DiffPD: Differentiable Projective Dynamics

We present a novel, fast differentiable simulator for soft-body learning and control applications. Existing differentiable soft-body simulators can be classified into two categories based on their time integration methods: Simulators using explicit timestepping schemes require tiny timesteps to avoid numerical instabilities in gradient computation, and simulators using implicit time integration typically compute gradients by employing the adjoint method and solving the expensive linearized dynamics. Inspired by Projective Dynamics ( PD ), we present Differentiable Projective Dynamics ( DiffPD ), an efficient differentiable soft-body simulator based on PD with implicit time integration. The key idea in DiffPD is to speed up backpropagation by exploiting the prefactorized Cholesky decomposition in forward PD simulation. In terms of contact handling, DiffPD supports two types of contacts: a penalty-based model describing contact and friction forces and a complementarity-based model enforcing non-penetration conditions and static friction. We evaluate the performance of DiffPD and observe it is 4–19 times faster compared with the standard Newton’s method in various applications including system identification, inverse design problems, trajectory optimization, and closed-loop control. We also apply DiffPD in a reality-to-simulation ( real-to-sim ) example with contact and collisions and show its capability of reconstructing a digital twin of real-world scenes.

Friction anisotropy of violet phosphorene and its surface structure direction identification

Violet phosphorene, a recently determined semiconducting two-dimensional elemental structure, is a promising electronic and optoelectronic material. The nano-tribological properties of violet phosphorene nanoflakes are essential for their micro device applications. A friction anisotropy has been demonstrated for the violet phosphorene nanoflakes by lateral force microscope due to the sub-nanorod components of violet phosphorus. The friction forces of the violet phosphorene nanoflakes have been demonstrated to be valley along sub-nano rod direction and peak across the sub-nanorod direction with a period of 180°, resulting in a fast identification of the surface structure direction of violet phosphorene. The friction of violet phosphorene nanoflakes has also been shown to increase with increasing scanning pressure. However, it is not sensitive to scanning speed or layers. The friction of the violet phosphorene nanoflakes have also been demonstrated to increase when exposure to air for hours. The friction and adhesion features of violet phosphorene nanoflakes provide valuable foundation for violet phosphorene based devices.

Bioinspired Granular Media Friction Pad: A Universal System for Friction Enhancement on Variety of Substrates

The granular media friction pad (GMFP) inspired by the biological smooth attachment pads of cockroaches and grasshoppers employs passive jamming, to create high friction forces on a large variety of substrates. The granular medium inside the pad is encased by a flexible membrane which at contact formation greatly adapts to the substrate profile. Upon applying load, the granular medium undergoes the jamming transition and changes from fluid-like to solid-like properties. The jammed granular medium, in combination with the deformation of the encasing elastic membrane, results in high friction forces on a multitude of substrate topographies. Here we explore the effect of elasticity variation on the generation of friction by varying granular media filling quantity as well as membrane modulus and thickness. We systematically investigate contact area and robustness against substrate contamination, and we also determine friction coefficients for various loading forces and substrates. Depending on the substrate topography and loading forces, a low filling quantity and a thin, elastic membrane can be favorable, in order to generate the highest friction forces.

ON RANDOM EXPECTED VALUES VARIATIONS OF TRIBOLOGY PARAMETERS IN HUMAN HIP JOINT SURFACES

This paper presents recent progress in the knowledge concerning the stochastic theory of bio- hydrodynamic lubrication with a phospholipids bilayer. On the basis of experimental measurements and analytical solutions, the research concerns the determination of the random expectancy values of load carrying capacity, the friction coefficient, and synovial fluid dynamic variations. After numerous measurements, it directly follows that the random density function of the gap height in the human joint usually indicates a disorderly increases and decreases in the height. Such irregular gap height variations have an important influence on the random synovial bio-fluid dynamic viscosity. This finally leads to the friction coefficient and cartilage wear changes of cooperating bio- surfaces. The main topic of this paper relates to the expectancy values of the tribology parameters localized inside the variable stochastic standard deviation intervals of the human joint gap height. The results obtained finally indicate the influence of the random roughness and growth of living biological cartilage surfaces on the expectancy values of the synovial fluid dynamic viscosity, load carrying capacity and friction forces in human hip joints.

