Author Center
Editor Work
Peer Review
Editorial Entry
Email Alert
RSS2019 Vol. 38, No. 3
Display Method:
2019, 38(3): 329-338.
doi: 10.13433/j.cnki.1003-8728.20180300
PDF 878KB
Abstract:
The rapid development of the latest information technology and artificial intelligence that are being integrated in the manufacturing field has prompted all advanced manufacturing countries to actively explore the development strategy of smart manufacturing. Through the plant-wide sharing of the production lifecycle data, the collaboration of business information, the dynamic optimization of decision-making processes, the manufacturing systems can eventually become smart, transparent, and green. In order to understand the connotation of smart manufacturing comprehensively, this paper presented the details about the history and origin of smart manufacturing, the development and evolution of manufacturing modes, as well as the connotation and characteristics of the typical smart manufacturing strategies. Besides, this paper extracted some key enabling technologies that have been implemented effectively, and reviewed the current research of them. Finally, by combining the research foundation on the smart manufacturing from our group, the future development trends of smart manufacturing were analyzed and discussed.
The rapid development of the latest information technology and artificial intelligence that are being integrated in the manufacturing field has prompted all advanced manufacturing countries to actively explore the development strategy of smart manufacturing. Through the plant-wide sharing of the production lifecycle data, the collaboration of business information, the dynamic optimization of decision-making processes, the manufacturing systems can eventually become smart, transparent, and green. In order to understand the connotation of smart manufacturing comprehensively, this paper presented the details about the history and origin of smart manufacturing, the development and evolution of manufacturing modes, as well as the connotation and characteristics of the typical smart manufacturing strategies. Besides, this paper extracted some key enabling technologies that have been implemented effectively, and reviewed the current research of them. Finally, by combining the research foundation on the smart manufacturing from our group, the future development trends of smart manufacturing were analyzed and discussed.
2019, 38(3): 339-343.
doi: 10.13433/j.cnki.1003-8728.20180193
Abstract:
According to the seven-gear planetary transmission system matching problem of a high-speed tracked vehicle, the equivalent torsional vibration model is established by the concentrated mass method. The natural frequencies of the undamped torsional vibration system and the sensitivity of the vibration mode to the moment of inertia and stiffness are analyzed. Combining with the inherent characteristics of the planetary gear mechanism and the internal and external excitation frequencies, the resonance conditions of each gear of the planetary transmission mechanism are analyzed. Based on the sensitivity analysis, the dynamic parameters of the torsion vibration model are optimized. The simulation results show that the optimized dynamic parameters can avoid torsional resonance better. The results of the study provide some guidance for the further analysis of the matching influence of the planetary transmission system on the torsional vibration and the improvement of torsional vibration performance.
According to the seven-gear planetary transmission system matching problem of a high-speed tracked vehicle, the equivalent torsional vibration model is established by the concentrated mass method. The natural frequencies of the undamped torsional vibration system and the sensitivity of the vibration mode to the moment of inertia and stiffness are analyzed. Combining with the inherent characteristics of the planetary gear mechanism and the internal and external excitation frequencies, the resonance conditions of each gear of the planetary transmission mechanism are analyzed. Based on the sensitivity analysis, the dynamic parameters of the torsion vibration model are optimized. The simulation results show that the optimized dynamic parameters can avoid torsional resonance better. The results of the study provide some guidance for the further analysis of the matching influence of the planetary transmission system on the torsional vibration and the improvement of torsional vibration performance.
2019, 38(3): 344-350.
doi: 10.13433/j.cnki.1003-8728.20180190
Abstract:
Taking the 600 Wh flywheel energy storage system as the research object, in order to calculate the critical rotational speed distribution of the flywheel rotor system and evaluate the stability of the movement at high rotational speed, the SolidWorks was used for 3D modeling and the finite element software SAMCEF Rotor was used to solve the critical rotational speeds and vibration modes. The dynamic experimental data of the flywheel energy storage system were collected and analyzed by the time domain diagram, frequency domain diagram and axial orbit diagram. Comparing the finite element analysis with the experimental data, the results show that the vibration amplitude of the flywheel rotor system at the critical speed is significantly greater than the steady speed; in the working speed, the recurrence of the axial trajectory is good, and it does not exceed the air gap value of 0.3 mm indicating good stability performance, which also provides reference and basis for the future improvement and design of different energy storage flywheel rotors.
Taking the 600 Wh flywheel energy storage system as the research object, in order to calculate the critical rotational speed distribution of the flywheel rotor system and evaluate the stability of the movement at high rotational speed, the SolidWorks was used for 3D modeling and the finite element software SAMCEF Rotor was used to solve the critical rotational speeds and vibration modes. The dynamic experimental data of the flywheel energy storage system were collected and analyzed by the time domain diagram, frequency domain diagram and axial orbit diagram. Comparing the finite element analysis with the experimental data, the results show that the vibration amplitude of the flywheel rotor system at the critical speed is significantly greater than the steady speed; in the working speed, the recurrence of the axial trajectory is good, and it does not exceed the air gap value of 0.3 mm indicating good stability performance, which also provides reference and basis for the future improvement and design of different energy storage flywheel rotors.
