Search

An articulated pipe composed of a flexible pipe and a rigid pipe is a typical hybrid rigid-flexible dynamical system involving flow-induced vibrations. Based on Hamilton′s principle, the governing equations of motion of the hybrid rigid-flexible pipe system were established. The partial differential equations of motion were discretized via Galerkin′s approach using the modal functions of a cantilevered beam. Instabilities and critical flow velocities were analyzed and calculated as a function of the stiffness of the rotational spring, mass ratio, length ratio of the rigid and flexible pipes. The results show that the critical velocities decrease as the increasing of length ratio and the system mainly suffers second mode instability. The hybrid pipe undergoes complicated transference of “second mode instability-regain stability-third mode instability” at a certain value of stiffness when the flow velocity increases. With the increasing of mass ratio of flexible pipe, a transition from second mode instability to fourth mode instability is observed.

In order to ensure safety and efficient state of nuclear power plant in the long time operation, it is necessary to figure out the acceptance table of defects of the primary loop to meet the requirement of rapid assessment in engineering. It is necessary to do the fracture mechanics and plastic limit load analysis for the cracked pipe. Through numerical simulation analysis, the rationality of adopting plastic limit load as the critical state of austenitic stainless steel structure with good fracture toughness is demonstrated. Based on the criterion of non-decreasing the plastic limit load in design state, the acceptance standard of defects for the austenitic stainless steel pipe is established in this paper. The results show that the defect acceptance size of the primary circuit pipe is close to that in the ASME code, but much larger than that of the RSE-M code, and the RSE-M code is too conservative.

A statistical model of static friction factor of joint surfaces was deduced, which was more in line with the actual situation and takes into account the interaction of asperities and the deformation of the substrate. Based on the contact load model of joint surfaces considering the interaction of asperities and the deformation of the substrate, the statistical model of the static friction factor of joint surfaces considering the interaction of asperities and the deformation of the substrate was deduced according to the definition of the static friction coefficient of joint surfaces. The parameters of joint surfaces measured by the experiment are substituted into the model for simulation, and it is found that the static friction coefficient of joint surfaces considering the asperities interaction and the substrate deformation is larger than that without considering the asperities interaction and the substrate deformation. The influence of the asperities interaction and the substrate deformation decreases with the increase of the dimensionless normal contact load and decreases with the decrease of the dimensionless distance of joint surfaces, and increases with the decrease of the root mean square of the profile height. The simulation results of the model are compared with the experimental data of the static friction factor. The results show that the static friction coefficient of joint surfaces considering the asperities interaction and the substrate deformation is closer to the experimental data, which shows that the model is correct.

The complex and changeable working conditions of mechanical equipment in industrial field lead to uneven distribution of fault samples, which brings great trouble to traditional machine learning. In order to solve this problem, proposes a bearing fault transfer diagnosis method based on domain adaptive neural network and balanced dynamic distribution adaptive. Firstly, According to the characteristics of bearing vibration fault samples, the convolution layer of convolutional neural network is improved by wavelet transform, and the characteristics of bearing samples are extracted adaptively. Then, Maximum Mean Discrepancy measure and weight regularization are used to process the generated features in the loss function to reduce the difference in sample distribution and obtain the domain adaptive neural network model. Finally, A-distance is used to improve the equilibrium distribution adaptive to make it have dynamic characteristics, further improve the difference of sample distribution, and realize bearing transfer diagnosis by KNN classifier. Through experimental verification, the proposed method can accurately migrate the bearing fault state in the same bench rig cases and cross bench rig cases, proving that the method can effectively solve the problem of uneven distribution of unlabeled samples under variable working conditions, and has the effectiveness and robustness.

Aiming at the problem of failure strength analysis and prediction of bolted composite connection, the strong nonlinear mapping ability of deep learning neural network was used to non-linear fit the influence of different parameters on the failure load of composite bolting, and the influence weight of each parameter was allocated. A prediction model was constructed based on limited training samples to predict the peak failure load of bolted composite joints. Using finite element software, the data set of peak failure load of bolted laminates was calculated to construct the deep learning neural network. Through the test, it is determined that the development effect of deep learning model is the best when the number of hidden layers is two. The mean square error between the predicted value and the finite element simulation value is taken as the loss function, and the learning rate is set at 0.01. When the mean square error is the minimum, the training is stopped, and the best deep learning prediction model is obtained. The model is used to predict the maximum value of all the prediction results of peak load failure and the corresponding parameter combination, and compared with the simulation results of the same parameters, the difference between them is 1.4%. Compared with the prediction methods of finite element simulation and empirical formula fitting, the deep learning prediction method has obvious advantages in time efficiency.

Aiming at the mutation in demodulation of the envelope estimation function, a local mean decomposition method for singular interval envelope reconstruction is proposed. This method determines that the reason for the sudden change in the demodulation of the envelope estimation function is the intersection of the envelopes, and defines the local area where the intersection exists as a singular interval. Combined with the extreme value symmetry theory, the interpolation points in this interval are expanded, and the cubic Hermitian interpolation is used to perform local reconstruction. A singular interval envelope reconstruction algorithm is formed. The simulation signal and the application of fault diagnosis of reciprocating compressor bearing prove that the proposed method solves the envelope crossing problem, suppresses the demodulation mutation phenomenon, and the decomposition result has more obvious fault characteristics.

