Search

Various complex and minor defects exist inside 3-D-printed lattice structures that are difficult to detect using traditional detection methods. This study focuses on enhancing the accuracy of YOLO V7 in identifying these intricate defects. Initially, a convolutional block attention module (CBAM) was integrated into the YOLO V7 network to emphasize key image details and enhance detection precision. Then, adaptively spatial feature fusion (ASFF) was used to enable the model to select and fuse features of different scales, as needed, to improve the detection performance of the model. These improvements address the issue of inconsistent attention to features at different scales in the classic feature pyramid network (FPN) of the YOLO V7. To assess the performance of the proposed model, this study utilized two datasets—the face-centered cubic (FCC) dataset and the combined body-centered cubic (BCC) and FCC datasets. The improved YOLO V7 demonstrated an average accuracy of 96.9%, a 2.4% enhancement compared to the unmodified model.

4-D printed structures have excellent application prospects in defense, aerospace, medical, and other fields due to their alterable shape and function. Current 4-D printed structures are limited to make predefined shape changes in response to particular external stimuli by the design of the materials and structures. However, it is crucial to achieve adaptive accurate control of its deformation in a dynamic environment, particularly in highly sensitive tasks. Therefore, we propose a deformation regulation strategy for a 4-D printed flexible intelligent structure based on an adaptive integral terminal sliding mode controller. The effectiveness of the algorithm is verified using a lab-made finger structure made of Liquid Crystal Elastomer as a 4-D printed structural part. Considering the condition of disturbed and nonpredetermined external excitation, the bending angle of the finger joint can be stable within $\mathbf {\pm 2.1\%}$ of the expected value based on the feedback of the self-sensing information, which enables collaborative tracking of the deformation state of the external human finger. This controller-based, adaptive, and accurate deformation regulation is not possible with the existing 4-D printed deformation structures and has not yet been reported nationally or internationally.

With the accelerating of electrification of ship power system, it enables a more efficient and flexible operation to alleviate the pressure on environment pollution and energy consumption. Hitherto, many existing literatures have achieved the optimal electric energy management on board, with the application of diesel generator and electric energy storage, while the thermal energy scheduling has been barely studied. In this paper, the electric energy and thermal energy scheduling are studied simultaneously, thus expanding the ship power system into hybrid power ship system (HPSS). Considering the total operating cost and Greenhouse Gases (GHG) emissions, the energy scheduling optimization problem is defined. To improve the quality of solution, an adaptive moth flame optimization based on multi-populations (AMFOMP) is proposed. Specifically, moth sorting strategy based on logistic mapping, nonlinear flames number update mechanism and modified update formula of moth are carried out to improve the exploration and exploitation ability in iterations. Moreover, modified opposition-based learning and population diversity strategy are implemented to prevent algorithm from stagnation. AMFOMP is carried out into electric and thermal energy scheduling and the simulation results show the economic and environmental benefits of the proposed HPSS.