2019, 38(3): 351-357.
doi: 10.13433/j.cnki.1003-8728.20180251
Abstract:
Automatic feature extraction plays a crucial role in the intelligent state monitoring of mechanical systems, which can adaptively learn features from raw data and discover new state-sensitive features. This research focuses on the ability of different depth convolution neural network (CNN) models to mine representative information and sensitive features from the excitation response signal without prior knowledge, and combine the feature extraction with assembly tightness classification process of the bolted structure. The effectiveness of the method is verified by excitation test data of car frame test bench bolt connection rotor. The results show that the feature learned adaptively by CNN can represent the complex mapping relationship between response signal and assembly state, and has higher accuracy than other methods.
Automatic feature extraction plays a crucial role in the intelligent state monitoring of mechanical systems, which can adaptively learn features from raw data and discover new state-sensitive features. This research focuses on the ability of different depth convolution neural network (CNN) models to mine representative information and sensitive features from the excitation response signal without prior knowledge, and combine the feature extraction with assembly tightness classification process of the bolted structure. The effectiveness of the method is verified by excitation test data of car frame test bench bolt connection rotor. The results show that the feature learned adaptively by CNN can represent the complex mapping relationship between response signal and assembly state, and has higher accuracy than other methods.
2019, 38(3): 358-364.
doi: 10.13433/j.cnki.1003-8728.20180323
Abstract:
In engineering applications, retrieval of existing 3D CAD models and reuse of their design information can reduce the time and cost of design and product. A classification and retrieval algorithm for three-dimensional CAD model with hidden Markov models (HMM) based on wavelet moment and affine invariant moment is proposed in this study. Firstly, the 3D model image is normalized, and the wavelet moment eigenvalues and affine invariant moment eigenvalues of the normalized image are extracted. Then the combined invariant moment features with high robustness and stability are selected by the K-W test algorithm and encoded. Finally constructing five type samples of three-dimensional models and using the above eigenvalues as input observations of HMM. The modified Baum-Welch (B-W) algorithm with scaling and multi-observation sequences is used to train and identify the model. The proposed algorithm is compared with other three algorithms. The results show that the proposed algorithm has better recognition rate and retrieval efficiency, and more practical engineering value.
In engineering applications, retrieval of existing 3D CAD models and reuse of their design information can reduce the time and cost of design and product. A classification and retrieval algorithm for three-dimensional CAD model with hidden Markov models (HMM) based on wavelet moment and affine invariant moment is proposed in this study. Firstly, the 3D model image is normalized, and the wavelet moment eigenvalues and affine invariant moment eigenvalues of the normalized image are extracted. Then the combined invariant moment features with high robustness and stability are selected by the K-W test algorithm and encoded. Finally constructing five type samples of three-dimensional models and using the above eigenvalues as input observations of HMM. The modified Baum-Welch (B-W) algorithm with scaling and multi-observation sequences is used to train and identify the model. The proposed algorithm is compared with other three algorithms. The results show that the proposed algorithm has better recognition rate and retrieval efficiency, and more practical engineering value.
2019, 38(3): 365-372.
doi: 10.13433/j.cnki.1003-8728.20180289
Abstract:
Aiming at the multi-solution problem of robot inverse kinematics, a solution method of multi-module radial basis function network based on singular trajectory is proposed. According to the singular trajectory theory, the joint space is strictly divided into multiple joint subspace regions with only single inverse solution. Then, through forward kinematics, each joint subspace is mapped to the same workspace to obtain multiple sets of training samples with the same working space but different joint regions. Each training sample only contains a single inverse solution. In a multi-module radial basis neural function network structure, each sub-module is responsible for learning a set of training samples. Therefore, the inverse kinematics multiple solution problem is transformed into the training problem of neural network weights of each sub-module. Multi-solution calculation of robot inverse kinematics is realized by inverse solution prediction of multiple modules in neural network structure. The example shows that multi-module neural networks based on singular trajectories can correctly output multiple sets of inverse solutions, moreover, it satisfies the requirements of solving accuracy and has a good generalization value in the inverse kinematics solution.
Aiming at the multi-solution problem of robot inverse kinematics, a solution method of multi-module radial basis function network based on singular trajectory is proposed. According to the singular trajectory theory, the joint space is strictly divided into multiple joint subspace regions with only single inverse solution. Then, through forward kinematics, each joint subspace is mapped to the same workspace to obtain multiple sets of training samples with the same working space but different joint regions. Each training sample only contains a single inverse solution. In a multi-module radial basis neural function network structure, each sub-module is responsible for learning a set of training samples. Therefore, the inverse kinematics multiple solution problem is transformed into the training problem of neural network weights of each sub-module. Multi-solution calculation of robot inverse kinematics is realized by inverse solution prediction of multiple modules in neural network structure. The example shows that multi-module neural networks based on singular trajectories can correctly output multiple sets of inverse solutions, moreover, it satisfies the requirements of solving accuracy and has a good generalization value in the inverse kinematics solution.
