Addressing the issue of autonomous vehicles’ instability on icy and snowy roads， an improved rapidly-exploring random tree （RRT） path planning algorithm is proposed. Firstly， a dynamic model introducing road adhesion coefficient on icy and snowy roads is established. Secondly， the global target deflection sampling combined with the front pointing and steering angle of the vehicle， combined with the collision avoidance detection and the maximum curvature constraint under the velocity-adhesion coefficient， is used to improve the traditional RRT algorithm problem.Finally， a double quintic polynomial is used for path smoothing to ensure stability， brake constraints， and comfort. The performance of the improved algorithm RRT is compared with that of the traditional algorithm under multi-scenario conditions through the joint simulation of MATLAB-Simulink and CarSim. The experiments show that the improved RRT algorithm significantly improves the path smoothness， reduces the curvature mutation， has short time， high success rate and good stability when driving on ice and snow.

In application scenarios like robot control and autonomous navigation of intelligent vehicle， path planning needs to account for factors including obstacles and terrain. To address the issues of directionless expansion target and low efficiency in rapidly-exploring random tree （RRT） algorithm in path planning， a particle swarm optimization for probabilistically homogeneous rapidly-exploring random tree （PSO-PH-RRT*） algorithm is proposed. This algorithm base on the probabilistically homogeneous rapidly-exploring random tree （PH-RRT*） algorithm by using the particle swarm optimization algorithm to update the probability of direction as the velocity direction for random tree nodes， thereby improving the node position update strategy. It also uses the distance between the node and the target vector， along with trajectory smoothness， as the fitness function in the particle swarm optimization algorithm. Finally， simulations across various scenarios demonstrate that the PSO-PH-RRT* algorithm can significantly reduce iteration time costs while improving path length and smoothness.

In fault diagnosis， the traditional time-frequency domain methods are easily affected by subjective factors while being used for feature extraction， so that the redundancy emerges. Deep learning algorithm is highly dependent on training data and has computation complexity. Fault diagnosis method for rolling bearings based on wavelet packet-thresholdless recurrence plot （WP-TRP）is proposed by combining time with frequency domains. Firstly， the decreasing information entropy criterion is developed to overcome the subjectivity of wavelet packet decomposition for acquisition of more accurate time-frequency feature. On this basis， the idea of thresholdless recurrence plot is introduced to extract the initial time domain feature. Moreover， by adopting the singular value decomposition to decrease the redundant feature， the computational efficiency can be increased. Secondly， the marine predator algorithm is introduced to obtain the optimal parameters of supporting vector machine， by which the more accurate classification can be realized. Finally， the effectiveness of the presented method is verified by using the simulation on the benchmark rolling bearing datasets.

In the semiconductor fabrication equipment， the coatings on aluminum alloy often fail due to the coupling effect of high-temperature， vacuum and aggressive gases， including their plasma. In the chlorine-based plasma， the anodized coating has a high etching rate， leading to rapid removal， while the etching rate of Y₂O₃ coatings is approximately one in 50 of that of the anodized coating. In the fluorine-based plasma， both the anodized coating and Y₂O₃ coatings experience particle contamination due to the fluoride layer peeling. The corrosion resistance of the anodized coating can be significantly enhanced by adjusting the composition and temperature of the electrolyte or depositing a pure aluminum layer on the aluminum alloy surface. Additionally， improving the density of Y₂O₃ coatings can reduce their etching rate. Combining these strategies with remote plasma cleaning techniques can minimize the impact of charged particles on chamber materials， significantly reducing particle contamination in the reaction chamber. During the etching and thin film deposition processes， changes in the chamber surface composition can alter the plasma state， leading to various process defects.

During the rolling process of tailor rolled blank（TRB） cold-rolled strips， the gauge exhibits characteristics of periodic， continuous， and multi-stage variations. To address the difficulty in achieving precise gauge control， a control strategy of dense sampling and point by point regulation is proposed. To prevent oscillation in the control system， a variable gauge control（VGC） model based on time and space constraints is developed and successfully implemented in the control system of a variable gauge cold rolling mill. On-site data analysis shows that prior to the use of VGC model， the gauge fluctuations in the equal gauge zone are relatively stable， while the transitional zones exhibited significant gauge deviations， with positive deviations during upwards rolling and negative deviations during downwards rolling. After using VGC model， the system demonstrates rapid and effective adjustment capabilities for gauge deviations in the transitional zones， ensuring that the gauge deviations of TRB are within ±0.025 mm， better than the existing TRB standards in the automotive industry.

Thermogravimetric analysis was carried out at different heating rates to determine the pyrolysis behavior and kinetic parameters of bischofite. Based on the differential thermogravimetry （DTG） and thermogravimetric analysis （TGA） curves， the bischofite pyrolysis was divided into three main mass-loss processes， namely， stage Ⅱ （73.92~278.14 ℃）， stage Ⅲ （278.14~540.27 ℃） and stage Ⅳ （540.27~849.52 ℃）. The multi-stage reaction separation （K-K ） method was utilized to further divide the overlapping DTG curves. The pyrolysis products were analyzed by XRD and Fourier transform infrared （FTIR） spectroscopy， and the pyrolysis mechanism of bischofite was clarified that the dehydration is triggered by the breakage of the Mg—O—H bond， while the enhancement of the Mg—O bond prompts the hydrolysis reaction and hinders the dehydration. Finally， the activation energy of the sub-reaction was determined by Friedman’s method， and the pyrolysis kinetic model was constructed by combining the customized reaction function with the kinetic compensation effect.

