To solve the reliability problem of multicast transmission in industrial Internet of things （IIoT） based on SDN （software defined network），the data retransmission mechanism based on cache is studied，and a cache-based reliable multicast scheme is proposed. A cache-based reliable multicast （CBRM） framework and the corresponding reliable multicast algorithm are designed. Cache nodes for temporarily storing multicast data are deployed in selected locations in SDN. If receiving or forwarding node detects packet loss，it will require its upstream nodes to retransmit the lost packet. To support data retransmission of cache nodes，a UDP （user datagram protocol）-based reliable transmission protocol （URTP） is designed. The results of Mininet simulation show that the CBRM framework and the corresponding reliable multicast algorithm can guarantee the reliability of multicast transmission at a much lower cost than traditional methods.

To address the mismatch between space-to-ground resources supply and demand caused by the diversified traffic requirements of communication terminals in the beam-hopping satellite system，as well as the challenge of limited energy resources of machine-type devices in upward transmission，a resource adaptation scheme is proposed based on a multi-agent soft actor-critic（MASAC）approach utilizing maximum entropy reinforcement learning. Firstly，a two-stage transmission system model is constructed to investigate the synergistic effect of beam-hopping and non-orthogonal multiple access（NOMA）on the basis of the space-to-ground resource mismatch problem. Additionally，an energy harvesting and collection mechanism is introduced to optimize the relationship between terminal device energy harvesting and signal transmission. On this basis，a multi-objective optimization problem is established for beam-hopping pattern selection，time slot allocation，and rate and power control by integrating the uplink and downlink transmission processes. MASAC maximum entropy reinforcement learning is employed for optimization，obtaining an optimal joint control strategy. Experimental results show that the proposed scheme can effectively allocate resources for space-to-ground resource matching and meet the signal transmission requirements of energy-constrained machine terminals. Compared with the benchmark algorithm，the proposed algorithm exhibits superior performance.

In order to represent and process time information， the extended research on temporal RDF （temporal resource description framework） model has been carried out， and how to effectively manage temporal RDF data has gradually become an important issue. Constructing a reasonable and effective index mechanism is one of the important ways to manage temporal RDF data and improve query efficiency. A two-stage indexing method for bitemporal RDF star query was proposed. First-stage indexing proposes a candidate set generation method based on bitemporal RDF node neighborhood information， and second-stage indexing further proposes a result set generation method based on bitmap indexing. With two-stage indexing， expensive join operations in star queries can be avoided and query results can be obtained quickly. This paper conducts comparative experiments in terms of query efficiency and index performance to verify the effectiveness of the proposed method.

The burning through point （BTP） is an important parameter in the sintering process， which directly affects the efficiency of the sintering machine. Due to the multi-working conditions and time-varying characteristics of the sintering production process， the prediction performance of the global model is insufficient. Therefore， a burning through point prediction model using XGBoost as a local model in the just-in-time learning framework is proposed， namely JITL-XGBoost. Firstly， the KL divergence similarity measurement method is used to extract the characteristics of the sample to be tested， and the most relevant data set of the sample to be tested is selected. Secondly， this dataset is used as input to the XGBoost model to predict the location of the burning through point. In addition， the impact of related dataset numbers on model prediction accuracy and model computation time is considered. Finally， by comparing with other models， the results show that the model built has the best prediction accuracy within a reasonable time， providing new guidance for improving the efficiency of sintering machines.

In order to establish the flow stress models of V-N microalloyed steel under multi-pass deformation conditions， multi-pass compression experiments were conducted on experimental steel using a DIL805 thermal expansion phase transformation tester， and the stress-strain curves were plotted. The flow stress of experimental steel was simulated with high accuracy by Hensel-Spittel model under single-pass deformation conditions. When the deformation temperature and strain rate remained constant， genetic algorithm was used to optimize the parameters of the Hensel-Spittel model under multi-pass deformation conditions. The support vector machine（SVM） algorithm was used to establish the corresponding relationships between the static recrystallization volume fraction before deformation， austenite grain size before deformation， dislocation density， deformation temperature， strain rate and the model parameters. The results show that the predicted flow stress under multi-pass deformation conditions is in good agreement with the measured values. The research results can provide strong support for accurately describing the flow stress of V-N microalloyed steel under multi-pass deformation conditions.