2019, 38(3): 373-378.
doi: 10.13433/j.cnki.1003-8728.20180189
Abstract:
A kind of smooth three-dimensional non-autonomous fast and slow coupled tufting oscillation circuit is established by introducing periodic excitation. By adjusting the frequency of the external excitation alternating current source, the system exhibits obvious two-scale effects, namely fast and slow variables. Combining with numerical simulation and two time scale method is used to analyze the cluster mechanism and bifurcation mode of the circuit model system under different excitation frequencies and amplitudes. The fractional chaotic circuit is used to control the evolution of the dynamic behavior of the system. The results show that if there is a difference in magnitude between the natural frequency of the system and the frequency of the external excitation, the system has obvious tufting phenomena of fast and slow coupling. The smaller the external excitation frequency is, the faster and slower the system effect when the amplitude is small or constant, and the more obvious the phenomenon is. At the same time, the fractional-order chaotic circuit is used to control the evolution of dynamic system effectively.
A kind of smooth three-dimensional non-autonomous fast and slow coupled tufting oscillation circuit is established by introducing periodic excitation. By adjusting the frequency of the external excitation alternating current source, the system exhibits obvious two-scale effects, namely fast and slow variables. Combining with numerical simulation and two time scale method is used to analyze the cluster mechanism and bifurcation mode of the circuit model system under different excitation frequencies and amplitudes. The fractional chaotic circuit is used to control the evolution of the dynamic behavior of the system. The results show that if there is a difference in magnitude between the natural frequency of the system and the frequency of the external excitation, the system has obvious tufting phenomena of fast and slow coupling. The smaller the external excitation frequency is, the faster and slower the system effect when the amplitude is small or constant, and the more obvious the phenomenon is. At the same time, the fractional-order chaotic circuit is used to control the evolution of dynamic system effectively.
2019, 38(3): 379-385.
doi: 10.13433/j.cnki.1003-8728.20180177
Abstract:
A numerical analytical model of the granular lubrication interface was established with the discrete element method (DEM). With the DEM, the dynamic characteristics of granular macro-flow, coordination number and force chain evolution as well as the bearing capacity of the pressure load were studied. The study results show that the average velocity of the granular lubrication interface decreases with the increasing height of the gap y, while its fluctuation velocity of is opposite to the average velocity. The microscopic contact structures and dynamic properties of the granular lubrication interface are transient and changeable, and its contact intensity and contact strength increase with the increasing pressure load P. The bearing capacity of the granular lubrication interface is closely related to the variation of force chain evolution and increases with the increasing pressure load P.
A numerical analytical model of the granular lubrication interface was established with the discrete element method (DEM). With the DEM, the dynamic characteristics of granular macro-flow, coordination number and force chain evolution as well as the bearing capacity of the pressure load were studied. The study results show that the average velocity of the granular lubrication interface decreases with the increasing height of the gap y, while its fluctuation velocity of is opposite to the average velocity. The microscopic contact structures and dynamic properties of the granular lubrication interface are transient and changeable, and its contact intensity and contact strength increase with the increasing pressure load P. The bearing capacity of the granular lubrication interface is closely related to the variation of force chain evolution and increases with the increasing pressure load P.
2019, 38(3): 386-391.
doi: 10.13433/j.cnki.1003-8728.20180188
Abstract:
In order to study the dynamic cutting processes and coal power transfer laws of a shearer's drum, its model drum was built based on the similarity theory; the prototype was similar to the parameters of similarity criterion. The structural parameters, movement parameters and cutting material characteristics of the drum to be simulated are determined. The Pro/E and EDEM were applied. The simulation analysis and verification of the prototype drum, the dynamic cutting process, coal particles' motion trail, quantitative change rule, coal rate and drum loading are the same. The cylinder model validation results show that the prototype drum's coal particle distributions are similar in the cutting process; the drum's coal rate prediction error is 1.82%; the loading error is less than 2.1%; the coal rate and loading are satisfactory, verifying the accuracy of the drum's similarity criterion.
In order to study the dynamic cutting processes and coal power transfer laws of a shearer's drum, its model drum was built based on the similarity theory; the prototype was similar to the parameters of similarity criterion. The structural parameters, movement parameters and cutting material characteristics of the drum to be simulated are determined. The Pro/E and EDEM were applied. The simulation analysis and verification of the prototype drum, the dynamic cutting process, coal particles' motion trail, quantitative change rule, coal rate and drum loading are the same. The cylinder model validation results show that the prototype drum's coal particle distributions are similar in the cutting process; the drum's coal rate prediction error is 1.82%; the loading error is less than 2.1%; the coal rate and loading are satisfactory, verifying the accuracy of the drum's similarity criterion.
2019, 38(3): 392-397.
doi: 10.13433/j.cnki.1003-8728.20180187
Abstract:
Compared with the conventional annular jet pump, the annular self-excited oscillating jet pump can form a circular oscillation effect in the self-excited oscillating chamber only by its own reasonable size design without external excitation. The self-excited oscillation chamber generates a strong self-oscillating vortex to enhance the effect of gas-liquid two-phase mass transfer. In this paper, a three-dimensional computational fluid dynamics of a ring-shaped self-oscillating jet pump is performed using the k-ε turbulence model and combining computational and experimental methods. The ability of dissolved oxygen in water after deoxygenation is studied. The results show that the mixing capacity of the annular self-oscillating jet pump is 15%~20% higher than that of the traditional annular jet pump. Compared with the traditional aerator, the mixing capacity is increased by more than 30%. In the mixing process, the pulse effect due to self-oscillation causes a sharp shear field between the gas phase and the liquid phase. Under the action of the shear force, the number of bubbles in the mixed liquid increases, and the bubble particle size becomes smaller, thereby increasing the area in which the gas and liquid phases are in contact with each other increases the mass transfer of the gas-liquid two-phase mixture.
