Bryce Cianciotto, Derek Price, Logan Spencer, Martin Garcia, A. Tekes
This study presents the design and development of a tendon-driven soft gripper manipulated by a 4 DOF robotic arm. The proposed robotic arm and gripper explore new areas focusing on increasing the grasping performance of the gripper as well as the workspace. The gripper is designed with 3 fingers and driven by tendons using two servo motors. The tension of the strings is adjusted using a pulley mechanism and a string. The opening and grasping of the soft gripper are accomplished by each motor. The wide opening allows the gripper to grasp larger objects. The parallel robotic arm motion is actuated using 4 motors. These motors are mounted on a spherical shoulder plate with attached circular plates with angled axles are. The axles are angled so that their axes of rotation converge to the center point of the shoulder plate. The vertical and lateral motion of the robotic arm is controlled by the series of radial linkages connected to the motors, with a parallel linkage attached to the radial linkages to actuate the forearm of the mechanism. The robotic arm is 3D printed in polylactic acid (PLA) and the monolithic soft gripper is 3D printed in thermoplastic polyurethane (TPU). The gripping force applied by the gripper is obtained using flexible sensors attached to the tip of the 3 fingers. The finite element analysis is performed in SoldWorks and the link lengths are optimized to trace the desired workspace. The mechanism is tested for its grasping and lifting of various objects showing promising superiorities in terms of its grasping capabilities mimicking the human hand. If the robotic arm is mounted on a moving platform, then it can serve as an assistive robot for the elderly.
{"title":"Design and Development of a Novel Soft Gripper Manipulated by a Robotic Arm","authors":"Bryce Cianciotto, Derek Price, Logan Spencer, Martin Garcia, A. Tekes","doi":"10.1115/imece2021-69880","DOIUrl":"https://doi.org/10.1115/imece2021-69880","url":null,"abstract":"\u0000 This study presents the design and development of a tendon-driven soft gripper manipulated by a 4 DOF robotic arm. The proposed robotic arm and gripper explore new areas focusing on increasing the grasping performance of the gripper as well as the workspace. The gripper is designed with 3 fingers and driven by tendons using two servo motors. The tension of the strings is adjusted using a pulley mechanism and a string. The opening and grasping of the soft gripper are accomplished by each motor. The wide opening allows the gripper to grasp larger objects. The parallel robotic arm motion is actuated using 4 motors. These motors are mounted on a spherical shoulder plate with attached circular plates with angled axles are. The axles are angled so that their axes of rotation converge to the center point of the shoulder plate. The vertical and lateral motion of the robotic arm is controlled by the series of radial linkages connected to the motors, with a parallel linkage attached to the radial linkages to actuate the forearm of the mechanism. The robotic arm is 3D printed in polylactic acid (PLA) and the monolithic soft gripper is 3D printed in thermoplastic polyurethane (TPU). The gripping force applied by the gripper is obtained using flexible sensors attached to the tip of the 3 fingers. The finite element analysis is performed in SoldWorks and the link lengths are optimized to trace the desired workspace. The mechanism is tested for its grasping and lifting of various objects showing promising superiorities in terms of its grasping capabilities mimicking the human hand. If the robotic arm is mounted on a moving platform, then it can serve as an assistive robot for the elderly.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84173360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Inspired by the muscle structures of elephant’s trunk, octopus’ arm and tongue, the soft continuous arm developed based on bionics can extend, contract and bend along any point of its structure. Compared with the rigid manipulator, it has the characteristics of high flexibility, complex environment adaptability and safe human-machine interactivity. In this work, a three-dimensional (3D) kinematic and dynamic model of the soft continuous arm based on the modal method is presented and analyzed. The spatial kinematics model of the soft continuous arm is established based on the modal method first, in which the deformation of the soft continuous arm is assumed to be an arc without torsion, and the complex nonlinear function is approximated by a simple mathematical function. The accurate, singularity free and unique modal transformation matrix can be deduced conveniently. Also, it overcomes the limitations of the previous models in which the curve parameters are used and the curvature of the cross section assumes to be not zero as well as the dynamic model does not include the pure linear motion. The modal method can be used to simulate the spatial bending motion, but also effectively overcome the singularity problem caused by using the curve parameters for simulating the pure linear motion. Then, the dynamics model of the soft continuous arm is established using the Lagrange method. Due to large number of complex nonlinear integral operations is inevitably involved, and the computational efficiency is low in the numerical solution, the modal method is used in the numerical simulation, which greatly simplifies the complex nonlinear integral operation and improves the efficiency of numerical calculation. The simulation results of the pure elongation and bending of the soft continuous arm show that the modal method has the characteristics of fast and accurate response.