To solve the heat dissipation problem of electronic devices， the flow and heat transfer features of elastic pillar with different initial inclinations in microchannel are numerically investigated. Furthermore， the influences of dynamic characteristics and inclination angle of the elastic pillar on fluid heat transfer performance， hydrodynamic friction factor， and overall hydrothermal efficiency are probed. The results indicate that 1.25 μm oscillation is generated when the inclination angle of the elastic pillar is 0.2π. This small amplitude is not conducive to the generation of vortices. On the other hand， an elastic pillar in a fully flapping mode can generate periodic vortices， which is more conducive to the disturbance to the boundary layer and the mixing of cold and hot fluids， thereby enhancing heat transfer. In the range of Re from 800 to 1 200， the elastic pillar has the optimal inclination angle of 0.6π. Under the optimum working condition， at the expense of high mechanical loss， the heat transfer performance of the fluid can be improved by 63.5%， and the total hydrothermal efficiency can be increased by 7.5%.

Sb/CNTs（carbon nanotubes） nanotubes were prepared by electrolysis in choline chloride-ethylene glycol eutectic solvent using solid Sb₂S₃/CNTs as cathode and graphite as anode. The electrochemical reduction behavior of solid Sb₂S₃ was studied by cyclic voltammetry and the effects of electrolysis voltage， electrolysis time and temperature on the electro-desulfurization of solid-state Sb₂S₃ were systematically investigated. The cyclic voltammetry results show that solid-state Sb₂S₃ can be directly electrochemically reduced metallic Sb in the deep eutectic solvent， and the electrochemical reduction of solid-state Sb₂S₃ to metallic Sb is a one-step reduction process. The XRD and EDS results indicate that increasing the electrolysis voltage and extending the electrolysis time lead to more comprehensive electro-desulfurization of solid-state Sb₂S₃. After electrolysis at a voltage of 2.4 V for 6 h at 353 K， solid-state Sb₂S₃/CNTs can be completely electrochemically reduced to Sb/CNTs. SEM results indicate that as the electrolysis voltage increases， the particles of the electrolysis product become smaller and the surface porosity increases. The findings provide a novel pathway for the direct electrochemical reduction of solid sulfides to metals at low temperatures and even at room temperature.

In view of the complex temperature changes in metal additive manufacturing that lead to residual stress in the parts and then form holes，cracks and other defects，TC4 titanium alloy parts with circular tracks were taken as the research object，and ultrasonic micro-forging technology was used to apply high-frequency micro-vibration on the surface of the workpiece，so as to achieve the effect of refining the grains and reducing the residual stress. By studying the thermal stress changes of different characteristic points of the single-layer laser-cladding layer，the thermal stress field results were obtained and imported into the ultrasonic rotaty micro-forging system， and then the effects of the diameter of the micro-forging head，the rotating speed of the tool head，the amplitude and the forging temperature on the residual stress of the cladding layer were studied. The results showed that laser fuse additive manufacturing has the characteristics of“fast cooling and fast heating”. In the forging system，different forging parameters have different effects on improving the performance of additive parts.

To address issues such as incompatibility caused by axis misalignment between rehabilitation robots’ exoskeletons and the human body， a configuration synthesis method for redundant rehabilitation exoskeletons was proposed. Firstly， the number of redundant motion pairs was determined through global static conditions， and geometric analysis was used to reduce the number of combinations of redundant motion pair types. Secondly， based on user comfort， possible redundant motion pair axis combinations were considered and appropriate combination modes were selected. Next， the possible position arrangement of redundant motion pairs was considered to determine the optimal position arrangement based on engineering constraints.Finally， the motion of redundant motion pairs was analyzed to determine the position and posture of the exoskeleton and human body. The results indicated that the optimal configuration can achieve the same position and posture between the exoskeleton and the human body under different axes misalignment conditions， effectively overcoming the problem of kinematic incompatibility.

