Industrial dust collector is an important device for industrial safety production and green manufacturing. How to improve dust removal efficiency and reduce energy consumption is the development direction of future products. For the core part of the filter silo of the dust collector,three factors and two levels of the filter cartridge installation position A,the bottom funnel structure B and the spoiler length C were selected to carry out the full-factor flow field simulation optimization test. The total filtration flow and the comprehensive flow uneven coefficient were taken as the evaluation index,and the optimal combination of the test data was obtained by range analysis. The simulation results show that compared with the prototype,the improved cartridge dust collector can increase the total filtration flow by 4.21 %,and reduce the comprehensive flow uneven coefficient by 75.75 %,and solve the problem that the filter cartridge located at the air inlet is vulnerable to airflow erosion and secondary dust in the filter silo. According to the numerical simulation results,an improved cartridge dust collector was produced,and the error of the actual total filtration flow of the improved cartridge dust collector was 0.54 %. The numerical simulation optimization method can effectively guide the optimization design of cartridge dust collector,and is conducive to extending the service life of the filter cartridge and improving the product quality.

Riveting is an important connection method for aircraft structural assembly,and most structural failures occur at the connection holes.Therefore,it is very important to comprehensively evaluate the quality of riveted joints.Taking single riveted lap joints as the research object,and through finite element simulation analysis and tensile shear experiments,the residual stress distribution and shear properties around the holes of riveted joints were studed under different combinations of squeeze force and hole diameter riveting process parameters.The finite element analysis results show that increasing the squeeze force can significantly expand the longitudinal residual compressive stress distribution area around the hole,and make the maximum longitudinal residual tensile stress occur far from the edge of the hole,while the influence of the hole diameter on the longitudinal residual compressive stress distribution area was only significant;the tensile shear test shows that the failure mode of the single riveted lap joint structure was the instantaneous fracture of the rivet along the shear surface.The shear properties of the riveted joint formed with a 21 kN squeeze force at a hole diameter of 4.20 mm was the best,with an average peak load of 4.09 kN,which is about 9.36 % higher than that of the riveted joint formed with a combination of hole diameter of 4.04 mm and 15 kN squeeze force parameters. Increasing the hole diameter can improve the shear properties of single riveted lap joints,but it will reduce the longitudinal residual compressive stress distribution area around the hole,which will have an adverse impact on the fatigue life of riveted joints. Increasing the squeeze force appropriately can not only expand the distribution area of residual compressive stress but also improve the shear properties of the joint.

Based on AdvantEdge FEM software,the finite element numerical simulation model of 45 steel drilling process was established. A set of simulation data of torque,drilling force,drilling temperature and chip morphology when drilling 45 steel workpiece with standard twist drill was obtained through single factor simulation experiment. The effects of five parameters of 45 steel J-C constitutive model (yield strength,hardening modulus,strain rate sensitivity,thermal sensitivity and strain-hardening exponent) on the numerical simulation results of drilling process were studied. The results show that the influence of five parameters of J-C constitutive model on the numerical simulation results of drilling process of 45 steel has obvious regularity.The torque,drilling force and drilling temperature are greatly affected by yield strength,hardening modulus and thermal sensitivity. The variation of chip crimp radius is mainly affected by yield strength,thermal sensitivity and strain-hardening exponent,and has little relationship with parameters hardening modulus and strain rate sensitivity. The results have a certain reference value for adjusting and optimizing the parameters of workpiece material J-C constitutive model used for finite element numerical simulation of drilling process and obtaining accurate numerical simulation results of drilling process.

In order to achieve integrated optimization of process planning and workshop scheduling with the goal of minimizing completion time, a novel coding Genetic Algorithm (GA) based integrated optimization method was proposed. The Integrated Process Planning and Scheduling optimization (IPPS) problem was described and an integrated optimization model with minimizing completion time was established; a chromosome encoding method with maximum flexibility space was designed to ensure the maximum flexibility of integrated optimization problems from an encoding perspective; crossover and mutation method based on specific constraints of IPPS problem was improved, ensuring feasible solutions before and after genetic operation and making algorithm iterations effective; furthermore, a novel encoding genetic algorithm based IPPS problem solving process was developed. According to the verification of Kim example, compared with the existing advanced algorithms such as Two-stage Hybrid Algorithm (THA), Enhanced Ant Colony Algorithm (EACA), Hybrid Genetic Algorithm (HGA), completion time optimized by the novel coding GA algorithm is the smallest (i.e. 343, 344, 372, 320, 427, 432 min respectively) in the case of small-scale and large-scale production. The experimental results verify that the novel coding GA algorithm in solving IPPS problems is feasibility and progressiveness.