Based on Hamilton principle, the nonlinear magneto-elastic vibration equation of ferromagnetic plate is established under the combined action of mechanical load and harmonic magnetic force induced by magnetization. For the boundary conditions with two long simply supported edges, the transverse vibration differential equations after variables separation is obtained by Galerkin method. The multi-scale method and Lyapunov stability theory are used to solve the first-order principal resonance problem under electromagnetic excitation, the amplitude-frequency equation and the stability criterion of the steady-state solutions are obtained. Through numerical examples, the amplitude-frequency characteristic curves, the amplitude-magnetic field intensity curves and the amplitude-velocity variation curves of the ferromagnetic plate are obtained. The results show that the amplitude of stable solutions increases with the increase of magnetic field intensity amplitude. The nonlinear stiffness increases with the increase of the velocity, the hard-spring characteristics are enhanced and the nonlinear characteristics are more significant.

The new rigid-flexible coupling system for vehicle-beam expansion joints is proposed in the context of all-steel load-bearing radial tire by using the RecurDyn multi-body system dynamics simulation platform, the system is used to simulate the whole contact process of 1/4 car body through expansion joint, and the dynamic response of car body and beam body expansion joint is compared with the traditional simulation results. The results show that the simulation waveform of the new coupling system is consistent with the traditional method and the waveform is sensitive to the two influence factors in the coupling system, which verifies the good contact dynamic characteristics of the coupling system, the contact stability of the beam body can be improved by strengthening the connection between the side beam and the beam body and the end structure, the car with the new type of tire should not exceed 80 km/h when passing the seam. The coupling system provides a new and effective method for joint dynamic simulation of vehicle-beam expansion joint.

Focusing on investigating the brake squeal problem of a disc brake at 3 000 Hz. To this end, the noise of this frequency is reproduced on the bench, the laser Doppler vibration measurement technology is used to obtain the operational deflection shape（ODS） at this frequency. Finite element analysis（FEA） model is established and modified according to the experimental modal. The complex eigenvalue analysis is adopted to retrieve the complex eigenvalues and mode shapes.The correlation analysis is used to confirm that the complex eigenvalue results are valid for the actual noise ODS results.Afterwards, analysis of the components contribution and the components modal contribution of this unstable mode are used to provide corresponding structural optimization plan, and a bench test has been carried out to verify the effectiveness of this method.

Aiming at the low accuracy with unbalanced data sets in existing bearing fault diagnosis methods, we proposed a bearing fault diagnosis method based on importance weighted auto-encoder（IWAE） in unbalanced data sets. It was trained by minority samples, and the generated samples were added into original data sets to obtain balanced data sets. Then, deep learning method was used as diagnose network, and the balanced data sets were fed into it as input, so as to adaptively learn fault characteristics and realize fault classification. A large number of qualitative experiments showed that when the imbalance rate was 1∶7, the method could correctly classify the balanced samples, and the accuracy rate was 98.90%. Based on various imbalance ratios, the proposed method had better convergence and generalization than other existing models.

In order to study the strength performance of the airship envelope and the influence of overpressure bearing on its residual strength, the sample forms and experiments plans for the uniaxial and biaxial tensile of the envelope material were designed, and the tearing situation of the airship was analyzed with the help of a high-speed camera. Based on the experimental data, fitted the uniaxial tensile strength curve when the airship bears air pressure, and the longitudinal and latitudinal strength envelope line of the material when it was biaxially stretched. The experimental results demonstrate that there is a negative correlation between the residual strength of the envelope and the airship bearing air pressure when uniaxially stretched, the bearing strength of the envelope is comprehensively affected by the longitudinal and latitudinal strengths when biaxially stretched. This experiment also analyzes the failure situation of the airship envelope, which can provide an experimental guidance for its strength prediction.

A type of two-degree-of-freedom vibration system with bilateral nonlinear constraints is established. Through the fourth-order Runge-Kutta numerical algorithm, the dynamic characteristics of the p/1 periodic motion of the system under low frequency excitation, the law of mutual transition and the corresponding law of the coexistence zone of gap and period are analyzed. The cell mapping method is used to study the distribution law of different attractors and attracting domains in the coexistence area of periodic motion. The results show that the periodic motions of the system are mainly transferred through Grazing bifurcation and Saddle-node bifurcation. Due to the irreversible transition process, there is a coexistence zone of periodic motion between adjacent motions. As the gap increases, the range of the coexistence zone of periodic motion of the system gradually decreases.

