The robust control of high precision electromechanical systems, such as micropositioners, is challenging in terms of the inherent high nonlinearity, the sensitivity to external interference, and the complexity of accurate identification of the model parameters. To cope with these problems, this work investigates a disturbance observer-based deep reinforcement learning control strategy to realize high robustness and precise tracking performance. Reinforcement learning has shown great potential as optimal control scheme, however, its application in micropositioning systems is still rare. Therefore, embedded with the integral differential compensator (ID), deep deterministic policy gradient (DDPG) is utilized in this work with the ability to not only decrease the state error but also improves the transient response speed. In addition, an adaptive sliding mode disturbance observer (ASMDO) is proposed to further eliminate the collective effect caused by the lumped disturbances. The sterling performance is revealed with intensive tracking simulation experiments and demonstrates the improvement in the accuracy and response time of the controller.

Handwriting is a form of biometric behavioral characteristic with evident individual distinctiveness. With the surge of the deep learning trend and the demand for forensic identification, handwriting identification has become one of the focal points of research in the field of pattern recognition. The research on handwriting identification in major world languages has reached a mature stage. However, there is still a notable lack of relevant research in the field of Mongolian handwriting identification, despite the fact that Mongolian is used by over 4 million people in China. This paper embarks on an initial exploration of Mongolian handwriting identification by constructing a convolutional neural network named MWInet-12. In this paper, the model evaluation experiments were conducted using a dataset comprising 156,372 samples contributed by 125 authors from the MOLHW dataset. The dataset was divided into training, validation, and test sets in an 8:1:1 ratio. The final results of the experiments reveal impressive accuracy on the test set, achieving a top-1 accuracy of 89.60% and a top-5 accuracy of 97.53%. Furthermore, through comparative experiments involving Resnet, Fragnet, and GRRNN models, this paper establishes that the proposed model yields the most favorable results for Mongolian handwriting identification.

This research investigates the fatigue crack propagation on the lower surface of rocket deflector troughs of offshore rocket launch platforms. Initially, a numerical model of the offshore rocket launch platform is established using ABAQUS based on the Extended Finite Element Method (XFEM). Subsequently, two variable parameters—namely, the initial crack length and initial tilt angle—are introduced. This research systematically analyzes the impact of these parameters on fatigue crack propagation patterns in both the maximum stress and maximum deformation regions of the deflector channels under the combined conditions of high temperature and impact. Finally, the research indicates that the propagation length of surface cracks in the deflector trough exhibits a trend of initial increase followed by a subsequent decrease with the increase in the preset inclination angle. Notably, the stable propagation rate of the crack in the region of maximum deformation surpasses that observed in the region of maximum stress. Through meticulous comparative analysis, it is evident that temperature loading significantly fosters the initiation and propagation of cracks, particularly in the upper region of the deflector channel's lower surface.

Sea ice extraction and segmentation of remote sensing images is the basis for sea ice monitoring. Machine learning-based image segmentation methods rely on manual sampling and require complex feature extraction. Deep-learning semantic segmentation methods have the advantages of high efficiency, intelligence, and automation. Sea ice segmentation using deep learning methods faces the following problems: in terms of datasets, the high cost of sea ice image label production leads to fewer datasets for sea ice segmentation; in terms of image quality, remote sensing image noise and Severe weather conditions affects image quality, which affects the ac-curacy of sea ice extraction. To address the quantity and quality of the dataset, this study used multiple data augmentation methods for data expansion. To improve the semantic segmentation accuracy, the SC-U2-Net network was constructed using multi-scale inflation convolution and a multi-layer Convolutional Block Attention Module (CBAM) attention mechanism for the U2-Net network. The experiments showed that (1) data augmentation solved the problem of an insuffi-cient number of training samples to a certain extent and improved the accuracy of image seg-mentation. (2) This study designed a multilevel Gaussian noise data augmentation scheme to improve the network's ability to resist noise interference and achieve a more accurate segmenta-tion of images with different degrees of noise pollution. (3) The inclusion of a multi-scale inflation perceptron and multi-layer CBAM attention mechanism improved the ability of U2-Net network feature extraction and enhanced the model accuracy and generalization ability.