With the improvement of intelligent requirements in aerospace manufacturing field,using industrial robots to achieve flexible auxiliary assembly of large components is of great significance for realizing intelligent manufacturing of aerospace vehicles. However,limited by robots' own performance,it is difficult to compensate for physically mechanical coordination during the high-precision assembly of large components. Therefore,a passive compliant device with variable stiffness based on magnetorheological elastomers was designed at the end of the robot,after conducting simulation and experimental test,the results show that increasing the input current may enhance the stiffness of the device,and the displacement difference of the device under 1 A current difference is roughly 20 %,which can adapted to different processes of the flexible auxiliary assembly.

Aiming at the problems of slow search speed,low precision of optimization and easy to get into local optimality of chameleon swarm algorithm in solving mobile robot path planning,a fractional order chameleon swarm algorithm was proposed. Firstly,Tent chaotic mapping was used during initialization to enrich the population diversity and to improve the global search ability of the algorithm.Secondly,fractional order was added to change the tongue speed updating formula of chameleon when attacking prey to avoid the algorithm falling into local optimal while accelerating the convergence rate of the algorithm. Finally,cubic B-spline curve was used to smooth the path to improve the smoothness of the actual path of the mobile robot. The simulation results showed that the improved chameleon swarm algorithm converged faster than other algorithms,and could avoid obstacles and find the optimal path.

A Multi-Degree-Of-Freedom (Multi-DOF) soft exoskeletal hand rehabilitation actuator was proposed to address the limitation of single-degree-of-freedom in existing soft exoskeletal hand rehabilitation devices. Serrated bellows chambers were strategically placed at three joints,and bidirectional square chambers were incorporated at the palm and finger joints to facilitate finger flexion-extension and adduction-abduction. The nonlinear relationship between the bending angle of the actuator chambers and input air pressure was established based on the segmented constant curvature assumption model. Finite element analysis investigated the impact of chamber wall thickness,length,spacing,and quantity on the actuator′s bending performance. A silicone molding process was employed for actuator fabrication,and an experimental platform was constructed for bending performance testing.Results demonstrate that at 60 kPa,the actuator achieves a 21.8° extension,and at 25 kPa,a 254.8° flexion,aligning with the natural range of human finger motion.

To solve the low efficiency and insufficient dynamics of cooperative planning of multi-warehouse robots in current intelligent warehouses,a path planning method combining Improved Snake Optimizer(ISO) and Time Window(TW) model was proposed. Firstly,in the static planning stage,the improved snake optimizer considering the blocking factor was used to plan the global optimal path for multi-warehouse robots,and the time window of each robot was calculated,so as to improve the planning efficiency of warehouse robots. Secondly,in the dynamic planning stage,the time window model of multi-warehouse robots was established and dynamic adjustment strategy was introduced to resolve conflicts and improve the dynamic performance of the algorithm. Finally,simulation and experiments were conducted,the simulation results showed that the ISO-TW algorithm could reduce the number of blocking points by up to 27~43 compared to other algorithms with a reduction of 22.50 %~31.62 % (with a robot size of 80). The proportion of replanned paths could be reduced by 39.82 %~41.05 % (with a robot scale of 40). The average running time could be shortened by up to 38~40 s with a reduction of approximately 18.27 %~19.05 % (with a scale of 60 robots). Experimental results showed that the ISO-TW algorithm reduces the average running time by 8.53 %~9.23 % and the number of conflicts by 11.63 %~15.56 % compared to other algorithms. It could achieve efficient collaborative planning of multi-warehouse robots in real scenarios.