Compared with the conventional annular jet pump, the annular self-excited oscillating jet pump can form a circular oscillation effect in the self-excited oscillating chamber only by its own reasonable size design without external excitation. The self-excited oscillation chamber generates a strong self-oscillating vortex to enhance the effect of gas-liquid two-phase mass transfer. In this paper, a three-dimensional computational fluid dynamics of a ring-shaped self-oscillating jet pump is performed using the k-ε turbulence model and combining computational and experimental methods. The ability of dissolved oxygen in water after deoxygenation is studied. The results show that the mixing capacity of the annular self-oscillating jet pump is 15%~20% higher than that of the traditional annular jet pump. Compared with the traditional aerator, the mixing capacity is increased by more than 30%. In the mixing process, the pulse effect due to self-oscillation causes a sharp shear field between the gas phase and the liquid phase. Under the action of the shear force, the number of bubbles in the mixed liquid increases, and the bubble particle size becomes smaller, thereby increasing the area in which the gas and liquid phases are in contact with each other increases the mass transfer of the gas-liquid two-phase mixture.
2019, 38(3): 398-403.
doi: 10.13433/j.cnki.1003-8728.20180196
Abstract:
The ride comfort can be greatly improved thought the control of suspension variable stiffness. Based on the electronically controlled air suspension system (ECAS), the front suspension adopts the feedback control method, and the rear suspension adopts the control method combined with pre-aim feedforward and feedback. The suspension stiffness can be adjusted by filling air spring. The ECAS model and control algorithm are built respectively in AMESim and Simulink, and the vehicle model is built in TruckSim. In order to improve the tracking accuracy, the flexible PID control strategy is adopted. The experimental results demonstrate that the proposed control algorithm in this paper is suit to solve the problem of the over charging and over discharging when the height control is realized. The control algorithm is verified by the combined simulation. By measuring the smoothness index, the acceleration of the body was reduced by 60% compared with the non-control vehicle, indicating that the comfort was greatly improved.
The ride comfort can be greatly improved thought the control of suspension variable stiffness. Based on the electronically controlled air suspension system (ECAS), the front suspension adopts the feedback control method, and the rear suspension adopts the control method combined with pre-aim feedforward and feedback. The suspension stiffness can be adjusted by filling air spring. The ECAS model and control algorithm are built respectively in AMESim and Simulink, and the vehicle model is built in TruckSim. In order to improve the tracking accuracy, the flexible PID control strategy is adopted. The experimental results demonstrate that the proposed control algorithm in this paper is suit to solve the problem of the over charging and over discharging when the height control is realized. The control algorithm is verified by the combined simulation. By measuring the smoothness index, the acceleration of the body was reduced by 60% compared with the non-control vehicle, indicating that the comfort was greatly improved.
2019, 38(3): 404-408.
doi: 10.13433/j.cnki.1003-8728.20180172
Abstract:
A new grinding technology based on five-bar slider mechanism is proposed, parts of movement structure are designed, and feasibility of the technical scheme is analyzed. The mathematical model for plane lapping and spherical grinding is established, and the equation of grinding trajectory curve is obtained with matrix algebra. With MATLAB calculation, the computer simulation is carried out to obtain the plane and spherical grinding trajectory under different parameters. The simulation results show that when the speed of driver and the speed of the workpiece are not multiplied, which can improve the complexity of trajectory and reduce the surface roughness. Meanwhile, the simulation results prove the feasibility of the technology scheme applied to the plane and spherical grinding process.
A new grinding technology based on five-bar slider mechanism is proposed, parts of movement structure are designed, and feasibility of the technical scheme is analyzed. The mathematical model for plane lapping and spherical grinding is established, and the equation of grinding trajectory curve is obtained with matrix algebra. With MATLAB calculation, the computer simulation is carried out to obtain the plane and spherical grinding trajectory under different parameters. The simulation results show that when the speed of driver and the speed of the workpiece are not multiplied, which can improve the complexity of trajectory and reduce the surface roughness. Meanwhile, the simulation results prove the feasibility of the technology scheme applied to the plane and spherical grinding process.
2019, 38(3): 409-414.
doi: 10.13433/j.cnki.1003-8728.20180194
Abstract:
It is difficult to observe the infiltration process of droplets in the powder bed through experiments, therefore a mathematical model is established with the horizontal set method, in order to improve the accuracy of three-dimension printing(3DP) process and to predict the bonding of binder and granular material. The numerical simulation of droplets infiltration process is carried out via COMSOL software. The forms of droplets infiltration are influenced mainly by the droplets horizontal spacing and drop height. The experiment platform is set up to verify the consistency of the mathematical model and simulation results. The high-speed camera is utilized to capture the images of droplets infiltration process and the electron microscopy is utilized to observe the final condensed forms. The experimental results are compared with the simulated, in which the accuracy is verified, and which provides a basis for further research.
