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.

Prediction models establishment and comparison for guiding force of high-temperature superconducting maglev based on deep learning algorithms

By the merits of self-stability and low energy consumption, high temperature superconducting (HTS) maglev has the potential to become a novel type of transportation mode. As a key index to guarantee the lateral self-stability of HTS maglev, guiding force has strong non-linearity and is determined by multitudinous factors, and these complexities impede its further researches. Compared to traditional finite element and polynomial fitting method, the prosperity of deep learning algorithms could provide another guiding force prediction approach, but the verification of this approach is still blank. Therefore, this paper establishes 5 different neural network models (RBF, DNN, CNN, RNN, LSTM) to predict HTS maglev guiding force, and compares their prediction efficiency based on 3720 pieces of collected data. Meanwhile, two adaptively iterative algorithms for parameters matrix and learning rate adjustment are proposed, which could effectively reduce computing time and unnecessary iterations. And according to the results, it is revealed that, the DNN model shows the best fitting goodness, while the LSTM model displays the smoothest fitting curve on guiding force prediction. Based on this discovery, the effects of learning rate and iterations on prediction accuracy of the constructed DNN model are studied. And the learning rate and iterations at the highest guiding force prediction accuracy are 0.00025 and 90000, respectively. Moreover, the K-fold cross validation method is also applied to this DNN model, whose result manifests the generalization and robustness of this DNN model. The imperative of K-fold cross validation method to ensure universality of guiding force prediction model is likewise assessed. This paper firstly combines HTS maglev guiding force prediction with deep learning algorithms considering different field cooling height, real-time magnetic flux density, liquid nitrogen temperature and motion direction of bulk. Additionally, this paper gives a convenient and efficient method for HTS guiding force prediction and parameter optimization.

Modeling and investigation of cutting force for SiCp/Al composites during ultrasonic vibration-assisted turning

The machining process of SiCp/Al composites is considerably difficult because of the addition of ceramic particles. As an effective machining method, ultrasonic vibration-assisted turning is used to process SiCp/Al composites, which can effectively reduce the cutting force, improve the surface quality, and reduce the tool wear. This study developed a cutting force prediction model for ultrasonic vibration-assisted turning of SiCp/Al composites, which comprehensively considers the instantaneous depth of cut and the instantaneous shear angle. This model divides the cutting force into the chip formation force considering the instantaneous depth of cut, the friction force considering the influence of SiC particles at tool-chip interface, the particle fracture force, and the ultrasonic impact force in the cutting depth direction. By comparing the predicted value of the main cutting force with the experimental values, the results present the same trend, which verifies the feasibility of the cutting force prediction model. In addition, the influence of vibration amplitude, depth of cut, and cutting speed on the main cutting force is analyzed. The systematic cutting experiments show that ultrasonic vibration-assisted turning can significantly reduce the cutting force and improve the machinability of SiCp/Al composites.

Profile Design of the Grooved Die and Rolling Force Prediction in the Cold Pilger Rolling Process

The objective of this study was to design the die groove profile and predict the rolling force produced when employing the variable curvature rolls and mandrel for manufacturing seamless pipes using the cold pilger rolling process. The parameters of the key process design were the diameter of the initial tube and final product, as well as the feed amount, reduction area, principal deformation zone, and roller radius. The rolling forces during the pilger rolling process were theoretically calculated to enable their prediction, and the characteristics of the cold pilger rolling process were identified. The calculated values were in close agreement with the experimental data. The die groove design is important in the prediction process because the dimensional accuracy of the tubes and the life of the dies are highly dependent on this design. The presented design method can be successfully applied to fulfill this objective. The tube shape and adequate tolerance can be attained by using the proposed design method. The mechanical properties of the pipe are evaluated by calculating the Q factor.

Cutting Force Prediction Model for Elliptical Vibration Cutting SiCp/Al Based on Three-Phase Friction Theory

The friction behavior in the tool-chip interface is an essential issue in aluminum matrix composite material (AMCM) turning operations. Compared with conventional cutting, the elliptical vibration (EVC) cutting AMCM has attractive advantages, such as low friction, small cutting forces, etc. However, the friction mechanism of the EVC cutting AMCM is still inadequate, especially the model for cutting forces analyzing and predicting, which hinders the application of EVC in the processing of AMCM. In this paper, a cutting force prediction model for EVC cutting SiCp/Al is established, which is based on the three-phase friction (TPF) theory. The friction components are evaluated and predicted at the tool-chip interface (TCI), tool-particle interface (TPI) and tool-matrix (TMI), respectively. In addition, the tool-chip contact length and SiC particle volume fraction were defined strictly and the coefficient of friction was predicted. Based on the Johnson-Cook constitutive model, the experiment was conducted on SiCp/Al. The cutting speed and tool-chip contact length were used as input parameters of the friction model, and the dynamic changes of cutting force and stress distribution were analyzed. The results shown that when cutting speed reaches 574 m/min, the tool-chip contact length decreases to 0.378 mm. When the cutting speed exceeds 658 m/min, the cutting force decreases to a minimum of 214.9 N and remains stable. In addition, compared with conventional cutting, the proposed prediction model can effectively reduce the cutting force.

