Using artificial neural networks for the prediction of dimensional error on inclined surfaces manufactured by ball-end milling

Arnaiz‐González, Álvar; Fernández-Valdivielso, A.; Bustillo, Andrés; Lacalle, Luís Norberto López de

doi:10.1007/s00170-015-7543-y

Cited by 86 publications

(48 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, in order to improve the compensation of latent factors (e.g., operational conditions, temperature, etc. ), non-linear approaches can be used involving, for example, neural networks (as found in [28,29]). Nevertheless, the Mahalanobis distance can be a good starting point as it proved to perform quite well with variable operating conditions (when their effect is to produce quasi linear relations of the features).…”

Section: Discussionmentioning

confidence: 99%

Machine Vibration Monitoring for Diagnostics through Hypothesis Testing

Daga

Garibaldi

2019

Information

View full text Add to dashboard Cite

Nowadays, the subject of machine diagnostics is gathering growing interest in the research field as switching from a programmed to a preventive maintenance regime based on the real health conditions (i.e., condition-based maintenance) can lead to great advantages both in terms of safety and costs. Nondestructive tests monitoring the state of health are fundamental for this purpose. An effective form of condition monitoring is that based on vibration (vibration monitoring), which exploits inexpensive accelerometers to perform machine diagnostics. In this work, statistics and hypothesis testing will be used to build a solid foundation for damage detection by recognition of patterns in a multivariate dataset which collects simple time features extracted from accelerometric measurements. In this regard, data from high-speed aeronautical bearings were analyzed. These were acquired on a test rig built by the Dynamic and Identification Research Group (DIRG) of the Department of Mechanical and Aerospace Engineering at Politecnico di Torino. The proposed strategy was to reduce the multivariate dataset to a single index which the health conditions can be determined. This dimensionality reduction was initially performed using Principal Component Analysis, which proved to be a lossy compression. Improvement was obtained via Fisher's Linear Discriminant Analysis, which finds the direction with maximum distance between the damaged and healthy indices. This method is still ineffective in highlighting phenomena that develop in directions orthogonal to the discriminant. Finally, a lossless compression was achieved using the Mahalanobis distance-based Novelty Indices, which was also able to compensate for possible latent confounding factors. Further, considerations about the confidence, the sensitivity, the curse of dimensionality, and the minimum number of samples were also tackled for ensuring statistical significance. The results obtained here were very good not only in terms of reduced amounts of missed and false alarms, but also considering the speed of the algorithms, their simplicity, and the full independence from human interaction, which make them suitable for real time implementation and integration in condition-based maintenance (CBM) regimes.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Vibration Monitoring for Diagnostics through Hypothesis Testing

Daga

Garibaldi

2019

Information

View full text Add to dashboard Cite

show abstract

“…Experiments are carried out by determining the levels of parameters. However, experiments can be analyzed by using methods such as artificial neural networks, where all the values considered can be tested [29,30]. …”

Section: Discussionmentioning

confidence: 99%

“…be analyzed by using methods such as artificial neural networks, where all the values considered can be tested [29,30]. 5.…”

Section: Discussionmentioning

confidence: 99%

Optimization by Using Taguchi Method of the Production of Magnesium-Matrix Carbide Reinforced Composites by Powder Metallurgy Method

Küçük

Elfarah

Islak

et al. 2017

Metals

View full text Add to dashboard Cite

Abstract:The aim of this study was to determine the optimum production parameters in the production of magnesium matrix carbide-reinforced composites by using the powder metallurgy method. The parameter levels maximizing density (%), hardness (HB10), and bending strength (MPa) values were found by using the Taguchi method. The type of reinforcement, the amount of reinforcement, the sintering time, the sintering temperature, additive type, and additive rate were selected as the production parameters. Since the production of Mg and its alloys by using casting methods is problematic, the hot pressing method, a powder metallurgy method, was preferred in this study. Ceramic-based carbide particles were used as reinforcing materials in Mg matrix composite materials. B 4 C, SiC, Mo 2 C, and TiC carbides were preferred as the carbide. Microstructure and phase composition of the produced materials was examined with scanning electron microscope (SEM), X-ray diffractogram (XRD) and X-ray energy dispersive spectrometry (EDS). The hardness of the materials was measured by using a Universal Hardness device. The relative densities of the materials were determined according to Archimedes' principle. The bending strength properties of the materials were determined by using the three-point bending test. The optimum conditions were a sintering temperature of 500 • C, sintering duration of 5 min, additive type of B 4 C and additive rate of 2.5%, and the results obtained at these conditions were found to be as follows; relative density of 98.74 (%), hardness of 87.16 HB10 and bending strength of 193.65 MPa. SEM images taken from the fracture surfaces showed that the carbides added to the matrix had a relatively homogeneous distribution. XRD analyses revealed that the matrix was oxidized very little, and no phase formation occurred between the matrix and the carbides. Carbide addition caused a distinct hardness increase by showing the effect of distribution strengthening in the matrix.

show abstract

“…Three independent factors were considered as shown in Table 3. Furthermore, the results of the FE simulations using the Box-Behnken method were trained to learn the nonlinear relationship between the shape parameters using an artificial neural network (ANN) [14][15][16]. A total of 15 FE simulations were utilized to train and develop the ANN.…”

Section: Conditions Of Fe Simulationmentioning

confidence: 99%

A Four-Roll Squeeze Pointing Machine for a Shape-Drawing Process

Kim

2018

Metals

View full text Add to dashboard Cite

Abstract:A pointing process is a pre-work operation to facilitate the feeding of a rod to pass through a drawing die. After performing the pointing process, the drawing material is inserted into the drawing die and the jaw pulls the end of the material. A bar pointing turning machine, which is universally used for the pointing process, causes a breaking of drawing material easily in the shape-drawing process. Because a shape-drawing process requires a higher drawing load and a smaller cross-sectional area of the pointed zone of drawing material, a pointing process which is to prevent the breaking of the drawing material through a work-hardening effect at an early stage of the drawing process is necessary. In this study, a four-roll squeeze pointing machine (RSPM) as a new automatic pointing machine is introduced. RSPM has been developed to improve the productivity of the pointing process as well as the shape-drawing process by preventing the breaking of the drawing material. The ductile fracture criterion based on Cockcroft-Latham's theory was used to predict the breaking of drawing material and any defects during the pointing process. A tool design method for the RSPM and a feasible pointing size for the conventional pointing machine are proposed. In addition, the drawing materials manufactured using a conventional pointing machine and the RSPM are compared through finite element (FE) simulations and experiments.

show abstract

Using artificial neural networks for the prediction of dimensional error on inclined surfaces manufactured by ball-end milling

Cited by 86 publications

References 51 publications

Machine Vibration Monitoring for Diagnostics through Hypothesis Testing

Machine Vibration Monitoring for Diagnostics through Hypothesis Testing

Optimization by Using Taguchi Method of the Production of Magnesium-Matrix Carbide Reinforced Composites by Powder Metallurgy Method

A Four-Roll Squeeze Pointing Machine for a Shape-Drawing Process

Contact Info

Product

Resources

About