Indirect online tool wear monitoring and model-based identification of process-related signal

Stavropoulos, Panagiotis; Papacharalampopoulos, Alexios; Souflas, Thanassis

doi:10.1177/1687814020919209

Cited by 37 publications

(7 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because the workpiece rotates during turning, optical roughness measurement during machining is not possible, so the in-process tool wear measurement may be a potential solution for that. Several researchers have already dealt with this topic (e.g., [88][89][90][91]), but it would be worthwhile to examine the applicability of this idea from the perspective of the introduced method.…”

Section: Discussionmentioning

confidence: 99%

Theoretical Roughness Modeling of Hard Turned Surfaces Considering Tool Wear

Felhő

Varga

2022

Machines

View full text Add to dashboard Cite

Surface roughness is an important factor in metal cutting, and usually different surface roughness characteristics are used to control the quality of the machined surfaces. However, as the cutting tool wears out during the cutting process, the roughness values change. In most cases, theoretical roughness values are calculated without taking the wear characteristics of the tool into account. For this reason, the calculated and measured roughness values may differ from each other, and the tendency of their change may also be different. This paper presents a method for the determination of the theoretical roughness of hard turned surfaces considering the wear of the cutting tool. The purpose of the analyses performed was to show the effect of wear trace on the tool and the roughness of the machined surface and to give a possible method to take the wear into account when calculating the theoretical roughness values. During the investigations, the shape of the actual (worn) edge section of the cutting tool was recorded by an optical microscope, and the theoretical surface roughness values were calculated with that profile by a CAD modeling method developed earlier. Cutting experiments were conducted on a lathe machine with two similar cutting tools, one of them has significant tool wear, while the other was a completely new one. The calculated theoretical roughness values were compared with real measured roughness values, and the error of the estimates was between 8.7 and 68.3%, larger errors were found at lower feeds.

show abstract

Section: Discussionmentioning

confidence: 99%

Theoretical Roughness Modeling of Hard Turned Surfaces Considering Tool Wear

Felhő

Varga

2022

Machines

View full text Add to dashboard Cite

show abstract

“…During laboratory signal acquisition, testing can be stopped at any time to obtain a small amount of data to facilitate the collection of labelled data. However, in engineering applications, there can be no real-time interruptions [24] [25]. Obtaining tagged and intercepted information is a challenge in the face of large amounts of engineering data [26] [27].…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised Prediction of Milling Cutter Wear Based on an Empirical Formula for Cutting Force and Wear

Yu,

Zhan,

et al. 2024

Preprint

View full text Add to dashboard Cite

Accurately predicting tool wear is essential for maintaining high machining quality. Currently, deep learning models are extensively utilized in predicting tool wear. However, deep learning models are prone to local optima, and limited tool wear sample data and multi-sensor feature fusion limit the model's ability to generalize and extract valid information. To address the above problems, this paper proposes a semi-supervised milling cutter wear prediction method based on an empirical formula for cutting force wear, which is modelled (IRM-CFAM) by the Inception-ResNet Module (IRM) and the Channel Feature Adaptive Module (CFAM). By introducing an empirical formula for cutting force wear, the cutting forces in the unlabeled samples are input to estimate wear values and construct wear curves. Based on the constructed wear curves, a monotonic loss function guided by an empirical curve of milling cutter wear is proposed. Meanwhile, the estimated wear values are combined with unlabeled samples to form a second data source, which is then merged with the labeled samples to expand the dataset. The constructed CFAM consists of channel attention and cross-attention mechanisms, which achieves adaptive fusion of weights. Following validation on the PHM2010 milling cutter dataset, the proposed method attained average RMSE and MAE values of 6.693 and 5.136, respectively, across three test sets, showing a significant advantage over other methods. This research facilitates accurate prediction of milling cutter wear and introduces a novel approach for expanding sample data.

show abstract

“…In the process of signals acquisition in the laboratory, it can be stopped at any time to check the status, and the amount of data is small, so the labeled data is easy to obtain. However, in engineering applications, it is impossible to outages in real time [8,9]. In the face of massive engineering data, it is difficult to obtain marked and intercepted information [10,11].…”

Section: Introductionmentioning

confidence: 99%

Tool Wear Recognition and Signals Labeling with Small Cross Labeled Samples in Impeller Machining

Ou¹,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Data driven deep learning method is the main way to study the condition monitoring of mechanical equipment, in which sufficient labeled signals to train the model parameters is a typical problem. The existing methods to obtain the labeled signals mainly focus on manual marking. For the non-batch impeller processing with variable working conditions, manually marking signals is not the wisest move. To solve this problem, this manuscript puts forward a deep conditional random field neural network (CRFNN) method. This framework fully utilizes the sensitivity of the conditional probability model to adjacent data marker information, and small cross labeled samples are used to predict the labels of unknown signals. At the same time, the variational autoencoder is used to convert the one-dimensional time series signal into three-dimensional image, which solves the problem that the empty tool signals have a great impact on the tool wear condition monitoring in the process of impeller blade machining. Experimental results on a CNC machining center demonstrate the effectiveness and feasibility of the proposed method, and outperforms the existing works under industrial small labeled samples.

show abstract

Indirect online tool wear monitoring and model-based identification of process-related signal

Cited by 37 publications

References 35 publications

Theoretical Roughness Modeling of Hard Turned Surfaces Considering Tool Wear

Theoretical Roughness Modeling of Hard Turned Surfaces Considering Tool Wear

Semi-supervised Prediction of Milling Cutter Wear Based on an Empirical Formula for Cutting Force and Wear

Tool Wear Recognition and Signals Labeling with Small Cross Labeled Samples in Impeller Machining

Contact Info

Product

Resources

About