It is difficult to observe the infiltration process of droplets in the powder bed through experiments, therefore a mathematical model is established with the horizontal set method, in order to improve the accuracy of three-dimension printing(3DP) process and to predict the bonding of binder and granular material. The numerical simulation of droplets infiltration process is carried out via COMSOL software. The forms of droplets infiltration are influenced mainly by the droplets horizontal spacing and drop height. The experiment platform is set up to verify the consistency of the mathematical model and simulation results. The high-speed camera is utilized to capture the images of droplets infiltration process and the electron microscopy is utilized to observe the final condensed forms. The experimental results are compared with the simulated, in which the accuracy is verified, and which provides a basis for further research.
2019, 38(3): 415-421.
doi: 10.13433/j.cnki.1003-8728.20180191
Abstract:
In the paper, a new algorithm for the adaptive layering of triangular facet of the stereolithography(STL) model is proposed in the additive manufacturing technology. In this method, the minimum normal vector of the triangular facet intersecting with the thickness t of the layer is found, and then the relationship between the x-y resolution in the three-dimensional model voxels and the triangular facet normal vector in the STL model and the change direction of the manufacturing direction is applied to achieve adaptive layer of the STL model. With this method, the complexity of layering parameter processing is reduced, the adaptive layering becomes simpler and easier to realize, and the obtained layer thickness is more accurate. Since the adaptive layering is the layer thickness obtained within the allowable errors range, layering in the contour change area of the STL model is more refined.
In the paper, a new algorithm for the adaptive layering of triangular facet of the stereolithography(STL) model is proposed in the additive manufacturing technology. In this method, the minimum normal vector of the triangular facet intersecting with the thickness t of the layer is found, and then the relationship between the x-y resolution in the three-dimensional model voxels and the triangular facet normal vector in the STL model and the change direction of the manufacturing direction is applied to achieve adaptive layer of the STL model. With this method, the complexity of layering parameter processing is reduced, the adaptive layering becomes simpler and easier to realize, and the obtained layer thickness is more accurate. Since the adaptive layering is the layer thickness obtained within the allowable errors range, layering in the contour change area of the STL model is more refined.
2019, 38(3): 422-427.
doi: 10.13433/j.cnki.1003-8728.20180174
Abstract:
In order to reveal the phenomenon of root digging on a tooth surface after numerical control power gear-honing with internal teeth honing wheel, the mechanical model of involute tooth surface in single abrasive cutting was built with the cutting force model of a single abrasive particle. The honing pressure distribution on tooth surface was analyzed, using 3D finite element method to further verify the maximum tooth root stress. The analysis results show that honing pressure is one of the causes for tooth surface root digging. According to the contact equation of internal meshing tooth surface and the analysis of the relative velocity of involute tooth profile, the relative velocity variation of the gear profile on the end surface of the honing gear is the largest at the root of the gear and produced by the internal teeth honing wheel with power gear-honing, thus causing the tooth surface root digging of multiple internal teeth honing wheels; the relative velocity of tooth surface is also one of key factors affecting tooth surface root digging. Lastly, experimental results prove in two ways that the power gear-honing and the phenomenon of tooth surface root digging are easily produced with variable pressure gear-honing with center-distance.
In order to reveal the phenomenon of root digging on a tooth surface after numerical control power gear-honing with internal teeth honing wheel, the mechanical model of involute tooth surface in single abrasive cutting was built with the cutting force model of a single abrasive particle. The honing pressure distribution on tooth surface was analyzed, using 3D finite element method to further verify the maximum tooth root stress. The analysis results show that honing pressure is one of the causes for tooth surface root digging. According to the contact equation of internal meshing tooth surface and the analysis of the relative velocity of involute tooth profile, the relative velocity variation of the gear profile on the end surface of the honing gear is the largest at the root of the gear and produced by the internal teeth honing wheel with power gear-honing, thus causing the tooth surface root digging of multiple internal teeth honing wheels; the relative velocity of tooth surface is also one of key factors affecting tooth surface root digging. Lastly, experimental results prove in two ways that the power gear-honing and the phenomenon of tooth surface root digging are easily produced with variable pressure gear-honing with center-distance.
2019, 38(3): 428-432.
doi: 10.13433/j.cnki.1003-8728.20180324
Abstract:
In order to evaluate the circularity errors and its uncertainty more accurately and easily, the circularity error model is established with the least square method, and the parameters of objective function are optimized via BP neural network algorithm. The principle and implementation method of BP neural network algorithm are described. An example is given to show that the method can obtain the optimal solution for the nonlinear optimization problem of circularity errors evaluation. The uncertainty of roundness error is calculated with the Guide to expression of uncertainty in measurement (GUM) and Monte Carlo method. The reliability and accuracy of Monte Carlo method are verified. This method does not need to calculate the transfer coefficient of the model. It is easy to realize via MATLAB and provides a more convenient method for evaluating uncertainty of circularity error measurement results.
