Abstract:To improve the accuracy with which human motion postures are recognized,this paper proposes a method based on an optimized Hidden Markov Model(HMM).Standards for recognizing these postures are set based on the positional relationship between joints and bones in different postures.Optical imaging is used to capture human motion.The initial images are pre-processed using steps such as grayscale conversion and noise reduction.Motion targets are then detected and tracked,and posture features are extracted using the optimized HMM.Human motion posture recognition is then achieved based on state probability and feature matching results.Performance tests conclude that the optimized design method achieves an average recognition error rate of 0.13％.making it superior to traditional methods.

Abstract:A Na2SiO3 precursor solution is formed through low-temperature hydrothermal dissolution with NaOH,using natural inorganic mineral diatomite as the silicon source.Under varying concentrations of Na2SO4,nano-silica is prepared via sulfuric acid precipitation and acidification.The microstructure and properties of the synthesized nano-silica are analyzed comparatively using X-ray diffraction(XRD),scanning electron microscopy(SEM),infrared spectroscopy(FTIR)and BET analysis of the specific surface area.The results reveal that the prepared silica exhibits abundant surface hydroxyl groups,indicating the structural characteristics of centric diatoms have disappeared.Optimal performance is achieved at a Na2SO4 concentration of 0.14 mol/L,where the primary particles have an average diameter of 9.506 nm and a BET specific surface area of 211.39 m2/g.The DBP absorption value is 2.647 cm3/g.These properties meet the performance criteria for Class A nano-silica products as defined by China's chemical industry standard(HG/T3061-1999).

Abstract:In order to address the issues of low recall rates and long response times in the current location positioning system for network users,this paper proposes a location recognition method based on a big data similarity model.Network data from multiple sources,such as base station signals and user trajectories,is collected and standardized.Multi-dimensional features,such as spatio-temporal,behavioral,and network features,are extracted.Different weight coefficients are assigned to these features based on an attention mechanism in order to construct a big data similarity model.The similarity of the spatio-temporal,behavioral and network features is then calculated using the Frechet,EMD and cosine similarity measurement methods.Comprehensive similarity is then obtained through weighted summation,combining attention weight coefficients to precisely locate and identify the user's network position.Experimental results demonstrate that applying the proposed method achieves an optimal user network location recognition accuracy of 99％with a minimum response time of 0.08 seconds.

Abstract:This paper presents a defect detection method for electronic connectors based on rotating template matching and local feature recognition.For images of electronic connectors captured on the test bench,preprocessing steps such as smoothing,binarization,and morphological operations are performed to obtain clear target contours.The template matching algorithm is enhanced by incorporating angular rotation in order to preliminarily locate the region containing electronic connectors.To achieve precise detection,a feature vector based on SIFT gradient magnitude and direction is established.A support vector machine model is then trained to classify and detect defects in the target connectors.Experimental validation shows that the proposed electronic connector defect detection algorithm achieves a detection accuracy of 96.3％and a real-time frame rate of 17.4 FPS,outperforming other general deep learning object detection models.This meets the requirements for defect detection in electronic connectors on production assembly lines.

Abstract:Currently,the support structure for residential precast slab roofs at power stations is mostly of the counterweight type,using counterweights to provide gavitation load and frictional resistance to prevent sliding and overturning.However,due to the limited bearing capacity of the roof,this scheme cannot be used in areas with high wind pressure.Increasing the bottom beam and back pull can effectively reduce the counterweight specification and improve the overall stability of the support structure.Taking an area with wind pressure of 0.42 kN/m2 and snow pressure of 0.48 kN/m2 as an example,the counterweight specifications are compared before and after optimization,provided that the safety requirements are met.The results show that the overall weight specification is reduced by more than 60％,and the cost by more than 5％.This provides a new form of power station installation for prefabricated slab roofs in areas with high wind pressure.