Based on the plate tectonic theory, using seismic, geology, and public literature data, the proto-type basins and lithofacies paleogeography during the main geological periods of the passive continental margin of East Africa are restored, and the architecture characteristics of the basins and the differences in sedimentary fillings are analyzed. Combined with the dissection of discovered oil and gas reservoirs, four types of hydrocarbon accumulation models are established, the favorable hydrocarbon accumulation combinations (plays), and the next exploration directions in this region are explored. The passive continental margin basins in East Africa experienced three proto-type stages: intracontinental aborted rift during the Late Carboniferous-Triassic Karoo period, intracontinental-intercontinental rift during the Jurassic period, and passive continental margin basin since the Cretaceous period. Affected by the difference in sedimentary filling thickness between the rift period and depression period, four types of passive continental margin basins are formed: rifted type, depressed type, faulted depression type, and reformed delta type. The maximum sedimentary thickness of the rifted basins during the rift period is greater than 3500 m, and that during the depression period is less than 4000 m, forming a "single-source structure type" hydrocarbon accumulation model. The exploration of rifted basins should focus on giant structural traps at the top of the rifted sequences. The maximum sedimentary thickness of the depressed basins during the depression period exceeds 4000 m, forming a "single-source fan group type" hydrocarbon accumulation model. The dominant exploration directions of depressed basins are sedimentary sand bodies and structural-lithologic traps during the depression and drift periods. The maximum sedimentary thickness of the faulted depression basins during the rift period is greater than 3500 m, and that during the depression period is greater than 4000 m, forming a "double-source and double-combination type" hydrocarbon accumulation model. The exploration of rifted basins should focus on giant slip-collapse-debris flow sedimentary sand bodies on the upper and middle slopes. In addition to the large sedimentary thickness of the early rift and the central depression, the delta thickness of the reformed delta basins during the late depression period since the Miocene is more than 4000 m, and four structural belts (growth fault, mud diapir, thrust fold, and foredeep gentle slope) were formed from onshore to offshore, forming a "three-source and multi-combination type" hydrocarbon accumulation model. All four structural belts can form giant oil and gas fields.

Disposal and treatment of excess sludge is a huge problem in waste water treatment plants. Discharge plasma oxidation is an effective approach for sludge dewatering and digestion. In this study, the excess sludge disintegration by non-thermal discharge plasma coupled with thiosulfate (TSA) was investigated. The soluble chemical oxygen demand (SCOD) increased to 404.93 mg L-1 after 20 min of single discharge plasma treatment, and it further increased to 549.08 mg L-1 after adding 15 mmol L-1 of TSA; the water content of filter cake also further decreased in the presence of TSA. There existed an appropriate TSA dosage. In the discharge plasma coupled with TSA system, reactive oxygen species (·OH and ·O2-) were produced and played important roles in sludge disintegration. The addition of TSA promoted the production of ·OH. These reactive oxygen species destroyed the floc structures and promoted the transformation of organic substances, leading to reduction in average size of sludge flocs. The ratio of soluble extracellular polymer substances (S-EPS) was enhanced, while the ratio of tightly bound fraction was reduced after treatment. Thus, discharge plasma coupled with TSA promoted the lysis of microbial cells and released intracellular organic matter and bound water, finally improving sludge dewaterability.

In the sphere of urban renewal in historic districts, preserving and innovatively reinterpreting traditional architectural styles remains a primary research focus. However, the modernization and adaptive reuse of traditional buildings often necessitate changes in their functionality. To cater to the demands of tourism in historic districts, many traditional residential buildings require conversion to commercial use, resulting in a mismatch between their external form and internal function. This study explores an automated approach to transform traditional residences into commercially viable designs, offering an efficient and scalable solution for the modernization of historic architecture. We developed a methodology based on diffusion models, focusing on a dataset of nighttime shopfront facades. By training a Low-Rank adaptation (LoRA) model and integrating the ControlNet model, we enhanced the accuracy and stability of generated images. The methodology’s performance was validated through qualitative and quantitative assessments, optimizing batch size, repetition, and learning rate configurations. These evaluations confirmed the method’s effectiveness. Our findings significantly advance the modern commercial style transformation of historical architectural facades, providing a novel solution that maintains aesthetic and functional integrity, thereby fostering breakthroughs in traditional de-sign thinking and exploring new possibilities for the preservation and commercial adaptation of historical buildings.

The rebound behaviors of rubber balls based on ethylene-propylene-diene monomer (EPDM) / chlorinated butyl rubber (CIIR) blends were investigated systematically. A novel method was proposed to characterize the damping capacity of the EPDM/CIIR blends. The damping capacity can be represented by the rebound height of rubber balls, and lower rebound height corresponds to better damping capacity. A mathematical model expressed by an equation obtained through theoretical derivation has been proposed to predict the rebound height of the rubber balls. The energy dissipation rate (EDR) defined by the ratio of the height loss to the rebound time was proposed to further characterize the damping capacity. The EDR value shows the highest for the pure CIIR and lowest for the pure EPDM, showing a decrease trend with the increase of EPDM content in the rubber blends. By comparison one can obtain that the damping capacity of the EPDM/CIIR blends decreases with the decrease of external excitation, the conclusion of which plays a key role in the formulation design of viscoelastic damping rubber materials.

Severe COVID-19 patients show impaired IFN-I response due to decreased IFN-β production, allowing persistent viral load and exacerbated inflammation. The SARS-CoV-2 nucleocapsid protein has been implicated in inhibiting IFN-I response through interfering with IFN-I signaling. This study reveals that SARS-CoV-2 nucleocapsid protein enhances interaction between human SUMO-conjugating enzyme UBC9 and MAVS. Increased MAVS-UBC9 interaction leads to enhanced SUMOylation of MAVS, inhibiting its ubiquitination, resulting in the inhibition of phosphorylation events involving IKKα, TBK1, and IRF3, disrupting IFN-I signaling. These results provide essential insights into the intricate regulation of the host's innate immunity during SARS-COV-2 invasion. Understanding these complex molecular mechanisms is crucial in developing effective therapeutic interventions against COVID-19 and potential future viral outbreaks.