Aiming at the problems of multiple types and numbers of parts in aerospace compartments and low assembly efficiency,an automatic gripping and positioning technology for multi-species part structures is proposed. The technology is based on the part assembly feature recognition method,part flexible clamping end and part position attitude accurate adjustment technology,which can be used for automatic robot gripping and high-precision fast positioning of bracket parts. By measuring the part positioning dimensions of a typical product assembly,the results show that the axial and circumferential positioning deviations of the parts are within the tolerance range,indicating that the automatic clamping and positioning technology for the part can provide new technological means and equipment for the digital positioning, assembly and manufacturing of composite rotary body segments.

Complex product systems are characterized by large scale and complex structure, the key nodes identification of complex product systems from the perspective of system structure is of great significance for locating potential risk sources in advance. In view of the current problems that it is difficult to ensure the reliable operation of complex product systems and forward-looking risk management by tracing risk sources ″afterwards″, a key node identification method for ″before the fact″ risk management is proposed considering the hierarchical structure of complex product systems and the interaction of risks among subsystems. First, a complex network is established based on the system structure. Secondly, the null model is constructed and combined with the network statistical characteristics to initially divide the set of nodes, based on which a linear threshold model is introduced to analyze the risk propagation influence of nodes, and key nodes are selected according to the dynamic risk propagation. Finally, the system structure of the Xinzhou 700 aircraft is used for empirical analysis. The result shows that the proposed method can effectively identify the key nodes of complex product systems.

A fuzzy variable weight factor adaptive control system for vehicle semi-active suspension was proposed to address the issue of fixed and immutable weight factors in traditional suspension control strategies. The control system consisted of three fuzzy control loops,and the three fuzzy controllers correspond to three suspension response parameters that affected the vehicle′s dynamic performance. The α fuzzy controller,β fuzzy controller and γ fuzzy controller controlled the weight factors of the sprung mass acceleration function,dynamic load function,and dynamic deflection function based on road conditions. The control rules and functional relationships of input and output variables for each fuzzy controller were established. The fuzzy variable weight factor adaptive control system of vehicle semi-active suspension was to increase the weight of sprung mass acceleration control on high-grade roads represented by A and B levels to improve vehicle comfort performance,increased the weight of dynamic deflection control on middle-grade roads represented by C levels to improve vehicle comprehensive performance,and increased the weight of dynamic load control on low-grade roads represented by D levels to improve vehicle safety performance.The simulation results of bump road excitation showed that compared to passive control and fixed weight factor control,the peak the sprung mass acceleration under fuzzy variable weight factor adaptive control had decreased by 40.2 % and 28.6 %,the peak dynamic deflection had decreased by 41.5 % and 21.0 %,and the peak dynamic load had decreased by 40.1 % and 31.8 %. The attenuation speed of each response had hardly decreased,and the attenuation time remained basically consistent.There were no obvious oscillation phenomenon during the attenuation process. The simulation results of random road excitation showed that compared to passive control and fixed weight factor control,the root mean square values of sprung mass acceleration under fuzzy variable weight factor control had been reduced by 48.8 % and 23.5 %,the root mean square values of dynamic deflection had been reduced by 23.6 % and 13.4 %,and the root mean square values of dynamic load had been reduced by 11.9 % and 9.7 %. The suspension performance had been significantly improved.When the road surface excitation input was at a lower level,the improvement of safety indicators as significant;when the road surface excitation input was at a higher level,the improvement of comfort indicators was significant;when the road surface excitation input is in the middle of the road level,the improvement of comprehensive indicators was significant,and the simulation results demonstrated the effectiveness and correctness of the proposed control method.

In order to carry out validation research on active or semi-active suspension actuators and control algorithms of intelligent vehicles,a vehicle suspension experimental platform is proposed here,which can realize real-time force tracking and monitoring of dampers and Rapid Control Prototype (RCP). A vibration evaluation scheme for vehicle suspension is firstly investigated,and the dynamic control equations and transfer functions for typical quarter suspension are developed. A McPherson quarter vehicle suspension test bench with real-time force tracking detection is developed to demonstrate the dynamic output characteristics of real vehicle suspensions and to real-time monitor the control force output characteristics of the actuators. The test bench provides the opportunity to conduct research the optimal trade-off between ride comfort and ride stability for passive suspensions.In addition,the test bench can carry out investigations on optimal control algorithms for active or semi-active suspensions with rapid control prototype technology,which can even conduct studies on actuator uncertainty semi-active control strategies and actuator state observer reliability verification based on the real-time actuator output force value tracking function unique to the test bench,and the validity of the suspension test bench has been examined by means of control algorithm prototype tests.

