Search

Undesirable microstructures, with thin or uneven compound layers, pores, pearlitic structures, vein-like nitrides, etc., are detrimental to mechanical and tribological properties of nitrocarburized components. Here, we report a novel heat process in which a thicker nanobainitic layer, consisting of γ′ lamellas and α-Fe laths with high-density α′′-phase GP-zones, is obtained by heating the undesirable nitrocarburized layer at 700 °C (austenitizing) followed by isothermal aging at 225 °C for 14 h. The total phase-transformation sequences of this process are described as: ε + γ′ → γ-N/C → γ′ + α-Fe + GP-zones. The thicker nanobainitic layer exhibits much higher hardness, load-bearing capacity, fracture toughness and wear resistance as compared to the nitrocarburized layer, and corresponding mechanisms for strengthening and toughening are shown. [Display omitted] [ABSTRACT FROM AUTHOR]

• A novel segmentation method for breast ultrasound is proposed. • Breast lesion is segmented by semantic classification of superpixels. • A K-NN based reclassification is adopted to refine the segmentation. • Experimental results show the advantages of the proposed method. Breast cancer is a great threat to females. Ultrasound imaging has been applied extensively in diagnosis of breast cancer. Due to the poor image quality, segmentation of breast ultrasound (BUS) image remains a very challenging task. Besides, BUS image segmentation is a crucial step for further analysis. In this paper, we proposed a novel method to segment the breast tumor via semantic classification and merging patches. The proposed method firstly selects two diagonal points to crop a region of interest (ROI) on the original image. Then, histogram equalization, bilateral filter and pyramid mean shift filter are adopted to enhance the image. The cropped image is divided into many superpixels using simple linear iterative clustering (SLIC). Furthermore, some features are extracted from the superpixels and a bag-of-words model can be created. The initial classification can be obtained by a back propagation neural network (BPNN). To refine preliminary result, k-nearest neighbor (KNN) is used for reclassification and the final result is achieved. To verify the proposed method, we collected a BUS dataset containing 320 cases. The segmentation results of our method have been compared with the corresponding results obtained by five existing approaches. The experimental results show that our method achieved competitive results compared to conventional methods in terms of TP and FP, and produced good approximations to the hand-labelled tumor contours with comprehensive consideration of all metrics (the F1-score = 89.87% ± 4.05%, and the average radial error = 9.95% ± 4.42%). Image, graphical abstract [ABSTRACT FROM AUTHOR]

With structural changes to electricity industry in recent years, there is an increasing emphasis on short-term load forecasting (STLF). STLF is the essential part of power system planning and operation. In deregulated environment, STLF greatly affects system marginal price. Therefore many kinds of models for STLF have been proposed in recent 30 years. However, an individual forecasting model cannot work well in every possible condition. To overcome this problem, an integrated model, based on the idea of simulative forecasting, which can synthesize individual forecasting model results, is proposed in this paper. In practice, STLF is always affected by a variety of nonlinear factors. Since it is difficult to identify the nonlinearity by conventional methods, a novel approach is proposed which is able to deal with most factors affecting power load. The major idea in the approach is to define a mapping function for each factor. Finally, an algorithm for optimizing mapping functions is presented. Case studies show that it is a successful solution to improve the accuracy of STLF. [Copyright &y& Elsevier]