{"title":"Kinematic and Dynamic Modelling and Simulation of Soft Continuous Arm Based on Modal Method","authors":"Qingfeng Kong, Z. Bai","doi":"10.1115/imece2021-70300","DOIUrl":"https://doi.org/10.1115/imece2021-70300","url":null,"abstract":"\u0000 Inspired by the muscle structures of elephant’s trunk, octopus’ arm and tongue, the soft continuous arm developed based on bionics can extend, contract and bend along any point of its structure. Compared with the rigid manipulator, it has the characteristics of high flexibility, complex environment adaptability and safe human-machine interactivity. In this work, a three-dimensional (3D) kinematic and dynamic model of the soft continuous arm based on the modal method is presented and analyzed. The spatial kinematics model of the soft continuous arm is established based on the modal method first, in which the deformation of the soft continuous arm is assumed to be an arc without torsion, and the complex nonlinear function is approximated by a simple mathematical function. The accurate, singularity free and unique modal transformation matrix can be deduced conveniently. Also, it overcomes the limitations of the previous models in which the curve parameters are used and the curvature of the cross section assumes to be not zero as well as the dynamic model does not include the pure linear motion. The modal method can be used to simulate the spatial bending motion, but also effectively overcome the singularity problem caused by using the curve parameters for simulating the pure linear motion. Then, the dynamics model of the soft continuous arm is established using the Lagrange method. Due to large number of complex nonlinear integral operations is inevitably involved, and the computational efficiency is low in the numerical solution, the modal method is used in the numerical simulation, which greatly simplifies the complex nonlinear integral operation and improves the efficiency of numerical calculation. The simulation results of the pure elongation and bending of the soft continuous arm show that the modal method has the characteristics of fast and accurate response.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84508708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Practically all marine vessels have fixed-geometry hulls. This limits their capabilities and high-performance regimes to a limited set of operational conditions. Having a transformable or adaptive hull structure can help maximize ship’s operational performance for various scenarios. In this work, a transformable concept boat is conceived that can change its configuration from monohull to twin-hulled configuration. A catamaran is desirable for carrying volumetric cargo or creating a large deck space that can serve, for example, as a launch pad for aircraft, while more compact monohulls can be more easily stored or operated in restricted environments. A monohull and a catamaran also have different stability, hydrodynamic, maneuvering and seakeeping characteristics. In the present effort, a small-scale model boat has been constructed with two hulls that can be brought together or separated using an expansion mechanism driven by a servo motor. This model setup has been equipped with propulsors, batteries, and control and communication modules for radio-controlled operations. In addition, a remote data acquisition system was assembled for measuring boat’s kinematic and powering characteristics. Results of initial tests with the small-scale transformable boat in an open water reservoir are reported and discussed in this paper.
