Defects detection from time series of cutting force in composite milling process by recurrence analysis

2020 ◽  
Vol 39 (23-24) ◽  
pp. 890-901
Author(s):  
Krzysztof Ciecieląg ◽  
Krzysztof Kecik ◽  
Kazimierz Zaleski
Keyword(s):  
Time Series ◽  
Milling Time ◽  
Milling Process ◽  
Minimal Size ◽  
Indicator Values ◽  
Recurrence Plots ◽  
Defect Depth ◽  
Recurrence Analysis ◽  
Nonlinear Methods ◽  
Different Diameters

The article presents possibility of defects detection from milling time series using nonlinear methods: recurrence plots and recurrence quantifications. The defects are modeled as the holes with different diameters and depths. This allows estimation of the minimal size of defect possible to detect. Based on the conducted research, it has been shown that some of the recurrence indicators enable detection of defects. These recurrence indicators have been tested on the reals damage. Additionally, we show influence of the defect depth on the recurrence indicator values.

scholarly journals Damage detection by recurrence and entropy methods on the basis of time series measured during composite milling

2020 ◽  
Vol 111 (1-2) ◽  
pp. 549-563
Author(s):  
Krzysztof Kecik ◽  
Krzysztof Ciecielag ◽  
Kazimierz Zaleski
Keyword(s):  
Time Series ◽  
Damage Detection ◽  
Approximate Entropy ◽  
Recurrence Plot ◽  
Milling Process ◽  
Fiber Reinforced Polymer ◽  
Reinforced Polymer ◽  
Glass Fiber Reinforced ◽  
Damage Indicators ◽  
Different Diameters

Abstract This paper presents methods for damage detection in machined material on the basis of time series measured during milling of glass-fiber–reinforced polymer (GFRP). Recurrence methods and different types of entropy have emerged as useful tools for detecting subtle non-stationarities and/or changes in nonlinear signals. In this research, a recurrence plot, recurrence quantifications, an approximate entropy, and sample entropy are used. By identifying changes in the cutting force measured during the composite milling process, the damage occurrence has been detected. Firstly, the damage has been modelled as the intentionally introduced hole with different diameters and depths in order to estimate the size detectable damages and to select proper recurrence measures as damage indicators. Next, the experiments with the real damage have been performed and the damage indicators have used.

Analysis of time series recurrence and cross recurrence in the relationship of climate with Coffee Leaf Rust

2021 ◽  
Author(s):  
Ana M. Tarquis ◽  
Emmanuel Lasso ◽  
Juan Carlos Corrales ◽  
Elias de Melo
Keyword(s):  
Time Series ◽  
Leaf Rust ◽  
Central America ◽  
Financial Support ◽  
Weather Conditions ◽  
Production Costs ◽  
Recurrence Plots ◽  
Recurrence Analysis ◽  
Coffee Rust ◽  
Coffee Leaf Rust

<p>Agroindustry in South and Central America is positioned as a traditional production sector, where exists a need for integration of processes for the implementation of contingency measures in a timely manner against events that create a risk for crops. Diseases affecting agricultural sectors are often closely related to weather conditions and crop management. In particular, for the coffee production, the Coffee Leaf Rust (CLR) is a disease that affects quality and production costs for farmers greatly. </p><p>Detecting the patterns that affect the disease can lead to early actions that lessen its impact. In this sense, some researchers in the sector have focused their efforts on determining over time the relationships between weather conditions and agronomic properties of crops with episodes of epidemics of diseases as coffee rust. </p><p>Different natural processes, such as the climate, can have different and recurrent behaviors in time. Despite its periodicity, climate change has impacted on recurring events, both in their temporality and intensity. Thus, climate variables have properties of dynamic deterministic or nonlinear systems. The recurrence analysis of states in these systems is one of the solutions to carry out a study of their behavior in the time-domain.  Eckmann et al. proposed the Recurrence Plots (RP) for the visualization of state recurrence, allowing to see the space phase trajectories in a bidimensional representation. This analysis, initially applied to a single time series and its recurrence with itself, can also be extended to compare two time series by Cross Recurrence Plots (CRP) and find the recurrence between them. Moreover, the elements of PR and CRP can be quantified, obtaining direct elements of comparison between series or pairs of time series.</p><p>The aim of this analysis was to find the times and conditions in which the time series of the climatic variables present events related to anomalies or extreme values in the CLRI time series. In addition, the recurrence analysis allows to know the time delay for which each climatic variable affects the disease.</p><p><strong>References</strong></p><p> J. Avelino et al., «The coffee rust crises in Colombia and Central America (2008–2013): impacts, plausible causes and proposed solutions», Food Secur.,  7(2), 303-321, 2015.</p><p> J. M. Waller, M. Bigger, y R. J. Hillocks, Coffee pests, diseases and their management. CABI, 2007.</p><p> A. C. Kushalappa y A. B. Eskes, «Advances in coffee rust research», Annu. Rev. Phytopathol., 27(1), 503–531, 1989.</p><p> E. Lasso, D. C. Corrales, J. Avelino, E. de Melo Virginio Filho, y J. C. Corrales, «Discovering weather periods and crop properties favorable for coffee rust incidence from feature selection approaches», Comput. Electron. Agric., 176, 105640, 2020.</p><p> J. P. Eckmann, S. O. Kamphorst, y D. Ruelle, «Recurrence plots of dynamical systems», World Sci. Ser. Nonlinear Sci. Ser. A, 16, 441–446, 1995.</p><p><strong>Acknowledgements</strong></p><p>Technical support of Telematics Engineering Group (GIT) of the University of Cauca, the Tropical Agricultural Research and Higher Education Center (CATIE) and the InnovAccion Cauca project of the Colombian Science, Technology, and Innovation Fund (SGR- CTI) for PhD scholarship granted to MSc. Lasso is acknowledge. Financial support by Fundación Premio Arce (ETSIAAB, UPM) financial support under contract FPA18PPMAT08 is greatly appreciated.</p>