In order to evaluate the circularity errors and its uncertainty more accurately and easily, the circularity error model is established with the least square method, and the parameters of objective function are optimized via BP neural network algorithm. The principle and implementation method of BP neural network algorithm are described. An example is given to show that the method can obtain the optimal solution for the nonlinear optimization problem of circularity errors evaluation. The uncertainty of roundness error is calculated with the Guide to expression of uncertainty in measurement (GUM) and Monte Carlo method. The reliability and accuracy of Monte Carlo method are verified. This method does not need to calculate the transfer coefficient of the model. It is easy to realize via MATLAB and provides a more convenient method for evaluating uncertainty of circularity error measurement results.
2019, 38(3): 433-439.
doi: 10.13433/j.cnki.1003-8728.20180178
Abstract:
In the pose adjustment of millimeter size and the precision positioning of Micro-nano size for the astronomical optical mirror, the workspace of the novel 3SPS+3(SP-U) parallel mechanism with six degrees of freedom (DOFs) which based on fully flexible hinges is mainly determined by the revolving space of flexible hinges. The motion degrees of freedom and partial decoupling characteristics of the novel parallel mechanism are analyzed. According to the geometric parameters of the physical prototype, the model of inverse kinematics is established. On this basis, the search method is used to analyze the workspace of the mechanism, and the laser tracker is used to measure the actual workspace. Both the results of theoretical calculation and measurement show that the translational space of the mechanism's moving platform can reach ±3 mm, as well as the corner space can reach ±1°. It ensures that the mechanism has sufficient workspace in the application, indicating the mechanism has achieved the design goals.
In the pose adjustment of millimeter size and the precision positioning of Micro-nano size for the astronomical optical mirror, the workspace of the novel 3SPS+3(SP-U) parallel mechanism with six degrees of freedom (DOFs) which based on fully flexible hinges is mainly determined by the revolving space of flexible hinges. The motion degrees of freedom and partial decoupling characteristics of the novel parallel mechanism are analyzed. According to the geometric parameters of the physical prototype, the model of inverse kinematics is established. On this basis, the search method is used to analyze the workspace of the mechanism, and the laser tracker is used to measure the actual workspace. Both the results of theoretical calculation and measurement show that the translational space of the mechanism's moving platform can reach ±3 mm, as well as the corner space can reach ±1°. It ensures that the mechanism has sufficient workspace in the application, indicating the mechanism has achieved the design goals.
2019, 38(3): 440-444.
doi: 10.13433/j.cnki.1003-8728.20180170
Abstract:
In order to improve the control precision of the system, it is necessary to improve the approximation accuracy of the fuzzy system, so a large number of fuzzy rules are needed. A large number of fuzzy rules will cause overburden of control system calculation and can not meet the requirement of real time control. In this paper, a new decomposition fuzzy system is used to compensate the uncertain of mechanical system, and a robust adaptive control law is designed to control the trajectory of the system. The simulation results show that, the adaptive decomposition fuzzy control designed in this paper can not only compensate for the unknown part of the mechanical system in real time, but also has higher control accuracy, faster convergence and faster real-time control than the traditional adaptive fuzzy control.
In order to improve the control precision of the system, it is necessary to improve the approximation accuracy of the fuzzy system, so a large number of fuzzy rules are needed. A large number of fuzzy rules will cause overburden of control system calculation and can not meet the requirement of real time control. In this paper, a new decomposition fuzzy system is used to compensate the uncertain of mechanical system, and a robust adaptive control law is designed to control the trajectory of the system. The simulation results show that, the adaptive decomposition fuzzy control designed in this paper can not only compensate for the unknown part of the mechanical system in real time, but also has higher control accuracy, faster convergence and faster real-time control than the traditional adaptive fuzzy control.
2019, 38(3): 445-451.
doi: 10.13433/j.cnki.1003-8728.20180175
Abstract:
In order to manufacture high precision printing package embedded in electronic products in 3D printing to achieve high viscosity conductive material, through theoretical analysis and experimental verification, the influence of the nozzle structure and the electric field of fluid velocity on the liquid droplets form is studied. It is concluded that under the same pressure, the shorter the nozzle tip is, the larger the section shrinkage and the faster the liquid flow rate. In addition to the effect of the shrinkage of the electric field on the droplets forming the Taylor cone, the liquid velocity can be affected. For high viscosity printing of conductive materials, the biggest influencing factor are the result of the printing pressure and platform movement speed, to a certain pressure and voltage range, to print all can achieve, and the movement speed by adjusting the platform can improve the quality of jet printing. The research results provide a theoretical foundation and direction for improving the 3D printing appearance, forming accuracy and controllability of high viscosity conductive materials.
In order to manufacture high precision printing package embedded in electronic products in 3D printing to achieve high viscosity conductive material, through theoretical analysis and experimental verification, the influence of the nozzle structure and the electric field of fluid velocity on the liquid droplets form is studied. It is concluded that under the same pressure, the shorter the nozzle tip is, the larger the section shrinkage and the faster the liquid flow rate. In addition to the effect of the shrinkage of the electric field on the droplets forming the Taylor cone, the liquid velocity can be affected. For high viscosity printing of conductive materials, the biggest influencing factor are the result of the printing pressure and platform movement speed, to a certain pressure and voltage range, to print all can achieve, and the movement speed by adjusting the platform can improve the quality of jet printing. The research results provide a theoretical foundation and direction for improving the 3D printing appearance, forming accuracy and controllability of high viscosity conductive materials.