In order to study the influence of load change on the meshing characteristics of cycloid pinwheel of RV reducer, considering the modification of cycloid tooth profile, the three-dimensional model of cycloid pinwheel mechanism was established by UG. The changes of meshing tooth pairs and meshing stress of cycloid pinwheel in a meshing transmission cycle were analyzed by using the transient dynamics module of Ansys, and the simulation was compared with the theoretical contact stress. Results show that the overall trend of contact stress between simulation and theory is consistent, the number of meshing teeth and meshing stress of cycloid pinwheel show a periodic law, and increase with the increase of load. The change trend of contact stress and equivalent stress is opposite to the number of meshing teeth, the more the number of meshing teeth, the smaller the meshing stress. The equivalent stress of the needle tooth in the length direction increases with the increase of load, and the stress has a trend of uniform distribution, but the change is small. The finite element analysis has certain reference significance for the design, analysis and manufacture of RV reducer.

To obtain the filling material with good compression mechanical properties, two kinds of orthogonal trapezoidal honeycomb aluminum with different structures were manufactured（90°orthogonal trapezoidal honeycomb aluminum and 45° orthogonal trapezoidal honeycomb aluminum）, the mechanical properties of the material under quasi-static axial loading was analyzed. Two kinds of orthogonal trapezoidal honeycomb aluminum were filled with polyurethane foam, and the effect of polyurethane filling on the orthogonal trapezoidal honeycomb aluminum was analyzed. The results show that compared with 45° orthogonal trapezoidal aluminum honeycomb, the performance of 90° orthogonal trapezoidal honeycomb aluminum is better, and the peak stress, plateau stress and unit volume absorption energy of the two kinds of orthogonal trapezoidal honeycomb aluminum are improved after polyurethane filling. However, the energy absorption efficiency is not significantly improved after being filled with polyurethane, and the strain corresponding to the maximum energy absorption efficiency decreases.

To reduce the raw material waste problems and repeated trial production in the matching process of rubber parts hardness and stiffness, the correlation analysis of rubber parts stiffness and hardness was carried out. Based on a rectangular rubber spring cushion, with uniaxial compression and tensile tests, the theoretical parameters of the Mooney-Rivlin model were calculated, and the results were verified by the least square method. The stiffness of the rectangular rubber spring pad increases nonlinearly with the increase of deformation, which Abaqus calculated. Based on the experimental data of rubber hardness, the exponential quadratic nonlinear function of hardness and Mooney-Rivlin model parameters was established. As a result, the hardness of rubber was positively correlated with the material parameters. The relationship between elastic modulus and hardness was transformed into elastic modulus and material parameters C2 and C1. Furthermore, the stiffness displacement curve of rubber material was obtained by simulation analysis to find the relationship between rubber material parameters and stiffness. Finally, with Mooney-Rivlin model parameters as the bridge, the correlation conclusion between stiffness and hardness of rubber parts was obtained.

In order to characterize the Ⅰ-Ⅱ mixed-mode fracture performance of 2D-C/SiC composites, Arcan butterfly specimens with unilateral cracks were used, and the mode-Ⅰ and mode-Ⅱ fracture shape factors of the specimen were calibrated through finite element model. The Arcan disc device was used to carry out mode-Ⅰ, mode-Ⅱ and Ⅰ-Ⅱ mixed-mode fracture tests upon 2D-C/SiC butterfly specimens, and the macroscopic and microscopic fracture modes and strength properties of the specimens under different loading paths were investigated. The results show that the fracture toughness of mode-II of 2D-C/SiC material is higher than that of mode-I. With increasing of the loading angle, the influence of shear stress on the crack propagation path increases, and meanwhile the fracture resistance of the specimens presents a tendency to increase. The applicability of several fracture criteria is demonstrated, and an Ⅰ-Ⅱ mixed-mode fracture criterion for 2D-C/SiC composites is suggested.

Definition based on the incremental theory of plasticity of J integral, combining the method of experiment and computer simulation, the method of crack propagation rate based on elastic-plastic mechanics is studied, the research shows that the crack growth rate curve obtained by the incremental plastic J integral is well described the material fracture mechanics performance characteristics, ΔJ integral as the crack growth driving force is verified the rationality of the parameters, in order to further research to explore mechanism of crack extension provides research foundation, a new model is proposed based on the 7N01 aluminum alloy test data, it takes into account the effect of stress ratio on crack propagation, and it is suitable for estimating fatigue life of materials in engineering practice.

In order to improve the stability of diagonal trot gait for quadruped robot, a foot end trajectory planning method based on Bezier curve is proposed. Aiming at the diagonal trot gait of quadruped robot, the kinematics model of parallel legs of quadruped robot is established, the forward and inverse kinematics are solved by geometric method, and the foot end trajectories of swing phase and support phase are optimized based on zero impact principle and PD controller. The comparative test is carried out by using Webots simulation platform and experimental prototype. The results show that when the proportional coefficient in X direction is 20, the differential coefficient is 0.75 s, the proportional coefficient in Y direction is 20 and the differential coefficient is 0.5 s, the foot end trajectory of quadruped robot is smooth without sudden change of speed and acceleration, and the gait of diagonal trot is more stable.