Abstract:Monitoring the stability and safety of caissons and surrounding buildings is significant in modern urban construction and infrastructure maintenance.However,global navigation satellite system real-time dynamic kinematic(GNSS-RTK)positioning technology can generate multiplexing effects and measurement errors when used for horizontal displacement monitoring.This paper proposes an improved,high-precision Kalman filtering baseline solving system.First,the input data are preprocessed for standardization and normalization,and adaptive noise reduction is achieved using the improved CEEMDAN method.An optimized Kalman filter employing the Sage-Husa algorithm is then used to process the signal through prediction and correction.Iterative conditions are then added to achieve systematic and observational noise reduction.Finally,a horizontal displacement monitoring device is built using multi-frequency and multi-mode receivers for a single fixed base station and multiple mobile base stations to obtain the horizontal displacement of the caissons and surrounding buildings.The experimental results indicate that the system achieves a high-precision of up to±2 mm in horizontal displacement monitoring,demonstrating excellent systematic stability and monitoring accuracy.

Abstract:In response to the challenges of target detection posed by dense,small targets and complex backgrounds in UAV aerial infrared images,this paper presents an improved algorithm named EUAV_YOLOv8s,which is based on YOLOv8.Firstly,the algorithm incorporates the SEGS module to enhance the extraction of contextual semantic information by embedding the Squeeze-and-Excitation(SE)attention mechanism in the C2f module and integrating GhostNet to reduce model size.Secondly,a tiny object detection head has been added and a lightweight decoupled head structure has been designed to reduce model complexity without compromising detection accuracy.Finally,the Wise Intersection over Union(WIoU)loss function has been incorporated to enhance the performance of bounding box regression.Experimental results demonstrate that the improved algorithm achieves an mAP@50％of 83.3％and a detection speed of 373 FPS,significantly surpassing the performance of mainstream algorithms.

Abstract:As digital intelligence develops,construction companies are facing the challenge of transforming and upgrading.To reasonably assess the impact of digital intelligence levels on the financial performance of construction companies,this study has developed a digital intelligence evaluation model based on a text analysis of annual reports from listed companies.An empirical study is also conducted on the transmission path between digital intelligence levels,environmental,social and governance(ESG)performance,and financial performance.The sample comprises panel data from 74 listed construction companies from 2018 to 2023.The results show that there is a positive relationship between digital intelligence levels and financial performance,meaning that improving digital intelligence can promote improvement in corporate financial performance.At the same time,ESG performance partially mediates the relationship between digital intelligence levels and financial performance.Based on these findings,recommendations are made to promote the development of digital intelligence comprehensively,prioritize ESG performance,and pursue coordinated development tailored to local conditions.

Abstract:Due to the rapid development of robotics and machine vision technology,service robots are being used in an increasingly wide range of settings,including hotels,supermarkets and hospitals.In this study,an intelligent medication delivery robot has been designed and fabricated to simulate the collection and delivery of medicines between the hospital pharmacy and patient wards.The robot is equipped with an STM32 microcontroller as the main control core and a DC deceleration motor and TB 6612 chip for driving.One OpenMV camera uses template matching to identify ward numbers,while another uses thresholding and binarisation image processing to perform road recognition and line tracking.These features enable the robot to recognize roads and wards automatically,as well as delivering medicine autonomously.

Abstract:In response to the limitations of existing network intrusion detection algorithms,such as low accuracy and excessively long detection times,an XGBoost algorithm optimized using the Tianying algorithm has been designed.First,the data set entering the local network is preprocessed to improve its dimensionality.Then,the tree structure of XGBoost is established to split the objective function synchronously and evaluate the gap between the expected value and the output value.Finally,the Tianying optimization algorithm is introduced to dynamically select the key parameter values at different iteration stages.A weighted histogram is then used to accurately select the splitting points,controlling the scale of the tree structure and reducing the model's complexity.Experimental results demonstrate that the proposed optimization algorithm exhibits strong iterative efficiency.The accuracy rates of the training and test sets are 99.3％and 99.4％,respectively.It also has certain advantages in terms of controlling the detection time.Therefore,it can be concluded that the Tianying-optimized XGBoost algorithm is applicable to intrusion detection.