scholarly journals A Refinement of Recurrence Analysis to Determine the Time Delay of Causality in Presence of External Perturbations

Entropy ◽  
2020 ◽  
Vol 22 (8) ◽  
pp. 865
Author(s):  
Emmanuele Peluso ◽  
Teddy Craciunescu ◽  
Andrea Murari
Keyword(s):  
Time Series ◽  
Recurrence Plot ◽  
Recurrence Plots ◽  
Recurrence Analysis ◽  
External Perturbations ◽  
New Type ◽  
Periodic Signals ◽  
Alternative Techniques ◽  
Definition Of ◽  
The Right

This article describes a refinement of recurrence analysis to determine the delay in the causal influence between a driver and a target, in the presence of additional perturbations affecting the time series of the response observable. The methodology is based on the definition of a new type of recurrence plots, the Conditional Joint Recurrence plot. The potential of the proposed approach resides in the great flexibility of recurrence plots themselves, which allows extending the technique to more than three quantities. Autoregressive time series, both linear and nonlinear, with different couplings and percentage of additive Gaussian noise have been investigated in detail, with and without outliers. The approach has also been applied to the case of synthetic periodic signals, representing realistic situations of synchronization experiments in thermonuclear fusion. The results obtained have been very positive; the proposed Conditional Joint Recurrence plots have always managed to identify the right interval of the causal influences and are very competitive with alternative techniques such as the Conditional Transfer Entropy.

Development of Biodegradable Mg-Ti Alloy Synthesized Using Mechanical Alloying

2016 ◽  
Vol 1133 ◽  
pp. 75-79 ◽  
Author(s):  
Emee Marina Salleh ◽  
Sivakumar Ramakrishnan ◽  
Zuhailawati Hussain
Keyword(s):  
Mechanical Alloying ◽  
Milling Time ◽  
Milled Powder ◽  
Milling Process ◽  
Planetary Mill ◽  
Ti Alloy ◽  
X Ray Diffraction ◽  
Argon Flow ◽  
Diffraction Patterns ◽  
Ti Powder

The aim of this work was to study the effect of milling time on binary magnesium-titanium (Mg-Ti) alloy synthesized by mechanical alloying. A powder mixture of Mg and Ti with the composition of Mg-15wt%Ti was milled in a planetary mill under argon atmosphere using a stainless steel container and balls. Milling process was carried out at 400 rpm for various milling time of 2, 5, 10, 15 and 30 hours. 3% n-heptane solution was added prior to milling process to avoid excessive cold welding of the powder. Then, as-milled powder was compacted under 400 MPa and sintered in a tube furnace at 500 °C in argon flow. The refinement analysis of the x-ray diffraction patterns shows the presence of Mg-Ti solid solution when Mg-Ti powder was mechanically milled for 15 hours and further. Enhancements of Mg-Ti phase formation with a reduction in Mg crystallite size were observed with the increase in milling time. A prolonged milling time has increased the density and hardness of the sintered Mg-Ti alloy.

Failure time prognosis in manufacturing process using multi-dislocated time series convolutional neural network

2020 ◽  
pp. 095440892097414
Author(s):  
Bohao Li ◽  
Liping Zhao ◽  
Yiyong Yao
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Convolutional Neural Network ◽  
Manufacturing Process ◽  
Failure Time ◽  
Great Difficulty ◽  
Milling Process ◽  
Machining Process ◽  
Massive Data ◽  
The Difference

Failure time prognosis in manufacturing process plays a crucial role in guaranteeing manufacturing safety and reducing maintenance loss. However, most current prognosis methods face great difficulty when handling massive data collected from manufacturing process. Convolutional neural network (CNN) provides an effective way to extract features with massive data. Due to the difference between images and multisensory signals, CNN is not suitable for machining process. Inspired by the idea of CNN, a novel prognosis framework is proposed based on the characteristics of multisensory signals, which is called multi-dislocated time series convolutional neural network (MDTSCNN). The proposed MDTSCNN is composed of multi-dislocate layer, convolutional layer, pooling layer and fully connected layer. By adding a multi-dislocate layer, this model can learn the relationship between different signals and different intervals in periodic multisensory signals. The effectiveness of proposed method is validated by a milling process. Compared to other prognosis method, the proposed MDTSCNN shows enhanced performances in prediction accuracy.