The grinding of spiral bevel gears is a nonlinear and dynamic complex process. It is significant to explore the complex grinding mechanism of spiral bevel gears by simulating the grinding process based on finite element method. A finite element model of composite abrasive grinding was established based on material constitutive equation of Johnson-Cook force-thermal coupling. And the changes in the tooth surface temperature field, grinding force and abrasive wear were compared and analyzed under different grinding speeds, grinding depths and abrasive spacing condition. The simulation results were also verified by the experimental results.The results show that the grinding depth is the most important factor to affect the grinding surface performance. The influence of the compound abrasive grain interference on the tooth surface temperature field and the resulting grinding vibration cannot be ignored. It could provide theoretical guidance for the research and practical processing of grinding surface performance of spiral bevel gears.

To quickly evaluate the safety of damaged bridge structures in the service process of cranes, a real-time prediction method for the fatigue life of bridge structures based on the optimal additive agent model is proposed. Based on the measured load data, the randomness of the relevant parameters of the service conditions is characterized by the combination of the working cycle process. In line with the most unfavorable service conditions, the stress weak position of the bridge structure is determined through finite element statics analysis and failure case statistics. And the structure finite element model is established for fatigue damage and life prediction. The MSC. Fatigue′s S-N module and crack propagation module are employed to simulate the residual life of the structure under random working conditions, and the influence of different crack depths on the residual life of the structure is analyzed. The Kriging model for the life prediction of the bridge structure that is optimized by the Pelican Optimization Algorithm and additive point criterion is created by the Latin Hypercube Sampling method and the black box theory. The real-time fatigue life prediction of bridge structures during trolley movement with load by Miner linear cumulative damage theory. The bridge structure of QD20/10 t × 43 m × 12 m general bridge crane is taken as an example to verify the feasibility of the proposed method. Comparing the proposed method with the traditional fatigue life calculation method based on finite element simulation, the results show that it can accurately assess the remaining life of the damaged bridge structure with fewer simulation calls and shorter simulation time, which can provide some reference significance for crane maintenance cycle formulation and scrapping decisions.

Since there are more systems to be observed on the sites and the observation time is shorter than the average life of the product, the life data actually collected is often heavily censored data. For such data, the classical parameter estimation methods can’t provide accurate and robust estimation. For this reason, a parameter estimation method based on the Weibull auxiliary model is proposed, and the numerical experiment results validate that this method can provide a more accurate and robust estimation of the gamma distribution than the classical parameter estimation method under the condition of heavily censored data.

The impact characteristics and residual compressive strength of the honeycomb sandwich structure under different impact energies and different boundary conditions were studied by testing and finite element analysis. The Vumat subroutine was used to establish a three-dimensional progressive failure criterion. The validity of the simulation model was verified by experiments. The results show that the simply supported specimens are more prone to damage and the peak force is lower, but the energy absorption effect is slightly better. At the same time, the residual compressive strength of the simply supported sample after impact is relatively low. When the impact energy is low, the boundary conditions have an important influence on the residual compressive strength. The influence of boundary conditions on the residual compressive strength can be ignored when the impact energy is large.

Under the bending load generated by automobile collision, the composite safety components should have the ability to resist deformation and dissipate crashworthiness energy. In order to determine the mechanical parameters of the progressive damage material model of a woven carbon fiber composite thin-walled beam in the three-point bending condition, the mechanical properties test of the woven carbon fiber composite material and the three-point bending experiment of square tubes made by this material were carried out, thus got the basic mechanical parameters, bending failure history and force-deformation characteristics. Then the effects of the non-test parameter in the progressive damage material model on the simulation results were analyzed and the effectiveness of the multilayer shell modeling method were discussed. By comparing with the experimental results, the non-experimental parameters and effective simulation modeling methods for the progressive damage model of composite materials were gotten. The obtained material properties can be used for the crashworthiness simulation analysis of this material structure in similar working conditions.

Gearbox is widely used in petrochemical, shipbuilding, electric power and construction machinery, etc., and its fault monitoring and diagnosis technology is always a hot issue. A gearbox fault diagnosis method based on DS evidence theory, which integrates oil and vibration with multiple parameters is proposed. Aiming at the problem that the traditional single vibration signal analysis method is not sensitive to gear wear in the early stage, radial basis function（RBF） neural network is used to establish the classification model of the fusion of vibration time domain features and oil features. Vibration frequency-domain features are extracted adaptively based on high-dimensional variational autoencoder and fault classification is completed. Different evidence bodies obtained by RBF neural network and high dimensional variational autoencoder are fused by weighted DS evidence theory to obtain the final diagnosis result. Compared with different fusion methods, the effectiveness of the proposed method in gear fault diagnosis is verified.

In view of the low accuracy of cutting force prediction caused by the difficulty of considering the tip arc radius, the main cutting edge arc radius and the tool working angle in the traditional turning force prediction model, an equivalent turning force model considering the tip arc, the main cutting edge arc radius and the working angle based on a small amount of turning force experimental data is established in this paper. The model is a turning force equivalent model of ideal turning tool with three nonlinear equations with equivalent principal deflection angle, equivalent edge inclination angle and equivalent normal forward angle as independent variables through the turning force experimental data and coordinate transformation matrix obtained by turning once. A turning force prediction model based on the equivalent model is established. Finally, the effectiveness of the prediction model is verified by turning force experiments.