To complete the fault characteristic extraction of blade cracks， a finite element model of a cracked blade incorporating breathing effect was first established based on the Mindlin-Reissner shell element. The dynamic response of the cracked blade under the combined action of centrifugal and aerodynamic loads was solved， providing excitation signal input for fault feature extraction. Then， fault indicators based on the nonlinear output frequency response function and energy indicators were established. Finally， the effectiveness of various indicators in extracting the fault characteristics of rotating blade cracks was analyzed. The results showed that the contribution rate indicator F_e（n） and weighted contribution rate indicator R_n（n） are unstable and insensitive in diagnosing blade crack faults， whereas the energy indicator effectively extracts blade crack fault characteristics under both resonant and non-resonant state. These results provide engineering guidance for feature extraction， analysis， and indicator selection of rotating blade crack faults.

Based on the gyroid three period minimum surface （TPMS）， a sine function square and cosine function square density gradient skin lattice structure was proposed to address the problem of a relatively single gradient variation law in gradient lattice structures in current research. Titanium alloy （Ti-6Al-4V） sample was prepared using selective laser melting （SLM） technology. Finite element analysis and compression experiments were conducted on the sample to study its mechanical properties， and the error between the two results was less than 6.3%. Compared with the traditional linear density gradient skin lattice structures， the mechanical properties， deformation behavior， and energy absorption characteristics of the three structures were compared and analyzed. It was found that among the three structures， the cosine function square skin lattice structure had the best load-bearing performance， while the sine function square skin lattice structure had the strongest energy absorption capacity. By optimizing the internal relative density of the skin lattice structure， its mechanical properties and energy absorption capacity can be improved.

Research is conducted on the ball end milling process with the aim of achieving reliable optimization of milling process parameters. Firstly， according to the motion trajectory of the ball end mill cutting edge， the surface morphology formed during machining is simulated using the Z-mapping （Z-MAP） algorithm， and the surface roughness （R_a） is introduced to measure the surface quality after machining. The accuracy of the surface morphology simulation model is validated through surface morphology analysis experiments. Then， considering the actual constraint conditions of the machining surface quality， the tool service life， and the uncertainty of process parameters during the machining process， a reliability optimization model for process parameters is established with spindle rotation speed， tool feed rate， axial cutting depth and radial cutting depth as the optimization variables， and maximizing the material removal rate （Q） as the optimization objective. Finally， the optimization model is solved using the grey wolf optimization algorithm to obtain the optimal process parameters， and the feasibility of the optimized results is verified through milling experiments.

Aiming at the characteristics of Inconel 718 material such as high work hardening rate and large cutting temperature variation， the machining mechanism and modeling of Inconel 718 were deeply studied by taking the discharge machining process of wire electrical discharge machining as the research object. The temperature field of single-pulse discharge is analyzed by the finite difference method and finite element simulation， and the theoretical and simulation temperature distribution results under given parameters are obtained. Furthermore， the law of the influence of pulse width on the size and shape of the corrosion pit is further explored. On the basis of considering the influence of the recast layer on the size of the pit， the surface roughness and material removal rate of machining are predicted and compared with the experimental data. The results show that with the change of pulse width， the variation trend of the theoretical and simulated electric pit dimensions is consistent. The maximum error between theoretical and simulation data and experimental results is 9.88%.

In order to explore the effect of temperature and load conditions on the retardation force of stator vanes’ regulating mechanism from an experimental level， a single-level regulating mechanism test system is built considering the design criteria of the mechanism experiment based on motion functions and load simulation at the schematic-level. A method for testing the force of the operation mechanism using a unidirectional tension force sensor is proposed. Tests are conducted on the strain of the pull rod under different axial force loads and environmental temperatures， as well as the force on the push rod. The results show that under individual effects of axial force load and temperature， the change in retardation force is closely related to axial force load， and the effect of axial force load on retardation force is larger than that of temperature； when axial force is combined with temperature， the effect of retardation force shows a coupled trend， and the influence of temperature on retardation force cannot be ignored during the mechanism’s return motion.

Hybrid off-road vehicles include multiple operating modes， and the inappropriate control strategies will impact the smoothness and clutch life when vehicles switch operating modes. Taking the power split hybrid gearbox as the research object， the transformation relationship of contradictory features is established using the extensible control algorithm. The contradictory feature quantities are transformed by adjusting the weight matrix of model predictive control （MPC） to interfere with the extensible controller. The simulation model of the transmission system of off-road vehicles is established， and the results show that the jerk of the vehicle in the mode switching process using the extensible controller is less than 10 m/s³， which is about 20% lower compared with MPC. The friction work of clutches is about 18 kJ， and the jerk and sliding work effect are improved compared with the traditional MPC. Extensible control improves the smoothness of the mode switching process and enhances the quality of mode switching for off-road vehicles.