An inspection robot tends to swing under wind load，which may reduce the accuracy and reliability of inspection results. The balancing mechanism of the inspection robot under wind load was designed with reference to the rotor aircraft. Firstly，the transfer function between the voltage and speed of the motor was derived by using the dynamic equation of the motor，and the relationship between the lift generated by the rotor and the rotor speed was established by using the blade element method，so as to establish the relationship between the input voltage and the output lift of the balancing mechanism.Secondly，the influence of wind load in different directions on the working state of the inspection robot was analyzed， and the swing mathematical model of the inspection robot under transverse wind load was established. Finally，the fuzzy PID （proportional integral derivative） was applied to the control of the balancing mechanism， and the numerical simulation and prototype experiment of the inspection robot were carried out. The results showed that the designed balancing mechanism can effectively restrain the swing of the inspection robot under wind load.

In order to solve the problem of poor surface quality consistency after the grinding of nickel-cobalt-based functional gradient materials （NiCo-FGMs） prepared by laser-directed energy deposition （L-DED） and forming process， the influence laws of the grinding process parameters on the grinding force and surface roughness of NiCo-FGMs were analyzed based on the orthogonal experiments， and the corresponding prediction model was established. For the different processing objectives of rough machining and finish machining， the multi-objective grinding process parameter optimization was carried out and verified by using the second-generation non-dominated sorting genetic algorithm （NSGA-Ⅱ） and the entropy weight technique for order preference by similarity to ideal solution （TOPSIS） method. The results showed that the rough machining parameters are used a_p = 53.61 μm， v_s = 29.99 m/s， v_w = 311.89 mm/min， and the finish machining parameters are used a_p = 14.96 μm， v_s = 29.99 m/s， v_w = 300.92 mm/min. After the two-stage machining， the standard deviation of surface roughness is reduced from 0.195 μm to 0.101 μm， which effectively improves the surface roughness consistency of NiCo-FGMs.

Taking the electrical discharge machining （EDM） of P-type monocrystalline silicon as the experimental subject， the surface morphology features and microscopic discharge pit characteristics of the monocrystalline silicon were investigated. It was revealed that the material removal mechanism of the monocrystalline silicon is dominated by stress erosion， which is more significant than melting and vaporization erosion. An experimental research on EDM of monocrystalline silicon microstructures， including micro-grooves and micro-holes， was done to explore the forming accuracy， size deviations， fluctuation height of the cross-sectional profiles， and surface quality of the EDM-processed microstructures， revealing the influence of the main processing parameters on the material removal rate， surface morphology， and surface roughness of the EDM-processed monocrystalline silicon.

In order to improve the machining accuracy of optical free-form surfaces， given the shortcomings of traditional machine tools in heat and force， a new five-axis machine tool topology is constructed， which can effectively reduce the influence of thermal deformation on the accuracy of machine tools， and a synchronous compensation method of gravity/inertial force error is proposed. The error compensation effect of different error compensation methods are compared and analyzed. The results show that the accuracy of milling machines can be greatly improved by using gravity balance combined with mechanical shunt column， and the error compensation effect of variable force （VF） is significantly better than that of constant force （CF）， the error compensation effect of two-dimensional motion gravity balance （TMGB） is significantly better than that of fixed gravity balance （FGB）. The worst compensation method is FGB combined with CF， while the best compensation method is TMGB combined with VF. The combination of FGB combined with VF and TMGB combined with CF have their own advantages and disadvantages under different conditions.