{"title":"Construction and Testing of Small-Scale Transformable-Hull Concept Boat","authors":"Phillip R. Whitworth, Cole James, K. Matveev","doi":"10.1115/imece2021-69563","DOIUrl":"https://doi.org/10.1115/imece2021-69563","url":null,"abstract":"\u0000 Practically all marine vessels have fixed-geometry hulls. This limits their capabilities and high-performance regimes to a limited set of operational conditions. Having a transformable or adaptive hull structure can help maximize ship’s operational performance for various scenarios. In this work, a transformable concept boat is conceived that can change its configuration from monohull to twin-hulled configuration. A catamaran is desirable for carrying volumetric cargo or creating a large deck space that can serve, for example, as a launch pad for aircraft, while more compact monohulls can be more easily stored or operated in restricted environments. A monohull and a catamaran also have different stability, hydrodynamic, maneuvering and seakeeping characteristics. In the present effort, a small-scale model boat has been constructed with two hulls that can be brought together or separated using an expansion mechanism driven by a servo motor. This model setup has been equipped with propulsors, batteries, and control and communication modules for radio-controlled operations. In addition, a remote data acquisition system was assembled for measuring boat’s kinematic and powering characteristics. Results of initial tests with the small-scale transformable boat in an open water reservoir are reported and discussed in this paper.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77740479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a new ultra-compliant and flexible structural sensing neural system for damage detection in wind turbine blades. The entire sensing cluster is integrated inside a flexible printed circuit (FPC), which complies the structural geometric features as a neural skin planted on the structural surfaces. It is worth of noting that the proposed sensing system can be mounted on arbitrary locations of a wind turbine blade, mimicking the neurons of the human biological system to detect and monitor the damage in the blade. Besides, all the sensing elements are interfaced with a laptop-controlled data acquisition board to receive the structural dynamic responses. In addition, a miniature electromagnetic actuator is utilized to excite the blade. Additional mass is employed to simulate the damage situation. Several signal processing techniques are implemented to analyze the oscillatory responses, such as wavelet analysis, fast Fourier transform (FFT), and short time Fourier transform (STFT). Furthermore, the damage indices (DIs) which correlate the structural spectral power density of both pristine and damaged cases can accurately identify the location of damage. Such a sensing neural system possesses the tremendous potential in future Structural Health Monitoring (SHM) and Nondestructive Evaluation (NDE) applications. This paper finishes with discussion, concluding remarks, and suggestions for future work.
{"title":"An Ultra-Compliant and Flexible Structural Sensing Neural System for Damage Detection in Wind Turbine Blades","authors":"Junzhen Wang, Yanfeng Shen, Xue Peng, Zhengyang Han, Shaopeng Jiang","doi":"10.1115/imece2021-70986","DOIUrl":"https://doi.org/10.1115/imece2021-70986","url":null,"abstract":"\u0000 This paper presents a new ultra-compliant and flexible structural sensing neural system for damage detection in wind turbine blades. The entire sensing cluster is integrated inside a flexible printed circuit (FPC), which complies the structural geometric features as a neural skin planted on the structural surfaces. It is worth of noting that the proposed sensing system can be mounted on arbitrary locations of a wind turbine blade, mimicking the neurons of the human biological system to detect and monitor the damage in the blade. Besides, all the sensing elements are interfaced with a laptop-controlled data acquisition board to receive the structural dynamic responses. In addition, a miniature electromagnetic actuator is utilized to excite the blade. Additional mass is employed to simulate the damage situation. Several signal processing techniques are implemented to analyze the oscillatory responses, such as wavelet analysis, fast Fourier transform (FFT), and short time Fourier transform (STFT). Furthermore, the damage indices (DIs) which correlate the structural spectral power density of both pristine and damaged cases can accurately identify the location of damage. Such a sensing neural system possesses the tremendous potential in future Structural Health Monitoring (SHM) and Nondestructive Evaluation (NDE) applications. This paper finishes with discussion, concluding remarks, and suggestions for future work.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82463206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The behavior of fluid flow in differentially rotating shells with radii a and b where a < b, is discussed in the magnetic field’s presence. Out of the two significant elements of the magnetic field which are toroidal and poloidal, the toroidal magnetic field plays the dominant role for determining the behavior of fluid flow in the case of the rotating bodies. The solution for modelled equations is observed by perturbation method. Also the solution for flow quantities are observed by variable separable method for specific value of constant used in the abovementioned method.