Modeling Analysis and 3D Force Prediction of a Novel Piezoelectric Tactile Sensor

To measure three-dimensional (3D) forces efficiently and improve the sensitivity of tactile sensors, a novel piezoelectric tactile sensor with a “sandwich” structure is proposed in this paper. An array of circular truncated cone-shaped sensitive units made of polyvinylidene fluoride (PVDF) is sandwiched between two flexible substrates of polydimethylsiloxane (PDMS). Based on the piezoelectric properties of the PVD F sensitive units, finite element modelling and analysis are carried out for the sensor. The relation between the force and the voltage of the sensitive unit is obtained, and a tactile perception model is established. The model can distinguish the sliding direction and identify the material of the slider loaded on the sensor. A backpropagation neural network (BPNN) algorithm is built to predict the 3D forces applied on the tactile sensor model, and the 3D forces are decoupled from the voltages of the sensitive units. The BPNN is further optimized by a genetic algorithm (GA) to improve the accuracy of the 3D force prediction, and fairly good prediction results are obtained. The experimental results show that the novel tactile sensor model can effectively predict the 3D forces, and the BPNN model optimized by the GA can predict the 3D forces with much higher precision, which also improves the intelligence of the sensor. All the prediction results indicate that the BPNN algorithm has very efficient performance in 3D force prediction for the piezoelectric tactile sensor.

Cutting Force Modeling and Experimental Study for Ball-End Milling of Free-Form Surfaces

In order to improve the machining quality and efficiency and optimize NC machining programming, based on the existing cutting force models for ball-end, a cutting force prediction model of free-form surface for ball-end was established. By analyzing the force of the system during the cutting process, we obtained the expression equation of the instantaneous undeformed chip thickness during the milling process and then determined the rule of the influence of the lead angle and the tilt angle on the instantaneous undeformed chip thickness. It was judged whether the cutter edge microelement is involved in cutting, and the algorithm flow chart is given. After that, the cutting force prediction model of free-form surface for ball-end and pseudocodes for cutting force prediction were given. MATLAB was used to simulate the prediction force model. Finally, through the comparative analysis experiment of the measured cutting force and the simulated cutting force, the experimental results are basically consistent with the theoretical prediction results, which proves that the model established in this paper can accurately predict the change of the cutting force of the ball-end cutter in the process of milling free-form surface, and the error of the cutting force prediction model established in this paper is reduced by 15% compared with the traditional cutting force prediction model.

PIV-Based Hydrodynamic Analysis of a 2D Lifting Body Using Near-Field Flow Kinematics

The ability to extract quantitative flow information from photographic images of the velocity field using Particle Image Velocimetry (PIV) is a powerful alternative to the more traditional invasive or integrated method techniques. The usage of PIV allows the complete characterization of the flow field, and not just at discrete points. Additionally, with PIV, it is possible to predict the hydrodynamic characteristics of a lifting body without measuring the forces and moments acting upon it. In this paper, the hydrodynamic performance of a NACA 0018 airfoil was determined by analyzing the flow kinematics from a 2D-2C (two-dimensional, two-component) PIV data set. The motivation for this work was to provide a canonical study to show that laser optical measurement techniques such as PIV, can be an attractive alternative to dynamic force testing. The hydrodynamic performance evaluated using PIV data was compared to the computational program XFOIL to assess the validity of the results. The analytical drag force prediction was carried out using the Von Kármán Momentum Integral approach for a flat plate and the Squire-Young boundary layer method as an improved method, whereas the analytical lift force prediction calculation was based on the Kutta-Joukowski theorem. The results show reasonable agreement with the numerical prediction tool XFOIL and they follow the expected trends across all operating conditions. These findings suggest that this methodology might be expanded to conduct hydrodynamic analyses on more complex geometries such as hydrofoils, turbines, propulsors, fin stabilizers, rudders, and other control surfaces using flow kinematics data from PIV.