Polypropylene fibers are widely used in high-performance cement-based materials to reduce the high-temperature damage of structures. However， the calculation and prediction of thermal transfer parameters for polypropylene fiber reinforced concrete at high temperature still lack an effective model. Based on the homogenization method of composite materials， a refined multiscale calculation model was proposed to estimate the equivalent thermal conductivity of polypropylene fiber reinforced concrete at high temperature， which fully considered the influence of the volume change of polypropylene fiber at high temperature and the relaxation effect after fiber melting. As verified by experimental results， the proposed calculation model can accurately predict the thermal conductivity of polypropylene fiber reinforced concrete as a function of temperature. In addition， the thermal transfer properties and shape of polypropylene fiber have little effect on the equivalent thermal conductivity， while the fiber content is the main influence factor through model parameter sensitivity analysis. The analysis results provide a reference for the thermal design of fiber reinforced concrete.

In order to effectively prevent the occurrence of risk accidents during bridge construction and give prospective suggestions for engineering safety，a bridge construction risk assessment method based on cloud model improved analytic hierarchy process（AHP） was proposed. Firstly，the fishbone diagram was used to identify risk factors and construct a bridge construction risk assessment system. Secondly，considering the uncertainty of expert opinions in the process of assigning weights， the cloud model improved AHP is used to calculate the weights of each index. Finally，the two-dimensional cloud model is constructed to evaluate each risk factor from two dimensions： the probability of risk occurrence and the loss caused by risk. Combined with the index weights，the two-dimensional comprehensive cloud model was calculated to quantitatively describe the bridge construction risk level through the close degree. The above assessment method was applied to evaluate the construction risk of a bridge in Changchun. The result shows that the method has high feasibility and effectiveness，and can provide insights and references for the prevention and control of bridge construction risk.

Tunnel excavation has a detrimental impact on the deformation of existing station structures. Taking the close-proximity crossing an existing metro station in Shenyang’s gravel layer as an example， the finite difference method was used to investigate the deformation mechanisms and control techniques for such excavations. The support parameters for advanced grouting with small conduits and the method of unequal strength grouting reinforcement were optimized. Through simulations of three grouting schemes （uniform strength， lateral unequal strength， and spatial unequal strength）， the vertical displacement， horizontal displacement， and stress of the existing structure were analyzed. The results indicate that spatial unequal strength grouting is the most reasonable approach， with reductions in vertical displacement， horizontal displacement， and stress of 44.7%， 53.7%， and 15.6%， respectively. Additionally， the influencing factors of the spatial unequal strength grouting method were discussed and analyzed， which can provide a reference for tunnel excavation near existing stations.

According to the theory of social exchange， the longitudinal survey data of 445 employees for three months was analyzed to deeply understand the relationship among secure-base leadership， employee work engagement and employees’ proactive innovation behaviors. The results showed that secure-base leaders can actively promote employees’ proactive innovation behaviors， who can anticipate employee engagement for a certain period， and work engagement will also drive employees to actively innovate at a specific stage in the future. Similarly， employees’ proactive innovation behaviors will also affect secure-base leadership， and employees work engagement will promote the change of secure-base leadership styles at a specific stage in the future. These conclusions both expand the field of secure-base leadership and provide guidance for organizations to promote employee’s proactive innovation behaviors.

Taking Harbin open-air market as an example，a nonlinear hedonic price model and a quantile regression model were constructed to explore the capitalization effect of open-air markets. It was found that： the proximity of open-air markets has a negative effect on the prices of neighboring houses，and the higher-priced housings are more sensitive to the negative externality of open-air markets； there is a non-linear relationship between the proximity of open-air markets and housing prices，and the impact of open-air markets on housing prices will shift from inhibition to enhancement as the distance increases； shopping malls have a moderating effect on the capitalization effect of open-air markets，and this interaction exists only in the markets for moderately-priced housing. The findings of the study provide an important reference for the local governments to consider residents’ preferences in planning decisions and to rationally define the location of semi-“not in my backyard（NIMBY）” facilities such as open-air markets.