In the process of plunge cylindrical grinding, it is difficult to directly measure the contact stiffness between the grinding wheel and the workpiece. In order to solve this problem, a simplified method for qualitatively determining the contact stiffness between the grinding wheel and the workpiece is proposed, and the grinding parameters are optimized. Based on the dynamic model of plunge cylindrical grinding process, the transfer function of grinding process is deduced, and the mathematical model of the time constant and contact stiffness is established. A simplified method for qualitative determination of grinding contact stiffness is proposed. Further, multiple tests are conducted by controlling the wheel feed speed and the workpiece speed. According to the experimental results, the variation of the system time constant and the grinding force with the grinding parameters are analyzed. The relationship model between the contact stiffness and the grinding parameters is established by regression analysis, and the optimal values of the wheel feed speed and the workpiece speed in the specific grinding process are determined under the condition of meeting the machining requirements.

When mechanical components are subjected to multiaxial cyclic loading during operation, the additional strengthening effect is easy to occur. Using uniaxial fatigue life calculation formula to predict their life will produce large errors. However, the existing multiaxial fatigue life prediction models are not convenient for engineering applications because it contains a large number of material constants. To solve the above problems, based on the initiation and promotion mechanism of fatigue microcracks under different stresses, combined with the shear damage model, a new expression of damage parameters is proposed, and the corresponding multiaxial fatigue life prediction model is derived combined with the Manson-Coffin-Basquin equation. The test data of two materials under the non-proportional loading and proportional loading path are used to verify the proposed life prediction model, and compared with the prediction results of the FS model. The results show that the proposed model has a good prediction effect. Finally, the model is applied to the fatigue life prediction of a big pendulum boom, and the results show that the boom meets the fatigue life requirements.

Active gear design is a method that can pre-control gear meshing performance directly. In order to study the influence of active gear design on the bearing characteristics of tooth surface, the relationship between active modification parameters of the face gear and bearing characteristics of tooth surface is studied. Firstly, the contact model of the gear pair is established by the finite element method, and then different groups of active modification parameters are designed for simulation analysis. The results show that the geometric transmission error greatly influences gear transmission stability and contact strength, and controlling its amplitude can control bearing transmission error amplitude. Pre-designed contact path inclination angle within 30° neighborhood, bearing contact characteristics are better. The contact ellipse length mainly affects the contact strength of tooth surface and bending strength of tooth root. According to the simulation results, when it takes 0.5～0.75 times of tooth width, face gear pair has better bearing contact characteristics.

The long-term continuous high-speed rotation of the rotor of the steam turbine generator set inevitably leads to the change of the rotor vibration state. The quality of the rotor vibration state directly affects the safe operation of the whole unit. Therefore, it is imperative to study the early warning method. Firstly, the running law of rotor vibration is analyzed. Secondly, based on the vibration operation law, the calculation method of rotor vibration abnormal threshold is proposed. Thirdly, based on the abnormal threshold and nonlinear state estimation technique（NSET）, the rotor vibration early warning method is proposed. Finally, it is tested and verified by the operation data of a power plant, and compared with the vibration early warning method based on back propagation（BP） neural network. The results show that the proposed method can accurately and effectively realize the rotor vibration early warning.

To solve the problems of complex calculation and weak applicability of the traditional interval estimation methods such as Weibull distribution with the different parameters, a simple interval estimation method suitable to the conditions of multi-parameters was proposed. Through the expression of Weibull distribution, the pivotal-variable obeying chi-square distribution was obtained and the degrees of freedom were modified. Combined with the point estimation results from maximum likelihood estimation and the empirical estimation shape parameter in the Weibull distribution, the interval estimation method of two parameters in Weibull distribution was established. The confidence of proposed interval estimation method was verified by Monte Carlo simulation, and the applicability of the method on different parameters was also analyzed. Furthermore, comparisons with the results computed from traditional least square and maximum likelihood estimation methods were carried out by using simulation. Simulation results indicated that the modified pivotal-variable method has simple calculation process and small deviation from predefined nominal confidence with the different parameters. Therefore, it can be concluded that proposed method executes the more effective estimation than the traditional methods.

With the increase of the load and running speed of heavy freight locomotives, the wheel wear becomes more and more serious, which can exacerbate the polygon formation of wheels. The influence of polygonal wheels on the dynamic performance of a wide-gauge six-axle heavy freight locomotive is examined, taking the ГОСТ standard as the main reference. The finite element model of the wheelset is established, and then a rigid-flexible coupling model is established by using the multi-body dynamics software. The polygonal wheel is described by the harmonic function. Effects of the parameters, unrounded amplitudes and polygon numbers are analyzed. The results show that the critical speed decreases because of the effect of the polygonal wheel. Within the design speed range of 120 km/h, the wheel polygon has little influence on the running comfort; the influence on the locomotive safety and the vehicle dynamic characteristics is closely related to the modal characteristics of the vehicle.