{"title":"Fluid Flow in Between the Differentially Rotating Spherical Shells in the Presence of Toroidal Magnetic Field","authors":"B. Sharma, N. Srivastava","doi":"10.1115/imece2021-66692","DOIUrl":"https://doi.org/10.1115/imece2021-66692","url":null,"abstract":"\u0000 The behavior of fluid flow in differentially rotating shells with radii a and b where a < b, is discussed in the magnetic field’s presence. Out of the two significant elements of the magnetic field which are toroidal and poloidal, the toroidal magnetic field plays the dominant role for determining the behavior of fluid flow in the case of the rotating bodies. The solution for modelled equations is observed by perturbation method. Also the solution for flow quantities are observed by variable separable method for specific value of constant used in the abovementioned method.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83915339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study proposes a novel model-free vibration controller based on a virtual controlled object (VCO) considering actuator parameter uncertainty. A proof-mass actuator, which is modeled as a single-degree-of-freedom (SDOF) system, is employed. A VCO, which is defined as an SDOF system, is inserted between the actual controlled object and the actuator model. Considering frequency transfer characteristic from actual controlled object to VCO, setting appropriate parameters of the VCO realizes model-free control. A state equation to design the model-free controller is derived based on the two-degree-of-freedom (2DOF) system composed of the actuator model and the VCO. The actuator parameter uncertainty is quantitatively modeled in the 2DOF plant. Traditional mixed H2/H∞ control theory is applied for the uncertain plant to design a model-free controller with high damping performance and robustness to the actuator uncertainty. The effectiveness of the proposed controller is confirmed by vibration control experiments.
{"title":"Experimental Verification of Model-Free Vibration Control Technique Based on a Virtual Controlled Object Considering Actuator Parameter Uncertainty","authors":"Ansei Yonezawa, Heisei Yonezawa, I. Kajiwara","doi":"10.1115/imece2021-69100","DOIUrl":"https://doi.org/10.1115/imece2021-69100","url":null,"abstract":"\u0000 This study proposes a novel model-free vibration controller based on a virtual controlled object (VCO) considering actuator parameter uncertainty. A proof-mass actuator, which is modeled as a single-degree-of-freedom (SDOF) system, is employed. A VCO, which is defined as an SDOF system, is inserted between the actual controlled object and the actuator model. Considering frequency transfer characteristic from actual controlled object to VCO, setting appropriate parameters of the VCO realizes model-free control. A state equation to design the model-free controller is derived based on the two-degree-of-freedom (2DOF) system composed of the actuator model and the VCO. The actuator parameter uncertainty is quantitatively modeled in the 2DOF plant. Traditional mixed H2/H∞ control theory is applied for the uncertain plant to design a model-free controller with high damping performance and robustness to the actuator uncertainty. The effectiveness of the proposed controller is confirmed by vibration control experiments.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87260605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The lateral-directional stability of four-wheeled ground vehicles with understeer and oversteer characteristics in off-road conditions were investigated using a locally linearized vehicle dynamics model with two degrees-of-freedom. Analytical results confirm the widely accepted understanding that the open-loop response for an understeering vehicle with the steered wheels fixed parallel to the vehicle centerline is stable for all vehicle speeds, while an oversteering vehicle is unstable above a certain ‘critical speed’ for the same fixed steering condition. However, results for both understeering and oversteering vehicles using a path-centered steer angle input, defined relative to the path tangent, were found to be stable for all forward speed conditions. These results also indicated that the path-centered steering can be used for lateral-directional control with suitable gain. Therefore these results indicate that a manual control strategy based on path-centered steer angle inputs can be used to stably control a vehicle for all forward speeds, in both understeering and oversteering conditions, provided the tire cornering force continues to increase with slip angle, which is generally the case for vehicle off-road or snow conditions. These results are also consistent with ground vehicle operator acceptance and preference of vehicles with oversteering characteristics in off-road conditions.