To address the unclear influence of vessel wall configuration and floating support on the overall finishing of bearing ring,the kinetic energy of particle media and the normal contact force distribution law of each surface of the bearing ring were investigated by EDEM and ADAMS co-simulation tests under different vessel wall configurations and floating support structures. The simulation results show that the vessel wall configuration and floating support have significant influence on the particle media flow. When a vessel with the regular hexagonal cross-section,the inner circle diameter of 300 mm,the support bar diameter of 15 mm,and a uniformly distributed circle diameter of 70 mm is used,the normal contact force on each surface of the bearing ring is at a higher level and the distribution of normal contact force is relatively uniform,which is suitable for finishing. The relevant parameters optimized by simulation were selected to build the overall finishing of bearing ring test platform,and the overall finishing tests were carried out.The experiment results indicate that the surface roughness values of the bearing ring decrease significantly after finishing,and the basic optimization of the equipment parameters of the process is achieved.

Aiming at the problem of difficult machining on the deep blind small hole of the aluminum alloy body of the cylinder combination valve,electrochemcial drilling method was applied.In order to improve the machining stability of electrochemical drilling of deep blind small holes in aluminum alloy,the distributions regularity of electric field and flow field in the machining gap under different electrode feed rates were simulated and analyzed. In order to study the effects of initial gap,electrode feed rate and electrode rotation on electrochemical drilling,the tube electrode with an inner diameter of 0.6 mm and an outer diameter of 1.0 mm was used for machining aluminum alloy 6061. Simulation and experimental results show that a too large or too small electrode feed rate is not conducive to the machining stability of deep blind small hole drilling. When the electrode feed rate is greater than 1.4 mm/min,the hole bottom is easy to form a protrusion,and when the electrode feed rate is less than 1.0 mm/min,the hole bottom is easy to form pits. The initial gap of 0.3 mm is benificial for the orifice morphology,and the electrode rotation facilitates the discharge of product and heat,thus improving processing accuracy. A hole array with the center distance of 2.46 mm and the averaged diameter of 1.28 mm was fabricated by using optimal parameters,and a deep blind small hole with the averaged diameter of 1.29 mm and the depth of 50 mm was obtained.

The pilot relief valve enabling the shock absorber to have continuous damping adjustability,is a crucial component of the continuously variable damping shock absorber. And the annular elastic valve plate is a key component that affecting the opening degree,pressure and other parameters of the pilot relief valve. To find out the deformation regularity of the annular elastic valve plate and its mechanism of mapping to the opening degree of the pilot relief valve,a mathematical model for the deformation of the annular elastic valve plate,which is conducted by calculating and analyzing under the conditions on the pilot valve opening degree′s range of 0~0.25 mm and the valve plate thickness range of 0.1~0.3 mm,was derived. The accuracy of the deformation model was verified,and the changing regularities of the valve opening pressure,inlet pressure,differential pressure for valve hole of pilot relief valve,and opening degree of relief valve were revealed. The curves of relief valve opening pressure,differential pressure,and opening degree versus the thickness of annular elastic valve plate were obtained.This provides theoretical guidance for the highly precise design of pilot relief valves and the matching and optimizing of shock absorbers′ damping.

To address equipment monitoring and management challenges in an aluminum foil factory,a visual equipment monitoring and management system integrating edge nodes,distributed storage and cloud computing was proposed and implemented. The system provides real-time monitoring,information management and fault diagnosis through edge nodes working with private cloud services,reliable storage of production data through distributed storage service clusters,and resource sharing and data analysis with the help of public cloud. The system provides a variety of services such as equipment and personnel management,3D animation support,integrated distributed storage,offline timing prediction,data visualization,etc.,and has the advantages of safety,reliability,easy to expand,and rich functionality.The proposed equipment monitoring and management system has been successfully applied to the plant-wide equipment monitoring of an aluminum foil plant,and the application results show that the system is able to effectively carry out equipment condition monitoring and maintenance.