The mechanical properties of fiber reinforced composite aluminum honeycomb sandwich structure were studied by three-point bending test. The effects of different types of panels（carbon fiber panel, glass fiber panel, carbon fiber/glass fiber panel）, panel thickness and core aperture on failure modes, ultimate load and energy absorption were analyzed. The results show that the carbon fiber/aluminum honeycomb sandwich structure has stronger ultimate load and energy absorption than the other two structures. The thicker the panel, the smaller the core aperture, the stronger the ultimate load and energy absorption of the structure. The thickness of the panel has a great influence on the energy absorption, and the aperture of the core has a great influence on the ultimate load. The finite element simulation of carbon fiber/aluminum honeycomb sandwich structure and the comparative analysis of its failure and deformation process verify the validity of the model, and provide guidance and help for experimental design and analysis.

In order to improve the load-carrying capacity of the hydrodynamic bearing, and reduce the calorific value and friction coefficient at the same time, a multi-objective optimization model of the hydrodynamic bearing was established. Aiming at the shortcomings of the traditional optimization algorithm that the convergence speed is slow and it is easy to fall into the local optimal solution, an adaptive strategy assisted the golden jackal optimization algorithm is proposed to improve the exploration and exploration ability of the golden jackal algorithm. The performance of the adaptive strategy assisted the golden jackal optimization algorithm is verified by an example with constraints. The results show that the adaptive strategy assisted the golden jackal optimization algorithm has good convergence performance. The algorithm is used to solve the multi-objective optimization problem of hydrodynamic bearings. The optimization results show that the optimized structure has a great improvement in performance compared with the initial structure, the load capacity is increased by 12.257%, and the calorific value and friction coefficient are reduced respectively by 15.610% and 33.333%.

For improving the mobility and obstacle climbing ability of the bionic gecko, a bionic moving joint configuration in legged style with two degrees of freedom is designed and developed. Based on the principle of mechanism, the sucker, steering engine and planar linkage, etc. are organically integrated with gecko morphology, and 3D solid modeling of the joint components based on features is constructed, which realize the motion function simulation of the legged joint on the basis of profiling design. The correctness and feasibility of the transmission design are verified through the trajectory simulation of the large and small leg joints. At the same time, the swing angle（0～73°） and obstacle climbing height（＞30 mm） of the bionic joint are analyzed. The results of static analysis show that the bionic joint under different load-bearing environments（such as flat ground, hang upside down, vertical wall） can meet the design requirements of strength and stiffness. The experimental test of the bionic gecko shows that the legged joint has not only stable movement and reliable force transmission performance, but also good bionic mobility. This study provides an new ideas for the mobile technology exploration of bionic gecko.

Due to low adsorption efficiency of magnetic adsorption unit（MAU） of crawler wall-climbing robot, the parameters analysis and structural optimization design are carried out based on DOE, RSM and sensitivity analysis method. The functional relationship between the parameters of the MAU and the magnetic force is obtained by the finite element simulation model, DOE and RSM. The sensitivity of parameters of the MAU to the magnetic adsorption force is determined by the sensitivity analysis. The length, the width, the height of the permanent magnet, the thickness of the yoke and the distance between the permanent magnet and the wall are selected as the design variables. After the optimized design, the adsorption efficiency of the MAU is increased from 2.029 7 to 3.938 1.

Due to the complex material and shape of synchronous belt, it has complex mechanical behavior in normal operation. To solve this problem, propose a matrix model to calculate the tangential pressure of the meshing part between tooth and pulley. In the matrix model, use spring to replace the contact between synchronous belt and pulley, synchronous belt was discretized and separated into independent parts. In order to verify the model correctness, multi-body dynamics software RecurDyn was used to simulate the synchronous belt transmission system. The results show that the calculated results is closed to simulation results at the beginning and the end of the meshing part where the tangential pressure is larger, and there are some difference in the meshing part where the tangential pressure is smaller. Comparing with other methods, the matrix model has less computation, which solves the difficulty of calculating the tangential pressure of the belt, and provides a theoretical basis for the selection and design optimization of synchronous belt.

In order to improve the working reliability of the biomimetic robotic fish, put forward a gradient reliability robust optimization design method method which is applied to the biomimetic robotic fish.Taken the bionic killer whale as the research object, based on the flapping wing motion theory, the dynamic load of the bionic killer whale was obtained by using MATLAB.Based on the theory of reliability sensitivity design, robust design theory and the theory of performance degradation, established optimization evaluation function of bionic killer whale, obtained the influence of the bionic killer whale’s design variables to reliability sensitivity gradient.The improved quantum evolutionary algorithm was used to optimize the design, the optimal solution of caudal fin propulsion mechanism was obtained.The results show that the absolute value of sensitivity of the design variables is decreased, the reliability is improved and the structure is more robust. This paper combine the flapping wing movement theory, the reliability sensitivity design theory, the reliability-based robust design theory, the theory of performance degradation and improved quantum evolutionary algorithm were combined, put forward a gradient reliability robust optimization design method method which is applied to the biomimetic robotic fish, it provides theoretical method and data support for the reliability analysis and design of bionic robot fish.