{"title":"Lateral-Directional Stability and Manual Control of Understeering and Oversteering Vehicles in Off-Road Conditions Based on a 2-DOF Cornering Compliance Vehicle Dynamics Model","authors":"R. M. Van Auken, S. A. Kebschull","doi":"10.1115/imece2021-71854","DOIUrl":"https://doi.org/10.1115/imece2021-71854","url":null,"abstract":"\u0000 The lateral-directional stability of four-wheeled ground vehicles with understeer and oversteer characteristics in off-road conditions were investigated using a locally linearized vehicle dynamics model with two degrees-of-freedom. Analytical results confirm the widely accepted understanding that the open-loop response for an understeering vehicle with the steered wheels fixed parallel to the vehicle centerline is stable for all vehicle speeds, while an oversteering vehicle is unstable above a certain ‘critical speed’ for the same fixed steering condition. However, results for both understeering and oversteering vehicles using a path-centered steer angle input, defined relative to the path tangent, were found to be stable for all forward speed conditions. These results also indicated that the path-centered steering can be used for lateral-directional control with suitable gain. Therefore these results indicate that a manual control strategy based on path-centered steer angle inputs can be used to stably control a vehicle for all forward speeds, in both understeering and oversteering conditions, provided the tire cornering force continues to increase with slip angle, which is generally the case for vehicle off-road or snow conditions. These results are also consistent with ground vehicle operator acceptance and preference of vehicles with oversteering characteristics in off-road conditions.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91502439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Two-dimensional numerical simulations for heated and unheated flows over two circular cylinders of different diameters in tandem are conducted. The study is conducted for Re = 150 and temperature ratio 1 ≤ (T* = Ts/T∞) ≤ 2.3. The study is limited to spacing ratios in the range 2 ≤ L/Dd ≤ 8 and diameter ratios 0.5 ≤ Du/Dd ≤ 2. The effect of heating on the flow is captured by modelling the thermophysical properties of the fluid fitted as polynomial functions of temperature. The flow features are visualized by plotting the streamlines and velocity magnitude contours. These flow patterns are also discussed based on the reattachment and co-shedding regimes which are classification found in literature. The effect of varying geometrical parameters on the flow in terms of change of regime is discussed. Furthermore, the drag coefficient (Cd) on both upstream and downstream cylinders is presented as a function of spacing and diameter ratios. Additionally the average surface heat transfer coefficient (h) is presented for the interval of spacing and diameter ratios stated. Results show that the presence of an upstream cylinder led to reduction of Cd on the downstream cylinder to a value lower than that of single cylinder for all cases studied. Heating both cylinders causes Cd on both cylinders to become less sensitive to a change in L/Dd, and increasing Du/Dd is found to decrease h on both upstream and downstream cylinders.
{"title":"Dynamics of Flow and Heat Transfer Around Two Circular Cylinders of Different Diameters in Tandem Subjected to Forced Convection","authors":"R. Homsi, Md. Islam, Yap Yit Fatt, I. Janajreh","doi":"10.1115/imece2021-72944","DOIUrl":"https://doi.org/10.1115/imece2021-72944","url":null,"abstract":"\u0000 Two-dimensional numerical simulations for heated and unheated flows over two circular cylinders of different diameters in tandem are conducted. The study is conducted for Re = 150 and temperature ratio 1 ≤ (T* = Ts/T∞) ≤ 2.3. The study is limited to spacing ratios in the range 2 ≤ L/Dd ≤ 8 and diameter ratios 0.5 ≤ Du/Dd ≤ 2. The effect of heating on the flow is captured by modelling the thermophysical properties of the fluid fitted as polynomial functions of temperature. The flow features are visualized by plotting the streamlines and velocity magnitude contours. These flow patterns are also discussed based on the reattachment and co-shedding regimes which are classification found in literature. The effect of varying geometrical parameters on the flow in terms of change of regime is discussed. Furthermore, the drag coefficient (Cd) on both upstream and downstream cylinders is presented as a function of spacing and diameter ratios. Additionally the average surface heat transfer coefficient (h) is presented for the interval of spacing and diameter ratios stated. Results show that the presence of an upstream cylinder led to reduction of Cd on the downstream cylinder to a value lower than that of single cylinder for all cases studied. Heating both cylinders causes Cd on both cylinders to become less sensitive to a change in L/Dd, and increasing Du/Dd is found to decrease h on both upstream and downstream cylinders.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81930812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rolling element bearings are critical components regarding the reliability and safety of rotating machinery. A reliable and continuous monitoring system with high prediction accuracy prevents machine downtime, increases productivity, and reduces maintenance costs. Vibration analysis via machine learning tools is a well-established approach. In recent years, deep learning methods have received increasing attention from researchers and engineers because of their inherent capability of dealing with big data, mining complex representations, and overcoming the disadvantage of traditional fault classification and feature selection algorithms based on hand-crafted features. However, the literature lacks a well-structured set of rules and comprehensive evaluation of the existing methods and resources and it is not clear how to choose the best algorithm for certain situations to achieve the optimal outcome. This work evaluates traditional machine learning and recent deep learning-based fault classification methods based on two benchmark rolling element bearing datasets and provides a comprehensive evaluation of the methods. Both time-frequency domain statistical features and raw inputs were used. The comparisons were made based on classification accuracy, training time, and hyperparameter tuning, Based on the evaluation results, we discuss technical challenges and provide suggestions for method selection and improvement.
