Becausethe fluctuation and uncertainty of wind power generation bring severe secure and economic challenges to power systems, wind power forecasting becomes the critical part of the management in power systems. In this paper, a hybrid approach based on unequal span segmentation-clustering is proposed to mine the variation trend information in the wind speed series for improving forecasting performance. Firstly, the wind speed series representation is proposed to accurately represent the major variation trend of the wind speed series. It reduces the influence of the non-stationary and irregular behaviors of the wind speed series on the unequal span segmentation-clustering. Secondly, the shape-position comprehensive evaluation is proposed to combine shape and position measures to evaluate the clustering of the trend segments with unequal length. It identifies the trend segments which have comprehensive similarity on both shape and position to better construct forecasting models. Thirdly, novel bayesian optimization mutation operators are proposed to optimally move or add cut points in the unequal span segmentation. It enhances the local search capability of the unequal span segmentation. By comparing different approaches using wind datasets from two wind farms, the effectiveness and advantages of the proposed approach are demonstrated.

This paper is concerned with the event-triggered practical fixed-time containment control for stochastic multi-agent systems, in which the input delay is considered. A novel event-triggered mechanism is proposed, where the containment error of stochastic MASs is considered as a part of the threshold function in the form of an exponential decreasing function, which can provide more freedom to balance system performance and resource utilization. Then, a practical fixed-time containment control scheme is presented, whose convergence time is not affected by the initial conditions. Furthermore, Pade approximation technique is utilized to convert the input delay to the case without delay. The design process of the controller is simplified, and the computational complexity is reduced. Finally, two examples are rendered to demonstrate the performance of the presented method.

Magnetic flux leakage (MFL) detection has been attracted much attention in evaluating the pipeline integrity due to its high efficiency. The precise calculation of the MFL signals plays an important role in the defect size quantification. Currently, finite element model (FEM) and magnetic dipole model (MDM) are two main methods, but FEMs are time-consuming and MDM cannot generate the dynamic MFL signals. To solve the above problems, a defect size quantification method for pipeline dynamic MFL detection via 3-D dynamic magnetic dipole forward model (3-D DMDFM) is proposed. First, a 3-D DMDFM is established, where the dynamic influencing factors, such as the speed, lift-off, and magnetization angle, are integrated into 3-D MDM, so that 3-D DMDFM can generate dynamic MFL signals. Second, to solve the problem of MFL signals distortion due to the dynamic influencing factors, a layer-skipping stacked denoising autoencoder (LS-SDAE) is proposed to reconstruct the MFL signals accurately, where two layer-skipping denoising autoencoders (DAEs) are used to supervise SDAE and learn the underutilized features from the nondirect connection feature layer. Simulated and experimental results show that 3-D DMDFM has both high efficiency and low error when ignoring the modeling complexity. Finally, the reconstructed MFL signals are used to quantify the defect size, and the minimum relative error of the defect depth quantification can reach 1.33%.

Efficient utilization of sustainable energy is imperative for supporting the globally escalating electricity demand. Because the unstable wind energy makes the wind power access challenging for power systems, the wind power forecasting becomes the critical part of the power dispatch. In this paper, a short-term wind power forecasting approach based on model configuration optimization via prequential-cross cooperative validation estimation (PCCVE) is proposed. It enables the hybrid ANN including the convolutional neural network, bidirectional long short-term memory network, and multi-head attention mechanism (CNN-BiLSTM-MHA) to better construct the wind speed–power mapping relationship for improving forecasting performance. Firstly, the box-plot local detection–correction combining the spatial–temporal optimal-weighted fuzzy clustering and the sliding window connected box-plot is proposed to reasonably detect and correct local outlier wind speed points. It prevents CNN-BiLSTM-MHA from being interfered with local outlier wind speed points. Secondly, PCCVE based on the prequential-validation estimation and cross-validation estimation is proposed to more accurately give the estimated error of CNN-BiLSTM-MHA, thus better assisting the optimization of the values of CNN-BiLSTM-MHA’s hyperparameters. It enables CNN-BiLSTM-MHA to efficiently construct the wind speed–power mapping relationship. By comparing different approaches on the actual wind farm dataset, the effectiveness and advantages of the proposed approach are demonstrated.