In order to obtain the optimal vibration response of a seven-axle elevated platform vehicle with hydro-pneumatic suspension（HPS）, a hybrid coupling scheme of parallel connection ＆ contralateral cross-linked for HPS was designed. The dynamic model with 20 degrees of freedom which can comprehensively reflect vehicle coupling vibration including vertical-lateral-longitudinal-transverse was established. The vibration response characteristics of the vehicle and the influence of key parameters were analyzed by joint simulation. A single objective function contains two evaluation indices of vehicle ride comfort and stability was established through analytic hierarchy process（AHP） and range transformation standardization. The parameters optimization calculation of HPS was carried out. The model reference adaptive control, based on the principle of global vibration, applying Lyapunov design adaptive law, was carried on the HPS. The simulation results show that the evaluation index of vehicle vibration response is significantly improved after parameter optimization and superposition adaptive control.

Aiming at the life data information of multiple types of testing sources for the same product, a mixed base distribution constructed to achieve three-source data fusion, and at the same time, the overall distribution density function of the life data source was estimated. And the iterative formula of the maximum likelihood estimation of each unknown parameter in the parameter vectors of the mixed basis distribution density function was obtained based on the EM algorithm. At the same time, the estimated value of each parameter was given by using the sample weighted by Bootstrap method, the density function analysis of the mixed basis distribution was determined, and the product reliability analysis was completed. The calculation example verified the feasibility and effectiveness of this type of model.

Aiming at the dynamic mechanical system with multiple failure correlation, the copula function is used to describe the correlation of failure modes among units, and a reliability model of dynamic mechanical system with multiple failure correlations is proposed. The reliability problem of complex system is transformed into multiple two-dimensional pair copula functions by using R-Vine copula function. The optimal R-Vine structure is selected by using MST algorithm. The time-varying parameters in the model are described by nonparametric kernel density estimation（KDE）. The common series parallel system（hybrid system） in mechanical system is modeled and analyzed. Finally, the effectiveness of the model is verified by an example of spacecraft open and close shafting.

With the development of structural topology optimization methods, some scholars had proposed strategies to integrate existing topology optimization methods to improve the manufacturability of topology optimization result. Specifically, a topology optimization method combined the Solid Isotropic Material Penalty（SIMP） method and the Moving Morphable Components（MMC） method was proposed. Firstly, the characteristics of power function in SIMP method was used to quickly obtain the main path of force transfer. Secondly, according to the gray-scale elements generated by the SIMP method, an Otsu threshold algorithm was proposed to maximize response of the binary topological structure. On this basis, morphological thought was used to construct a component, whose boundary was the inflection point of the skeleton, then the MMC method was used to complete the optimization to finally achieve the precise control of structure size. By calculating the minimum compliance topology optimization problem of the 2 D and 3 D numerical examples, the optimization results of smooth boundary and precise size of structure are obtained, verifying the effectiveness of proposed method.

Foil hydrodynamic air bearing is a kind of self-acting hydrodynamic gas bearing which the foil is used for elastic support. Because of its superior dynamic performance, it is widely used in high-speed turbine machinery, high-precision machine tools, aviation, aerospace, national defense and other high-end equipment fields. The foil hydrodynamic air bearing uses the air as the lubricating medium, according to the principle of dynamic pressure lubrication, a certain relative velocity should be reached between the journal and the bearing, so that the dynamic pressure film formed can overcome the radial force of the rotor and make the journal and bearing separate from the mechanical contact, that is, the rotational speed should reach the takeoff velocity of the foil hydrodynamic air bearing. In order to reduce the mechanical contact time between journal and bearing, determine the acceleration time of the motor and accurately measure its takeoff velocity, which has important engineering significance. A method to determine the takeoff speed of foil hydrodynamic air bearing by evaluating the vibration characteristics is proposed.The analysed result shows that accords with the actual result, which verifies the accuracy of the theoretical calculation method and is of great significance to the engineering application of air bearing.

Aiming at the driving comfort problem of a sporty large-displacement motorcycle, an optimization scheme was proposed. Through the simulation analysis of the motorcycle on the C-level road, the pitch vibration and vertical vibration acceleration, which characterize the riding comfort of the motorcycle, were obtained, and the above two index peaks were used as the target response. Sensitivity analysis and approximate modeling of related parameters were carried out, and the radial basis function neural network approximate model was used for multi-objective optimization, based on the error analysis of the approximate model. The results show that after optimization, the actual vehicle verified that the pitch vibration acceleration peak was reduced by 17.21%, and the front and rear wheel vertical vibration acceleration peaks were reduced by 8.11% and 24.18%, respectively, which improved the ride comfort of the motorcycle and verified the effectiveness of the optimization scheme.

In order to investigate the impact resistance of the 2A12 aluminum alloy plate, blunt-nosed projectiles were launched with the impact velocity ranging from 120 m/s to 330 m/s to hit aluminum alloy target plates with a thickness of 6 mm, and the influence of the initial impact speed on the impact resistance of targets was analyzed. The ballistic limit and velocity curve of projectiles were obtained by fitting the data of the experiment. The results show that the energy absorption of the target first decreases sharply and then keeps stable with the increase of the initial velocity of the projectile. Meanwhile, the target plate mainly undergoes shear failure, and the structural deformation of the target plate decreases as the velocity of the projectile increases. Moreover, based on the ABAQUS/Explicit finite element software, the modified J-C constitutive relation and fracture criterion were used to calculate the projectile impact target model. And the failure mechanisms of targets were analyzed on the base of verifying the validity of the simulation model.It has been found that the failure mechanisms of targets are similar under different impact speeds, and the initial velocity of the projectile has little effect on the failure mechanism of target.

