Reliable and accurate aeroengine remaining useful life (RUL) prediction plays a key role in the aeroengine prognostics and health management (PHM) system. However, due to the epistemic uncertainties associated with aeroengine systems, prediction errors are unavoidable and sometimes significant in traditional deterministic point prediction methods. To improve the accuracy and credibility of RUL prediction, a novel prediction interval (PI) estimation method is proposed to quantify the uncertainties in RUL prediction. The proposed method involves data clustering, mathematical–statistical analysis, and deep learning techniques and is achieved through off-line and online phases. In the off-line phase, an enhanced fuzzy c-means (FCM) algorithm is proposed to divide the aeroengine health status into several discrete states. After labeling the health state of each sampling point, PIs are computed for them. This step is achieved by the empirical distributions of errors associated with all instances belonging to the health state under consideration. In the online phase, a bidirectional long short-term memory (Bi-LSTM) network is employed to estimate the boundaries of point prediction, and thus, the PI of aeroengine RUL is generated. The aeroengine degradation dataset from NASA is used to validate the proposed RUL PI estimation method. The results obtained indicate that the proposed method is a promising tool for providing reliable aeroengine RUL interval estimates, which can inform maintenance-related decisions.

Airborne laser scanning (ALS) point cloud has been widely used in the fields of ground powerline surveying, forest monitoring, urban modeling, and so on because of the great convenience it brings to people’s daily life. However, the sparsity and uneven distribution of point clouds increases the difficulty of setting uniform parameters for semantic classification. The PointNet++ network is an end-to-end learning network for irregular point data and highly robust to small perturbations of input points along with corruption. It eliminates the need to calculate costly handcrafted features and provides a new paradigm for 3D understanding. However, each local region in the output is abstracted by its centroid and local feature that encodes the centroid’s neighborhood. The feature learned on the centroid point may not contain relevant information of itself for random sampling, especially in large-scale neighborhood balls. Moreover, the centroid point’s global-level information in each sample layer is also not marked. Therefore, this study proposed a modified PointNet++ network architecture which concentrates the point-level and global features on the centroid point towards the local features to facilitate classification. The proposed approach also utilizes a modified Focal Loss function to solve the extremely uneven category distribution on ALS point clouds. An elevation- and distance-based interpolation method is also proposed for the objects in ALS point clouds which exhibit discrepancies in elevation distributions. The experiments on the Vaihingen dataset of the International Society for Photogrammetry and Remote Sensing and the GML(B) 3D dataset demonstrate that the proposed method which provides additional contextual information to support classification achieves high accuracy with simple discriminative models and new state-of-the-art performance in power line categories.

Interval estimation of landslide displacement prediction is crucial for the landslide early warning. To improve the accuracy of landslide displacement prediction, a dynamic prediction model based on double moving average (DMA) method and long short-term memory (LSTM) network is proposed. The DMA method is used to decompose the cumulative displacement of landslide into trend and periodic displacements, while the LSTM network is used to predict the trend and periodic displacements. The predicted values of cumulative displacement are obtained by reconstructing the predicted trend and periodic displacements. To quantify the uncertainties associated with the predicted values, the prediction intervals are derived according to the probability estimation theory. The proposed interval estimation method was validated on the Baishuihe landslide in the Three Gorges Reservoir area of China. The results show that the LSTM network performs better than the support vector machine (SVM) and Elman network, while the DMA decomposition method outperforms the single moving average (SMA) method. Besides, the proposed estimation method can construct high-quality prediction intervals.

Integration of different models may improve the performance of landslide susceptibility assessment, but few studies have tested it. The present study aims at exploring the way to integrating different models and comparing the results among integrated and individual models. Our objective is to answer this question: Will the integrated model have higher accuracy compared with individual model? The Lvliang mountains area, a landslide-prone area in China, was taken as the study area, and ten factors were considered in the influencing factors system. Three basic machine learning models (the back propagation (BP), support vector machine (SVM), and random forest (RF) models) were integrated by an objective function where the weight coefficients among different models were computed by the gray wolf optimization (GWO) algorithm. 80 and 20% of the landslide data were randomly selected as the training and testing samples, respectively, and different landslide susceptibility maps were generated based on the GIS platform. The results illustrated that the accuracy expressed by the area under the receiver operating characteristic curve (AUC) of the BP-SVM-RF integrated model was the highest (0.7898), which was better than that of the BP (0.6929), SVM (0.6582), RF (0.7258), BP-SVM (0.7360), BP-RF (0.7569), and SVM-RF models (0.7298). The experimental results authenticated the effectiveness of the BP-SVM-RF method, which can be a reliable model for the regional landslide susceptibility assessment of the study area. Moreover, the proposed procedure can be a good option to integrate different models to seek an “optimal” result.

In recent decades, landslide displacement forecasting has received increasing attention due to its ability to reduce landslide hazards. To improve the forecast accuracy of landslide displacement, a dynamic forecasting model based on variational mode decomposition (VMD) and a stack long short-term memory network (SLSTM) is proposed. VMD is used to decompose landslide displacement into different displacement subsequences, and the SLSTM network is used to forecast each displacement subsequence. Then, the forecast values of landslide displacement are obtained by reconstructing the forecast values of all displacement subsequences. On the other hand, the SLSTM networks are updated by adding the forecast values into the training set, realizing the dynamic displacement forecasting. The proposed model was verified on the Dashuitian landslide in China. The results show that compared with the two advanced forecasting models, long short-term memory (LSTM) network, and empirical mode decomposition (EMD)–LSTM network, the proposed model has higher forecast accuracy.

Accurate PM2.5 concentration prediction is crucial for protecting public health and improving air quality. As a popular deep learning model, deep belief network (DBN) for PM2.5 concentration prediction has received increasing attention due to its effectiveness. However, the DBN structure parameters that have a significant impact on prediction accuracy and computation time are hard to be determined. To address this issue, a modified grey wolf optimization (MGWO) algorithm is proposed to optimize the DBN structure parameters containing number of hidden nodes, learning rate, and momentum coefficient. The methodology modifies the basic grey wolf optimization (GWO) algorithm using the nonlinear convergence and position update strategies, and then utilizes the training error of the DBN to calculate the fitness function of the MGWO algorithm. Through the multiple iterations, the optimal structure parameters are obtained, and a suitable predictor is finally generated. The proposed prediction model is validated on a real application case. Compared with the other prediction models, experimental results show that the proposed model has a simpler structure but higher prediction accuracy.