FORMATION OF QUALITATIVE INDICATORS OF PRODUCTS BY ROLLING STAMPING PROCESSES

The influence of rolling stamping on the service characteristics of products is investigated in the work. Based on the analysis of deformation kinematics, stress-strain state, microstructure and evaluation of deformability of workpiece material, ways to increase geometric accuracy, vacuum tightness, electrolytic stability and mechanical characteristics of product material, as well as ways to improve the material of workpieces for their subsequent processing. The influence of active friction forces on the nature of the material flow during SHO was established, which contributed to the development of new processes that allow to bring the shape and dimensions of the workpiece as close as possible to the finished part. Thus accuracy of the sizes of details corresponds to 7-11th qualities of accuracy, and roughness of the processed surfaces makes Ra = 2,5… 0,63 microns. The process of reshaping the square billets into round ones by the method of SHO is effective, which increases the utilization factor of the metal and reduces the anisotropy of its mechanical properties. The characteristic of flat anisotropy λr, adopted in sheet metal stamping as a characteristic of the ability of the material to form scallops, decreases as a result of reshaping by 70-80%. The relative difference in yield strength in the plane of the sheet decreases from 0.10-0.15 to 0.03-0.05. The ultimate tensile strain increases by 8-10%, and the uniform uniform strain - by 5-8%. This improvement in the characteristics of the material reshaped by rolling blanks leads to the fact that when drawing cylindrical products, the value of scallops decreases by 2-2.5 times, and the value of the maximum degree of drawing increases by 10-15%. This reduces the relative difference in wall thickness along the perimeter of the elongated workpiece, and the change in wall thickness along its height becomes linear. Thus, the use of SHO processes significantly improves the quality characteristics of products.

Defining patterns in the dynamic load and strength of the bearing structure of a covered freight car with a filler in the girder beam

This paper reports a study into determining the dynamic load and strength of the bearing structure of a covered freight car under operational modes. A feature of the freight car's bearing structure is that the girder beam has a closed cross-section. To reduce the dynamic load of the frame, the girder beam is filled with a material with viscoelastic properties. Such a solution could contribute to the transformation of the kinetic energy of impact (due to jerk, stretching, compression) into work of viscoelastic friction forces, and, consequently, to reducing the load on the bearing structure. To substantiate the proposed improvement, the dynamic load on the bearing structure of a covered freight car was mathematically modeled. The calculation was performed for the case of joint impacts at shunting. The study was carried out in a flat coordinate system. It was established that the maximum accelerations acting on the bearing structure of a covered freight car were about 37 m/s2. The calculated acceleration value is 3.2 % lower than that obtained for the bearing structure of a covered freight car without filler. The results of calculating the strength of the load-bearing structure of a covered freight car are given. In this case, a finite-element method was applied. The maximum equivalent stresses occur in the zones of interaction between the girder beam and the pivot beams, and amount to 319.5 MPa, which is 8 % lower than permissible. The calculation was also performed regarding other operational modes of loading the freight car's bearing structure. The model of the dynamic load on the bearing structure of a covered freight car was verified according to the F-criterion. The research reported here could contribute to designing innovative rolling stock structures, thereby improving the efficiency of their operation.