Abstract:The width of the waterproof coal pillar is an important factor affecting water inrush from fault activation,as well as the caprock of the coal seam.Numerical test results show that as the width of the waterproof coal pillar gradually narrows,the stress concentration phenomenon becomes more pronounced,the roof plastic zone gradually changes from an"elliptical arch"to a"saddle shape",and the floor plastic zone fails in a planar manner.When the width of the waterproof coal pillar is less than 20 meters,the fault becomes prone to activation and bursting.Field measurements verify the rationality of the width of the waterproof coal pillar.The results demonstrate that a monitoring system can analyze rock mass failure caused by fault activation when the working face advances towards the reserved fault waterproof coal pillar.Microseismic events with high energy are stable overall during this period.Ultimately,the 1026 working face of Wugou Coal Mine was successfully mined without any accidents involving fault activation and water inrush occurring.Field application results show that:it is reasonable to set the width of the fault waterproof coal pillar at 20 meters,which improves the mining rate of coal resources and provides reference for similar engineering practices.

Abstract:To address the issue of excessive overall mass in the main frame of horizontally mounted ceramic membrane plate water purification units for mines,structural optimization studies were conducted on the main frame using ANSYS.Firstly,the three-dimensional model of the main frame was imported into ANSYS Workbench to conduct a static analysis.According to the deformation and stress cloud maps,it was concluded that the support frame of the main frame contained too many invalid parts.The variable density method was then used to optimize the topology of the support frame to meet the design requirements.Finally,the support frame was found to be almost stress-free based on the topology optimization results.The optimized support frame model was then established and static test analysis was carried out on the optimized support frame and the main frame.The results show that the overall deformation of the mine's main frame with a transverse ceramic membrane plate water purification device designed on this theoretical basis is 0.063 mm.This meets the actual working conditions and reduces the mass by 7.4％compared to the original,proving the method's rationality and effectiveness.This approach provides a new idea for the lightweight design of the main frame of a mine with a transverse ceramic membrane plate water purification device and has certain practical application value.

Abstract:Electrochemical energy storage technology is pivotal in energy conversion and storage.In an innovative approach to enhancing the energy density and thermal stability of lithium batteries,this study employs coaxial electrospinning-a subset of in-situ polymerisation techniques-to fabricate nanofibre membranes.Cathode electrodes,lithium anodes and precursor solutions are also prepared and encapsulated within the membranes to create a new composite solid-state electrolyte.Experimental testing revealed that the conversion rate of the 1,3-dioxolane monomer in the composite solid electrolyte is 99.1％,the lithium ion migration number is 0.82 and the ion conductivity at 30℃ is 1.04×10-4 S/cm.It only begins to lose weight at a high temperature of 462℃ and can still maintain 24％of undecomposed components above 600℃.These results confirm the successful enhancement of the electrochemical,energy,and thermal stability of lithium batteries.The multifunctional composite materials developed offer a promising new solution for the manufacturing of advanced lithium batteries.

Abstract:This paper is concerned with the multiplier stability and second-order optimality conditions for mathematical programs with switching constraints(MPSC).In the context of the No Nonzero Abnormal Multiplier Constraint Qualification(MPSC-NNAMCQ),it proved that the M-multiplier mapping and the stationary point mapping are both locally bounded and upper semicontinuous.Moreover,it proved that the strong second-order sufficient condition for the refinement of the MPSC problem at a feasible point is equivalent to the strong second-order sufficient condition for the refinement of the S-multiplier,as well as that the strong second-order necessary optimality condition holds without any constraint specification condition.

Abstract:With increasingly severe global environmental issues,achieving green and sustainable development in the construction industry has become a common concern.Using data from 2014 to 2022,this study measures the green total factor productivity(GTFP)of the construction industry with the super-efficient SBM-GML model.It also analyzes the impact of three types of environmental regulations on the GTFP through a mixed panel Tobit model.The results indicate that the GTFP of the construction industry is growing at a slow but positive rate,with technological progress being the dominant driving factor.Command-based and market-based environmental regulations promote the GTFP significantly,while the effect of public-based environmental regulations is less pronounced.Additionally,market-based regulation has a more significant promotional effect in the eastern region,command-based regulation plays a more prominent role in the central region,and the effects of all three types of regulations are not significant in the western region.