In mechanical systems, the gap between mechanical components causes nonlinear elastic collisions during their movement. Taking the two-degree of freedom vibro-impact system with gap as the research object, the mathematical model was established and the numerical simulation analysis was carried out in Matlab software. Based on this, the application of function generators, resistors, capacitors, operational amplifiers and oscilloscopes and other analog electronic components in Multisim software to establish the equivalent circuit model for simulation analysis. The results show that the simulation results of equivalent circuit are consistent with those of numerical simulation analysis.The difference is that the equivalent circuit simulation not only has faster and more efficient operation, but also can realize dynamic parameter adjustment and conversion analysis under different modes. The Equivalent circuit simulation provides a reference method for the dynamics research of nonlinear vibration system.

In order to obtain the stiffness-damping feature of cross-arm damping seat of special vehicles, the dynamic simulation model of cross-arm seat is established. The time-frequency responses on the level D road and different vehicle speed conditions（5 km/h, 15 km/h, 25 km/h and 35 km/h） are analyzed. With considering the weighted root-mean square（RMS） acceleration in criterion ISO 2631-1:1997 as an evaluation index, the ride comfort of the seat with different matching relationships of stiffness and damp are discussed. Furthermore, the real vehicle validation experiment is conducted. The results show that the seat vertical acceleration under different speeds has the non-stationary random characteristic. The seat vibration intensifies with the increase of the speed. Under different vehicle speed conditions, the first two model frequencies for the seat concentrates in 3 Hz～10 Hz, which includes the human sensitivity frequency ranges, inducing the poor ride comfort performance. Additionally, the simulation results agree with the experimental data, which verifies the correctness of the dynamic simulation model. The research has certain reference for multi-objective matching optimization design about the stiffness-damping parameters for the seat comfort of special all-terrain vehicles and engineering applications.

In order to solve the reliability evaluation of small samples for the high reliability, long-life products and to improve the accuracy of reliability evaluation, the Bayesian reliability evaluation model of multiple information source data fusion was carried out. First, a set of data as the initial prior information of another information source in this model was selected, by using the posterior density according to Bayesian estimation. Then, it as the prior distribution of the next information source was used. Through iteration, the joint posterior density and reliability index data of multi-source heterogeneous data are obtained, and the corresponding reliability analysis is completed. Finally, a simulation example verifies the rationality and effectiveness of the model.

Some characteristics of bearing initial signals are considered including unobvious features, susceptibility to noise interference and strong nonlinearity when a bearing was damaged. Based on the fractal box dimension, an improved variational mode decomposition for nonlinear features extraction of bearing fault signals（IVMD-NFE） is proposed. Because of the multi-measurement of nonlinear signals, the multifractal detrended fluctuation analysis（MF-DFA） method is used to study the multifractal characteristics of each fault signal. Taking the experimental data of rolling bearing as the research object, IVMD-NFE and MF-DFA method were used to analyze and diagnose the initial bearing signal. The results show that the signal extracted by the IVMD-NFE method can filter out noise to a greater extent and has lower fractal box dimension, and the extracted nonlinear characteristics are more representative; The fault signals of bearing exhibit multi-fractal features, the outer ring fault has the largest singularity index and strongest nonlinearity, and the cage fault is the smallest and weakest nonlinearity, indicating that the complexity of the data can better reflect the running state of the bearing. However, the method of using VMD or directly processing the original signal fails to extract effective nonlinear features, resulting in failure to distinguish faults.

Multi-scale fuzzy entropy can well measure the complexity of the vibration signal, but it lacks the effective use of other channel information. To make full use of the vibration information of other channels, the multivariate sample entropy theory that characterizes the multivariate complexity of synchronized multi-channel data is applied to the bearing fault diagnosis. To accurately extract fault features of bearing signals, a bearing multi-fault diagnosis method based on complete ensemble empirical mode decomposition with adaptive noise（CEEMDAN） and refined composite generalized multivariate multiscale fuzzy entropy（RCGmvMFE） is proposed. First, CEEMDAN is used to decompose multi-channel raw signals to obtain IMF without mode mixing. Then the correlation analysis method is used to screen the IMF components, and the IMF sensitive to the fault characteristics is selected as the multi-channel data to constitute the multivariate variable, and the RCGmvMFE is calculated to constitute the fault feature. Then, t-distributed stochastic neighbor embedding（t-SNE） is used to reduce the dimensionality of high-dimensional features. Finally, the whale optimization algorithm（WOA）is used to optimize the kernel extreme learning machine（WOA-KELM） so as to classify the low-dimensional fault features. Experimental results show that this method can effectively diagnose different fault severity of bearings, and provides a supplementary method for fault diagnosis of rolling bearings.