2019, 38(3): 452-456.
doi: 10.13433/j.cnki.1003-8728.20180186
Abstract:
To improve the load performance of the electric load simulator, the corresponding mathematical model was established according to the system configuration of the electric load simulator, The influence of stiffness and load inertia of the system for the surplus torque, the system stability and the response speed was analyzed, so as to analyze and define the optimal stiffness and load inertia. The simulation results were verified by MATLAB software. The simulation results showed that the sinusoidal curve could be reproduced well, the response speed was fast and the stability was good, and it is necessary to consider not only the suppression of surplus torque, but also the stability of the system and the rapid response when designing the optimal stiffness and load inertia of the electric load simulator.
To improve the load performance of the electric load simulator, the corresponding mathematical model was established according to the system configuration of the electric load simulator, The influence of stiffness and load inertia of the system for the surplus torque, the system stability and the response speed was analyzed, so as to analyze and define the optimal stiffness and load inertia. The simulation results were verified by MATLAB software. The simulation results showed that the sinusoidal curve could be reproduced well, the response speed was fast and the stability was good, and it is necessary to consider not only the suppression of surplus torque, but also the stability of the system and the rapid response when designing the optimal stiffness and load inertia of the electric load simulator.
2019, 38(3): 457-464.
doi: 10.13433/j.cnki.1003-8728.20180179
Abstract:
In terms of the traditional search method of workspace having the boundary problem of low accuracy and low efficiency, a general fast search method was put forward via the theory of cellular automata. This algorithm based on the thought of neighborhood of cellular automata and the golden section method. First, the traditional numerical methods of the workspace of parallel mechanism were studied and summarized their deficiency; the method introduced the neighborhood of cellular automata so as to extract roughness boundary and the search direction. Then, the golden section method was introduced into dimension search so as to realize search of arbitrary precision of boundary point. The quickness and the effectiveness of algorithm are tested by simulation examples, and compared with polar coordinate searching method, Monte Carlo method and the process of precision processing show that the algorithm has obvious advantages in the accuracy and efficiency, and is suitable for 2D or 3D the workspace problems of parallel manipulator.
In terms of the traditional search method of workspace having the boundary problem of low accuracy and low efficiency, a general fast search method was put forward via the theory of cellular automata. This algorithm based on the thought of neighborhood of cellular automata and the golden section method. First, the traditional numerical methods of the workspace of parallel mechanism were studied and summarized their deficiency; the method introduced the neighborhood of cellular automata so as to extract roughness boundary and the search direction. Then, the golden section method was introduced into dimension search so as to realize search of arbitrary precision of boundary point. The quickness and the effectiveness of algorithm are tested by simulation examples, and compared with polar coordinate searching method, Monte Carlo method and the process of precision processing show that the algorithm has obvious advantages in the accuracy and efficiency, and is suitable for 2D or 3D the workspace problems of parallel manipulator.
2019, 38(3): 465-471.
doi: 10.13433/j.cnki.1003-8728.20180192
Abstract:
In order to achieve the goal of rapid and accurate inspection of the inner cone angles of fuel nozzles in the aero engine, a precision measuring system for this parameter is built up according to the measuring principle of two spheres. Based on the three-axis motion platform, linear variable differential transformer, floating holder and multi-axis motion controller etc, the system could be utilized to solve the automatic measurement of the inner cone angles of fuel nozzles. In the measuring system, the compact column structure is adopted and the motion mechanism is composed of X, Y and Z linear axes. At the aspect of the control section, the digital composite control mode of position closed loop of two levels is employed, which is based on the combination of the mainstream industrial computer and the special programmable multi-axis controller. Also, the new special probe for the inner cone angle is designed on account of the floating entry technology. Finally, the inner cone features of three work-pieces and a fuel nozzle are measured several times on the measuring system established. The experimental results show that the system could accomplish the appointed measuring task and the measuring accuracy for meeting the inspecting requirements. Therefore, the precision measuring system for inner cone angles of fuel nozzles could serve as a solution for the measurement of the shapes and dimensions of fuel nozzles.
In order to achieve the goal of rapid and accurate inspection of the inner cone angles of fuel nozzles in the aero engine, a precision measuring system for this parameter is built up according to the measuring principle of two spheres. Based on the three-axis motion platform, linear variable differential transformer, floating holder and multi-axis motion controller etc, the system could be utilized to solve the automatic measurement of the inner cone angles of fuel nozzles. In the measuring system, the compact column structure is adopted and the motion mechanism is composed of X, Y and Z linear axes. At the aspect of the control section, the digital composite control mode of position closed loop of two levels is employed, which is based on the combination of the mainstream industrial computer and the special programmable multi-axis controller. Also, the new special probe for the inner cone angle is designed on account of the floating entry technology. Finally, the inner cone features of three work-pieces and a fuel nozzle are measured several times on the measuring system established. The experimental results show that the system could accomplish the appointed measuring task and the measuring accuracy for meeting the inspecting requirements. Therefore, the precision measuring system for inner cone angles of fuel nozzles could serve as a solution for the measurement of the shapes and dimensions of fuel nozzles.