In order to meet the bearing capacity and deformation requirements of the joint between the prefabricated steel reinforced concrete frame column and the steel plate shear wall （SPSW）， the new flange-bolt connection mode were designed and applied between the prefabricated steel reinforced concrete column and the steel plate shear wall. The flange-bolt connection mode was established the finite element model of the prefabricated and cast-in-place joints respectively， and the bearing performance of the two joints under the same load condition was compared. The effects of concrete strength grade， strength grade and thickness of built-in steel plate， thickness of flange plate， axial compression ratio and number of bolts on the bearing capacity and deformation capacity of the connected joint were analyzed. The results show that the mechanical properties of the prefabricated joints are similar to those of the cast-in-place joints. Increasing the concrete strength grade， the number of bolts and the thickness of the flange plate can improve the bearing capacity of the joints， while changing the strength grade of the built-in steel plate has little effect on the mechanical properties of the joints. It is recommended to use 34 bolts， 20 mm thick built-in steel plate， 15 mm thick flange， and the axial compression ratio is controlled at about 0.7.

There are few studies on the correlation between the weighting and counting efficiencies of filter media of bag dust collector， which leads to the fact that its filtration performance cannot be accurately evaluated. Two factors that affect the weighting and counting efficiencies of the filter media were analyzed through building a filtration performance test platform. The results show that the filtration efficiency decreases with the increase of filtration velocity， and the filtration efficiency of low-efficiency filter media decreases more than that of high-efficiency filter media. Compared with alumina powder， the filtration efficiency of filter media under atmospheric dust source environment is more easily affected by the filtration velocity. The weighting and counting efficiencies of low-efficiency filter media under the two kinds of dust sources environment are quite different. The filtration efficiency of the filter media under alumina powder environment is significantly higher than its filtration efficiency under atmospheric dust source environment. The dust source has little effect on the filtration efficiency of high-efficiency filter media. It is more reasonable to use both the weighting and counting efficiencies to characterize the filtration performance of filter media.

The stability evaluation of rock mass relies on reasonable rock discontinuities grouping. However， traditional methods are susceptible to boundary and outlier points. To address this issue， an improved density peak clustering algorithm was proposed for rock discontinuities grouping. Firstly， the rock discontinuities orientations were converted into spatial coordinates， and the squared sine of the angle between unit normal vectors was used as a similarity metric. Then， an objective function was constructed based on validity evaluation indices， and the cutoff distance was optimized using the crow algorithm to obtain the optimal grouping results. Validation with simulated datasets demonstrates that the proposed algorithm effectively reduces human intervention， avoids interference from outliers， and ensures more reliable and reasonable clustering outcomes. The results show that the proposed method not only maintains good consistency with traditional methods but also exhibits higher applicability， providing a reliable reference for dominant joint grouping in engineering applications.

In order to further optimize the layout of prefabricated vertical drains （PVD） and explore the influence of PVD parameters on the vacuum consolidation effect， based on the traditional PVD theory， a small-width， small-spacing PVD-vacuum preloading method was proposed. This method adjusted the width and spacing of the PVDs appropriately to improve the vacuum consolidation efficiency， while ensuring that the amount of PVD material remained the same value. The ratio of the time impact factor （T_h） and superimposed influencing factor （F） in vaccum preloading was defined as the characterization parameter for consolidation analysis. The changes in drainage volume， surface settlement， and other indicators during the vacuum consolidation process under different PVD parameters were compared. The results show that the consolidation efficiency of the soil increases as the T_h/F increases. Using PVDs with a width of 25 mm and a spacing of 17 mm can effectively improve the consolidation efficiency， increasing the total drainage volume by approximately 6.9% compared to the traditional layout. In addition， the use of small-width， small-spacing PVDs can effectively alleviate the attenuation of shear strength along depth and radial direction. Therefore， it can conclude that with a fixed material consumption， the use of small-width， small-spacing PVDs combined with vacuum preloading method has certain practical significance.

By using the latent profile analysis method， based on 564 questionnaire survey data， the types of paternalistic leadership in enterprises were divided， and further exploration was conducted on the impact of different types of paternalistic leadership on differences in employee work behavior， including both in-role behaviors and extra-role behaviors. It was shown that the paternalistic leadership in enterprises can be divided into four types， which are named high authoritarian leadership， enlightened leadership， low authoritarian leadership and balanced leadership. Among them， the level of employees’ in-role behaviors and extra-role behaviors under the influence of enlightened leadership is the highest， followed by the low authoritarian leadership. There is no significant difference in the level of employees’ in-role behaviors under the balanced leadership and high authoritarian leadership is higher. However， the level of employees’ extra-role behaviors under the balanced leadership. This study not only reveals the heterogeneity of paternalistic leadership， enriches the theoretical system of paternalistic leadership， but also provides the beneficial guidance for enterprises management practices.