Designing an outer toothed gear whose wheel teeth are outlined by the logarithmic spiral arcs

Toothed gears are the most common mechanical gears in machine building, which are characterized by high reliability and durability, a constant transfer number, and which can transmit high torque. During toothed gear operation, the surfaces of the teeth slide, which gives rise to friction forces and wears their working surfaces. To prevent this, the surfaces of the teeth need constant lubrication. This paper considers the design of a gear tooth engagement, which does not have friction between the surfaces of the teeth since they roll over each other without slipping. The profile of the tooth of such a gear is outlined by congruent arcs, symmetrical relative to the line that connects the center of rotation of the toothed wheel with the top of the tooth. These symmetrical curves at the top of the tooth intersect at the predefined angle. In the depressions of the wheel, adjacent teeth also intersect at the same angle. Such a condition can be ensured by a curve that at all its points crosses the radius-vector emanating from the coordinate origin, also at a stable angle equal to half of the given one. This curve is a logarithmic spiral. If the number of teeth of the drive and driven wheels is the same, then their teeth are congruent. Otherwise, the profiles of the teeth would differ but they could be outlined by congruent arcs of the same logarithmic spiral of the same length taken from different areas of the curve. The minimum possible angle at the top of the teeth is straight. At acute angle, the toothed gear operation is impossible. To build gear wheels with a right angle at the top of the tooth, it would suffice to set the number of teeth of the drive and driven wheels. The center-to-center distance is calculated using the derived formula. The transfer number of such a gear is variable but, with an increase in the number of teeth, the range of its change decreases. The algorithm of wheel construction is given.

Comparative Analysis of Machine Learning Methods for Predicting Robotized Incremental Metal Sheet Forming Force

This paper proposes a method for extracting information from the parameters of a single point incremental forming (SPIF) process. The measurement of the forming force using this technology helps to avoid failures, identify optimal processes, and to implement routine control. Since forming forces are also dependent on the friction between the tool and the sheet metal, an innovative solution has been proposed to actively control the friction forces by modulating the vibrations that replace the environmentally unfriendly lubrication of contact surfaces. This study focuses on the influence of mechanical properties, process parameters and sheet thickness on the maximum forming force. Artificial Neural Network (ANN) and different machine learning (ML) algorithms have been applied to develop an efficient force prediction model. The predicted forces agreed reasonably well with the experimental results. Assuming that the variability of each input function is characterized by a normal distribution, sampling data were generated. The applicability of the models in an industrial environment is due to their relatively high performance and the ability to balance model bias and variance. The results indicate that ANN and Gaussian process regression (GPR) have been identified as the most efficient methods for developing forming force prediction models.

Flexure based gripper to grasp hollow objects by internal surface interaction

The paper focuses on the design, development, and evaluation of a gripper intended to hold hollow objects by interacting with the inner surface. The gripper moves towards the inside of the hollow object and grips it using the friction forces applied on the surface of the object. The design also ensures the application of variable normal forces on the surface of the object to be grasped. The mathematical architecture is verified using prototypes and experiments.

FEM-based comparative study of SQUARE & RHOMBIC INSERT machining performance during turning of AISI-D3 steel

Nowadays, employment of the Finite Element Method (FEM) in machining simulation is a common practice to decrease development times and costs, as well as to investigate numerous parameters that affect machining processes. In the present work, the 3D modelling of AISI-D3 hard turning with both square and rhombic inserts is being presented by utilizing a commercially available Finite Element Analysis (FEA) software. Eighteen tests were carried out based on cutting conditions that are recommended for the used tools. Specifically, three levels of cutting speed (75m/min, 110m/min and 140m/min), three levels of feed (0.12mm/rev, 0.16mm/rev and 0.20mm/rev) and depth of cut equal to 0.40mm for all tests, were applied. In order to describe the complex factors that define the model, such as the friction forces, the heat transfer and the pressure due to contact between the tool and the workpiece, a number of acknowledged models were utilized. A comparison of the performance between the two types of tools was made with respect to the developed machining forces and temperature distribution on the workpiece. The findings of the investigation indicate that the specific square tools produce higher values of forces compared to the rhombic ones and approximately the same temperature patterns on the workpiece. The average increase on the produced cutting forces is about 26.4%.