2019, 38(3): 472-479.
doi: 10.13433/j.cnki.1003-8728.20180173
Abstract:
The calibration of kinematic error of parallel robot is one of the main problems in the application of parallel robot. The selection and identification algorithm of measuring positions has important influence on the result of parameter identification and the effect of error compensation. In engineering practice, in order to improve the efficiency of measurement or to be restricted by the measurement environment, measurement data are often obtained with simple position and less number of positions, which may lead to the strong complex collinearity in the linear regression model. Using the means of residual proportion index and the principal component analysis (PCA), the algorithms for optimal measurement configuration selection and robust source error identification are investigated to realize the dimensionality reduction of the variable space, two important issues for improving the measurement efficiency as well as identification accuracy. Computer simulation shows that the proposed method is correct and feasible.
The calibration of kinematic error of parallel robot is one of the main problems in the application of parallel robot. The selection and identification algorithm of measuring positions has important influence on the result of parameter identification and the effect of error compensation. In engineering practice, in order to improve the efficiency of measurement or to be restricted by the measurement environment, measurement data are often obtained with simple position and less number of positions, which may lead to the strong complex collinearity in the linear regression model. Using the means of residual proportion index and the principal component analysis (PCA), the algorithms for optimal measurement configuration selection and robust source error identification are investigated to realize the dimensionality reduction of the variable space, two important issues for improving the measurement efficiency as well as identification accuracy. Computer simulation shows that the proposed method is correct and feasible.
2019, 38(3): 480-486.
doi: 10.13433/j.cnki.1003-8728.20180185
Abstract:
In order to trace the cause of gearbox body cracks of a diesel locomotive, a finite element model for the gearbox was established. A modified quasi-static superposition method according to the real dynamic stress was proposed to correct the response produced by unit acceleration and obtain the stress-time history. The maximum absolute principal-stress-time history was respectively acquired by the quasi-static superposition method and modified quasi-static superposition method. Then, the rain flow counting method was used to count the cycle of stress-time history. Finally, the fatigue life of the gearbox body was predicted with Miner's cumulative damage theory. The result of the latter method is in a good agreement with the experimental. The significant difference between the result via two methods is preliminarily believed to be caused by the resonance of gearbox body. It is proved by comparison between the acceleration spectrum analysis and the gearbox body modal analysis. This investigation provides references for predicting the fatigue life of gearbox body in structural design and optimization.
In order to trace the cause of gearbox body cracks of a diesel locomotive, a finite element model for the gearbox was established. A modified quasi-static superposition method according to the real dynamic stress was proposed to correct the response produced by unit acceleration and obtain the stress-time history. The maximum absolute principal-stress-time history was respectively acquired by the quasi-static superposition method and modified quasi-static superposition method. Then, the rain flow counting method was used to count the cycle of stress-time history. Finally, the fatigue life of the gearbox body was predicted with Miner's cumulative damage theory. The result of the latter method is in a good agreement with the experimental. The significant difference between the result via two methods is preliminarily believed to be caused by the resonance of gearbox body. It is proved by comparison between the acceleration spectrum analysis and the gearbox body modal analysis. This investigation provides references for predicting the fatigue life of gearbox body in structural design and optimization.
2019, 38(3): 487-492.
doi: 10.13433/j.cnki.1003-8728.20180182
Abstract:
Under the guaranteeing the computational accuracy, to ensure the simulation is more in line with the real practical results and increase the calculation efficiency, the classical hybrid interface substructure component modal synthesis method (HISCMSM) is improved. The finite element model (FEM) of the mistuned bladed disk assemblies (MBDA) with film holes is built by this methodology and the modal is analyzed. Comparing with the high fidelity integral structure finite element model method (FEMM), the modal deviation is not more than 0.79%, but the computational time is shortened by 31.01%~55.78%, the computational efficiency is higher by 0.89%~5.45% than that via classical HICMSM. So the computational efficiency and the computational accuracy of this method are both superior to the classical HICMSM. The effect of the film holes on the modal is investigated, comparing with the solid MBDA, the modal is declined. Firstly the modal frequencies decrease after an increase with the increasing of size, and they almost do not alter with the arrangement and the angle changing, but they are impacted by the number of the film holes. The above investigations lay the foundation for the further study on the dynamic response of the MBDA and have a certain guiding significance for the design of the BDA.
Under the guaranteeing the computational accuracy, to ensure the simulation is more in line with the real practical results and increase the calculation efficiency, the classical hybrid interface substructure component modal synthesis method (HISCMSM) is improved. The finite element model (FEM) of the mistuned bladed disk assemblies (MBDA) with film holes is built by this methodology and the modal is analyzed. Comparing with the high fidelity integral structure finite element model method (FEMM), the modal deviation is not more than 0.79%, but the computational time is shortened by 31.01%~55.78%, the computational efficiency is higher by 0.89%~5.45% than that via classical HICMSM. So the computational efficiency and the computational accuracy of this method are both superior to the classical HICMSM. The effect of the film holes on the modal is investigated, comparing with the solid MBDA, the modal is declined. Firstly the modal frequencies decrease after an increase with the increasing of size, and they almost do not alter with the arrangement and the angle changing, but they are impacted by the number of the film holes. The above investigations lay the foundation for the further study on the dynamic response of the MBDA and have a certain guiding significance for the design of the BDA.