Abstract:This study analyses the seismic torsional response of prefabricated reinforced structures with existing framework on high-slopes,focusing on the impact of the torsional response on the stability of the connection parts of prefabricated structures.Based on a specific high-slope engineering project,the high-slope environment was simulated using actual rock mass data.Rayleigh damping was used to add overall damping,and a dynamic stiffness matrix was used to add shear elements.A numerical model of the prefabricated reinforcement structure with the existing framework was then established using the finite element flexibility method to add beam column elements.The results showed that the displacement deformation of the prefabricated reinforced structure with an existing frame under seismic torsional response was mainly in the middle layer,while the displacement deformation in the upper layer was relatively small.The main reaction area was located at the top of the slope and in the backfill area.The seismic torsional response of the existing frame-assembled reinforced structure on the high slope exhibited excellent elastic-plastic hysteresis characteristics under earthquake,suppressing the transmission of seismic forces and controlling the sliding instability of the high-slopes.

Abstract:In order to implement the Outcome-Based Education(OBE)concept effectively,intelligent test paper generation software for the course Principle of Automatic Control has been independently developed using an improved Particle Swarm Optimization(PSO)algorithm.After establishing parameters including the correlation coefficients of knowledge points,question types for assessing knowledge points,support for course objectives,the number as well as scores of question types in the test paper,weights for assessing course objectives and difficulty coefficients,the software takes the following as optimization objectives:assessment of course objectives,correlation coefficients of knowledge points and difficulty coefficients of test questions.The subsequent implementation of the enhanced PSO algorithm facilitates the intelligent generation of the course test papers.The results of teaching practice have validated the effectiveness of the intelligent test paper generation software.Not only can it achieve "differentiated assessment"through the setting of difficulty coefficients,but it can also re-assess students whose individual course objective attainment levels fail to meet expectations by adjusting the weights of assessment course objectives,thereby realizing "continuous improvement".

Abstract:Due to the varying characteristics of different soil types in complex geological environments and the associated challenges of high groundwater levels,loose soil layers and strong permeability,this paper explores the application of the freezing method in the construction of subway connecting passages.Based on a specific subway connecting passage construction project,a comprehensive geological survey of the construction area was conducted.Soil samples were collected to measure moisture content and density parameters,and to calculate soil geology's physical characteristics indicators.Temperature and pressure relief hole measuring points were set up,freezing wall intersection circles were designed,and technical parameters for the freezing method construction were set to complete the construction of subway connecting passages in complex geological environments.The results show that the variation pattern of the frost heave force of the frozen wall in the proposed construction technology aligns with the relevant standards.The construction results demonstrate high stability and excellent performance,meeting construction safety standards for subway connecting passages.This technology has practical value and can be promoted and applied in real-life situations.

Abstract:To address the issues of low efficiency and poor stability in the automatic sorting of electronic components,this study proposes a hybrid method for an automatic sorting equipment for electronic components based on fuzzy logic control.This method also employs genetic algorithms to enhance the fuzzy logic control and utilizes the classic Harris corner detection algorithm for visual color capture.Experimental results demonstrate that the genetic algorithm-optimized fuzzy logic control model maintains an automatic control frequency deviation fluctuating between-0.005 and 0.005 Hz within 7～11 min.The accuracy rates of the proposed hybrid method and three other hybrid methods,namely,improved particle swarm algorithm,tunicate swarm algorithm and dung beetle algorithm,are 98.45%,91.63%,88.45%,and 89.12%,respectively.In conclusion,the proposed hybrid method for the automatic sorting equipment can effectively control fuzzy parameters and exhibits high sorting efficiency.

Abstract:The current intelligent mode of physical education teaching is developing rapidly.However,relying solely on manual observation by teachers to identify students'sports actions during the teaching process can easily lead to issues such as untimely feedback and subjective evaluations.This study therefore proposes a sports action recognition method that integrates pose features with the dynamic time warping(DTW)algorithm.Based on the skeletal point feature output from OpenPose,the method fuses features such as the center of gravity,limb angles and orientations of students during sports activities and uses DTW for alignment and evaluation.Experimental results show that,of the 400 action samples in the training dataset,the OpenPose+DTW model correctly identified 372 samples with a total recognition rate of 93%,which is higher than that of other models.In an experiment involving 50 actions in broadcast gymnastics teaching,the OpenPose+DTW model misjudged five samples,achieving a recognition accuracy of 90%.These results suggest that the sports action recognition model integrating pose features and the DTW algorithm exhibits excellent performance and can support online physical education teaching scenarios.