{"title":"An Empirical Study of Machine Learning and Deep Learning Algorithms on Bearing Fault Diagnosis Benchmarks","authors":"Behnoush Rezaeianjouybari, Y. Shang","doi":"10.1115/imece2021-69994","DOIUrl":"https://doi.org/10.1115/imece2021-69994","url":null,"abstract":"\u0000 Rolling element bearings are critical components regarding the reliability and safety of rotating machinery. A reliable and continuous monitoring system with high prediction accuracy prevents machine downtime, increases productivity, and reduces maintenance costs. Vibration analysis via machine learning tools is a well-established approach. In recent years, deep learning methods have received increasing attention from researchers and engineers because of their inherent capability of dealing with big data, mining complex representations, and overcoming the disadvantage of traditional fault classification and feature selection algorithms based on hand-crafted features. However, the literature lacks a well-structured set of rules and comprehensive evaluation of the existing methods and resources and it is not clear how to choose the best algorithm for certain situations to achieve the optimal outcome. This work evaluates traditional machine learning and recent deep learning-based fault classification methods based on two benchmark rolling element bearing datasets and provides a comprehensive evaluation of the methods. Both time-frequency domain statistical features and raw inputs were used. The comparisons were made based on classification accuracy, training time, and hyperparameter tuning, Based on the evaluation results, we discuss technical challenges and provide suggestions for method selection and improvement.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76142299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The converse flexoelectric effect has been applied to precision actuation and vibration control of flexible structures. High stress concentration caused by a single flexoelectric actuator can be alleviated by placing multiple actuators on the structure. In the presented work, a neural network model was established to optimize the positions of multiple flexoelectric actuators on a cantilever beam. It was proved that the neural network can recognize the relationship between actuator position and tip displacement and forecast the tip displacement of the beam accurately with reduced computational effort and higher effectiveness. By using the neural network, the displacement data generated by all possible combinations of actuator positions were predicted, and the optimal positions of multiple flexoelectric actuators can be obtained in term of maximum transverse tip displacement. The effect of actuator size with various actuator numbers was discussed. The results showed that applied voltage can be reduced by increasing the number of flexoelectric actuators when placed on optimal positions.
{"title":"An Artificial Neural Network Model for Flexoelectric Actuation and Control of Beams","authors":"Pengcheng Yu, Xiaogang Fu, M. Fan","doi":"10.1115/imece2021-69392","DOIUrl":"https://doi.org/10.1115/imece2021-69392","url":null,"abstract":"\u0000 The converse flexoelectric effect has been applied to precision actuation and vibration control of flexible structures. High stress concentration caused by a single flexoelectric actuator can be alleviated by placing multiple actuators on the structure. In the presented work, a neural network model was established to optimize the positions of multiple flexoelectric actuators on a cantilever beam. It was proved that the neural network can recognize the relationship between actuator position and tip displacement and forecast the tip displacement of the beam accurately with reduced computational effort and higher effectiveness. By using the neural network, the displacement data generated by all possible combinations of actuator positions were predicted, and the optimal positions of multiple flexoelectric actuators can be obtained in term of maximum transverse tip displacement. The effect of actuator size with various actuator numbers was discussed. The results showed that applied voltage can be reduced by increasing the number of flexoelectric actuators when placed on optimal positions.","PeriodicalId":23585,"journal":{"name":"Volume 7A: Dynamics, Vibration, and Control","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78235542","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：