A novel Swin Transformer semantic segmentation optimization algorithm with a codec structure is proposed to solve the problems such as difficulty in identifying the edge of carbon sliding plate by semantic segmentation model,large interference in complex background,and serious feature information loss. Firstly,the backbone network adopts U-shaped codec structure to realize multi-scale information fusion.Secondly,the attention local enhancement module is added to expand the sensing field and improve the model generalization ability. Then,the upsampling structure with data correlation is used to enhance the quality of upsampling,eliminate the impact of resolution on prediction results,and improve image reconstruction capability. Finally,the skip connection is replaced by the residual path to make the semantic information in the codec structure more closely connected and improve the training efficiency. The experimental results show that the Swin Transformer semantic segmentation algorithm improves the measurement prediction accuracy by 3.63 %,and the average accuracy of pixel classification in all categories is improved by 7.29 %. The research results confirm the superiority and robustness of the Swin Transformer semantic segmentation model in identifying and handling carbon slide tasks.

In response to the problems of inaccurate positioning,low efficiency,and low safety during the assembly of battery cell electrodes in block lithium batteries, the assembly elements and production characteristics of lithium battery cell electrodes were studied. A cell electrode assembly process flow and matching execution mechanical structure were proposed,and fieldbus network communication technology was combined to design a cell electrode assembly control system. The six level linkage device of the cell flip supply structure,orientation correction and transmission mechanism,multi work displacement feeding structure,cell electrode assembly structure,insulation paper pasting structure,and cutting structure in the assembly of cell electrodes was focused from the perspective of mechanical structure. Then,based on Siemens PLC,a communication network topology structure was constructed,and control program algorithm steps were designed to achieve automatic feeding of parts,automatic assembly,automatic transfer,and cutting process of cell electrodes. A visual online management system for the entire process of cell electrode assembly was established. The automatic assembly control system for battery cell electrodes has been applied in lithium battery production enterprises,with significant economic benefits.

A fault diagnosis method based on the AFF-Stablenet model is proposed to address the issues of low diagnostic accuracy and weak generalization of rolling bearings under small sample conditions.Initially,the samples are decomposed into sub-signals of multiple frequencies using EMD. The cross-correlation coefficients between the sub-signals and the original signal are computed. The top three sub-signals with higher coefficients are selected.These sub-signals are transformed into time-frequency representations using CWT. Through attention-based feature fusion,the time-frequency features are integrated.Finally,the fused features are input into the Stablenet model for training and prediction. To validate the effectiveness of the proposed model,comparative experiments are conducted using the Case Western Reserve University bearing dataset and verified using the Politecnico di Torino bearing dataset. Experimental results demonstrate that the AFF-Stablenet model exhibits superior generalization and robustness under small sample conditions compared to other models,affirming the superiority of the proposed approach.

The lightweight design of the wind turbine gearbox house is of great significance to reduce the cost of wind turbines,improve the power density of wind turbines and the mechanical properties of wind turbine structure.Affected by extreme wind conditions and climatic conditions,the housing has been subjected to the impact of irregular loads from the transmission system for a long time,which leads to the problems of incomplete load cases coverage,unreasonable model simplification and low material utilization rate in the optimization design process. Therefore,based on considering multiple extreme service load cases and reasonable equivalence of models,the lightweight design of the housing was carried out.Based on the premise of verifying the accuracy of the model through modal testing,with the minimum mass of the housing as the optimization goal,and with the maximum stress of the housing under different extreme load cases as the constraint condition,multi-condition optimization design was carried out on the main shell,transition plate and rear shell in the components of the housing,and a comparison of the performance of the housing before and after improvement was made.The results show that after optimization,the mass of the housing was reduced by 4.7 %(549.8 kg),and under different load cases,both maximum displacement and stress were significantly reduced. The lightweight design has also improved the strength and stiffness of the housing,which provides a reference for the design improvement of wind turbine gearboxes.