Accurate estimation of tool wear levels during milling, drilling and turning operations by designing novel hyperparameter tuned models based on LightGBM and stacking

Mahmood, Jawad; Mustafa, Ghulam-e; Ali, Muhammad Ammar

doi:10.1016/j.measurement.2022.110722

Cited by 23 publications

(9 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A grid or random search approach for hyperparameter tuning is generally defined as the use of a range of values to evaluate the optimal parameters of the model. While random search is more effective but may not come up with the optimal combination, grid search is comprehensive but computationally challenging [ 66 ]. Numerous validation sets using cross-validation for an exhaustive assessment of the model were utilized.…”

Section: Materials and Methodsmentioning

confidence: 99%

Assessing the Influence of Sensor-Induced Noise on Machine-Learning-Based Changeover Detection in CNC Machines

Biju,

Schmitt,

Engelmann

2024

Sensors

View full text Add to dashboard Cite

The noise in sensor data has a substantial impact on the reliability and accuracy of (ML) algorithms. A comprehensive framework is proposed to analyze the effects of diverse noise inputs in sensor data on the accuracy of ML models. Through extensive experimentation and evaluation, this research examines the resilience of a LightGBM ML model to ten different noise models, namely, Flicker, Impulse, Gaussian, Brown, Periodic, and others. A thorough analytical approach with various statistical metrics in a Monte Carlo simulation setting was followed. It was found that the Gaussian and Colored noise were detrimental when compared to Flicker and Brown, which are identified as safe noise categories. It was interesting to find a safe threshold limit of noise intensity for the case of Gaussian noise, which was missing in other noise types. This research work employed the use case of changeover detection in (CNC) manufacturing machines and the corresponding data from the publicly funded research project (OBerA).

show abstract

Section: Materials and Methodsmentioning

confidence: 99%

Assessing the Influence of Sensor-Induced Noise on Machine-Learning-Based Changeover Detection in CNC Machines

Biju,

Schmitt,

Engelmann

2024

Sensors

View full text Add to dashboard Cite

show abstract

“…State-of-the-art machine learning techniques offer a diverse range of options for sequence data, such as ensemble learning models, such as XGBoost [ 48 ], LightGBM [ 49 ] and CatBoost. XGBoost stands out for its high prediction accuracy and interpretability.…”

Section: Methodsmentioning

confidence: 99%

Sunflower Oil Winterization Using the Cellulose-Based Filtration Aid—Investigation of Oil Quality during Industrial Filtration Probe

Nedić Grujin,

Lužaić,

Pezo

et al. 2023

Foods

View full text Add to dashboard Cite

Waxes, phospholipids, free fatty acids, peroxides, aldehydes, soap, trace metals and moisture present in crude sunflower oil have a negative effect on the oil quality and are, therefore, removed during the refining process. Waxes crystallizing at low temperatures are removed during winterization by cooling and filtration. Waxes have poor filtration characteristics and an industrial filtration process must be enhanced by the use of filtration aids, which improve filter cake structure and properties, and consequently prolong the filtration cycle. Today, traditional filtration aids (diatomite, perlite, etc.) being used in the industry are frequently replaced by cellulose-based aids. The aim of this study is to examine the effect of oil filtration assisted by two cellulose-based filtration aids on the chemical parameters (wax, moisture, phospholipids, soaps, and fatty acids), oil transparency, carotenoids, and Fe and Cu content of sunflower oil obtained in an industrial horizontal pressure leaf filter. In order to investigate the mentioned parameters, the following techniques were used: gravimetric (waxes and moisture content), spectrophotometric (phospholipids and carotenoid content and oil transparency), volumetric (soaps and free fatty acids content) as well as inductively coupled plasma mass spectrometry (ICP-MS) for Fe and Cu content. An artificial neural network model (ANN) was employed for the prediction of removal efficiency based on the chemical quality, oil transparency, Fe and Cu content in oils before filtration, as well as filtration aid quantity and filtration time. Cellulose-based filtration aids had multiple beneficial effects; on average, 99.20% of waxes, 74.88% of phospholipids, 100% of soap, 7.99% of carotenoids, 16.39% of Fe and 18.33% of Cu were removed.

show abstract

“…Researchers used many methods to determine the tool life of a drilling tool [7]. Among them is determining the number of holes produced until the drilling tool fails or breaks, the total machining time used until the drilling tool breaks, chips formation, force signal, condition of drilling tool by using scanning electron microscope (SEM) [8] [17] [18]. The total time they are being used until the drilling tool is broken is sometimes challenging to be used when involving problems during the machining process, where there is a possibility researcher tend to forget to record the machining time before the problems occur.…”

Section: Introductionmentioning

confidence: 99%

Determination of Tool Wear Drilling Process of Aluminum 6061 and Brass C3604 by using CNC Robodrill Machine

Salaam,

Mu’tasim

2024

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Hole making process is a stimulating process as it impacts the drilling tool’s performance and wear. The main objective of this paper is to determine the drilling tool wear by using step-by-step drilling approaches used by most of the manufacturing companies. Two types of materials involved; Aluminum 6061 & Brass C3604. Firstly, the raw material grade was confirmed by testing it using an Oxford Spectrometer. Four types of drilling tools are used; center drill diameter 3.00 mm; drill diameter 10.00, 13.00 and 14.50 mm. The drilling tool material used is High-Speed Steel (HSS). The cutting parameter used is stated in detail in the methodology section. Two approach are used to determine the tool life; measuring the holes diameter produced by using Vernier caliper; graphical form with the help of the camera. The major finding of this research is the understanding on how the cutting process react with the drilling tool during the cutting process as discussed in the results and discussions. The tool life of a center drill, drilling tool diameter 10.00 mm, 13.00 mm and 14.50 mm in drilling Aluminum 6061 is 476, 250, 485 and 499 holes while for Brass C3604 is 360, 190, 360 and 375 holes.

show abstract

Accurate estimation of tool wear levels during milling, drilling and turning operations by designing novel hyperparameter tuned models based on LightGBM and stacking

Cited by 23 publications

References 53 publications

Assessing the Influence of Sensor-Induced Noise on Machine-Learning-Based Changeover Detection in CNC Machines

Assessing the Influence of Sensor-Induced Noise on Machine-Learning-Based Changeover Detection in CNC Machines

Sunflower Oil Winterization Using the Cellulose-Based Filtration Aid—Investigation of Oil Quality during Industrial Filtration Probe

Determination of Tool Wear Drilling Process of Aluminum 6061 and Brass C3604 by using CNC Robodrill Machine

Contact Info

Product

Resources

About