scholarly journals The Structural Evolution of Al86Ni9La5 Glassy Ribbons during Milling at Room and Cryogenic Temperatures

Materials ◽  
2018 ◽  
Vol 11 (10) ◽  
pp. 1956
Author(s):  
Zhicheng Yan ◽  
Yan Liu ◽  
Shaopeng Pan ◽  
Yihua Hu ◽  
Jing Pang ◽  
...  
Keyword(s):  
Milling Time ◽  
Shear Bands ◽  
Structural Evolution ◽  
High Energy ◽  
Milling Process ◽  
Cryogenic Temperatures ◽  
Dynamic Simulations ◽  
Tricapped Trigonal Prism ◽  
Medium Range ◽  
Energy Ball

Melt-spun metallic Al86Ni9La5 glassy ribbons solidified at different circumferential speeds (Sc) were subjected to high-energy ball milling at room and cryogenic temperatures. Crystallization induced by milling was found in the Al86Ni9La5 solidified at lower circumferential speed (Sc = 14.7 m/s), while the Al86Ni9La5 with Sc = 36.6 m/s kept amorphous. Besides, a trend of structural rejuvenation during milling process was observed, as the onset temperatures (Tx1, Tx2) and the crystallization enthalpies (ΔH1, ΔH2) first decreased and then increased along with the milling time. We explored the structural origin of crystallization by ab initio molecular dynamic simulations and found that the tricapped trigonal prism (TTP) Ni-centered clusters with a higher frequency in samples solidified at a lower cooling rate, which tend to link into medium-range orders (MROs), may promote crystallization by initiating the shear bands during milling. Based on the deformation mechanism and crush of metallic glasses, we presented a qualitative model to explain the structural rejuvenation during milling.

scholarly journals Electrical Conductivity Behavior of La0.5Ca0.5Mn0.5Fe0.5O3 Synthetized by the Solid State Reaction

2018 ◽  
Vol 12 (12) ◽  
pp. 174
Author(s):  
H. A. Martínez-Rodríguez ◽  
J. F. Jurado ◽  
E. Restrepo-Parra
Keyword(s):  
Solid State ◽  
Thermal Treatment ◽  
Solid State Reaction ◽  
Milling Time ◽  
Electrochemical Impedance ◽  
Solid State Reaction Method ◽  
Milling Process ◽  
Secondary Phases ◽  
Two Samples ◽  
Semiconductor Behavior

La0.5Ca0.5Mn0.5Fe0.5O3 was synthesized using the solid state reaction method. This method consists of two main processes: a milling process and a subsequent thermal treatment. Two samples at different conditions were produced: one using 2 h of milling time and 900°C (M-I), and the other using 6 h of milling time and 1200°C of thermal treatment (M-II).  X-ray diffraction analysis indicated, in both cases, an orthorhombic crystalline ordering of the space group Pbnm. For the case of M-I, the material exhibited secondary phases, different than the desired phase; on the contrary, in M-II, these secondary phases were not present. The dielectric response determined using electrochemical impedance spectroscopy (EIS) performed in a temperature range between 20°C and 300°C exhibited a thermally activated semiconductor behavior with activation energies of Eg= 0.11±0.05 eV and Eg= 0.47±0.06 eV for M-I and M-II, respectively.

scholarly journals Encoding Time Series as Multi-Scale Signed Recurrence Plots for Classification Using Fully Convolutional Networks

Sensors ◽  
2020 ◽  
Vol 20 (14) ◽  
pp. 3818
Author(s):  
Ye Zhang ◽  
Yi Hou ◽  
Shilin Zhou ◽  
Kewei Ouyang
Keyword(s):  
Time Series ◽  
State Of The Art ◽  
Recurrence Plot ◽  
The State ◽  
Recurrence Plots ◽  
Convolutional Networks ◽  
Multi Scale ◽  
Fully Convolutional Networks ◽  
Benchmark Datasets ◽  
Visualization Evaluation

Recent advances in time series classification (TSC) have exploited deep neural networks (DNN) to improve the performance. One promising approach encodes time series as recurrence plot (RP) images for the sake of leveraging the state-of-the-art DNN to achieve accuracy. Such an approach has been shown to achieve impressive results, raising the interest of the community in it. However, it remains unsolved how to handle not only the variability in the distinctive region scale and the length of sequences but also the tendency confusion problem. In this paper, we tackle the problem using Multi-scale Signed Recurrence Plots (MS-RP), an improvement of RP, and propose a novel method based on MS-RP images and Fully Convolutional Networks (FCN) for TSC. This method first introduces phase space dimension and time delay embedding of RP to produce multi-scale RP images; then, with the use of asymmetrical structure, constructed RP images can represent very long sequences (>700 points). Next, MS-RP images are obtained by multiplying designed sign masks in order to remove the tendency confusion. Finally, FCN is trained with MS-RP images to perform classification. Experimental results on 45 benchmark datasets demonstrate that our method improves the state-of-the-art in terms of classification accuracy and visualization evaluation.

Sign in / Sign up

Export Citation Format

Share Document