discussed.