A feature selection method based on an improved fruit fly optimization algorithm in the process of numerical control milling

Yuan, Min; Wang, Mei

doi:10.1177/1687814018778227

Cited by 7 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Feature selection is to select features with high scores in necessary rows from a large number of features, thereby reducing the number of features. The main feature selection methods include feature elimination, mutual information, and the Fisher criterion [9][10][11] . Jiang et al [12] screened the characteristics of HE statistical parameters by mutual information method and established multiple machine learning models.…”

Section: Introductionmentioning

confidence: 99%

Tool wear condition monitoring method based on Relevance Vector Machine

Jia¹,

Yue

Liu³

et al. 2023

Preprint

View full text Add to dashboard Cite

During the machining process, the tool wear state is closely related to the quality of the workpiece, which will directly affect the performance of the equipment. Not timely replacement of tools will lead to increased processing costs, low workpiece surface quality, and even damage to processing equipment. Therefore, research on tool wear monitoring is necessary for the tool processing industry. By analyzing the relationship between tool wear and sensor signals to determine the required acquisition signal. Aiming at the problem that the original sensor data cannot be directly used in the machining process, the signal processing technology is used to preprocess the original signal, remove the invalid signal collected during the cutting process, and use the filtering method to eliminate the singular points in the original signal. The time domain and frequency domain features of the data are extracted. Firstly, the features are optimized by the extreme random tree(ET), and the tool wear is taken as the target vector. The Pearson correlation coefficient(PCC) between the target vector and the filtered features is calculated, and the features with solid correlation with the target vector are selected. The results show that the relevance vector machine(RVM) model proposed in the research can effectively monitor tool wear.

show abstract

Section: Introductionmentioning

confidence: 99%

Tool wear condition monitoring method based on Relevance Vector Machine

Jia¹,

Yue

Liu³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…For many years, off-line 1–7 and on-line 8–12 optimization methods are researched respectively by scholars according to the optimization of CNC machining process, but the basic theory and application of the combination of the two optimization methods are relatively limited. 13–15 But in the actual processing process, there are many factors affecting the processing, in addition to the off-line optimization parameters, but also by the motor current, tool temperature, cutting force, and other online factors.…”

Section: Introductionmentioning

confidence: 99%

Application of combined optimization strategy and system of CNC machining parameter

Liu

Wang

Dong

2022

Advances in Mechanical Engineering

View full text Add to dashboard Cite

This paper presents an efficient optimization method of combining off-line optimization with on-line adaptive control for CNC (Computerized Numerical Control) machining parameter, and develops the corresponding system, aims to multi-objective optimization and cutting force stability. Firstly, the multi-objective optimization solution of machining parameters is obtained by off-line optimization based on Pareto genetic algorithm and TRIZ theory. Then, on this basis, feed speed is adjusted online based on adaptive control system. Next, current is used as feedback to control cutting force, and cutting force can be kept at a set value by the on-line adjusting. We obtain the relationship between off-line optimization and on-line adaptive control. Finally, we build the combined optimization module and embed the self-developed CNC machining system. Based on the system, the experimental results verify that the optimized process parameters make the processing time reduced by 14 s, and no matter how cutting conditions change, cutting force can be stabilized as soon as possible by adjusting feed speed online for keeping the cutting process stability. From the perspective of building the parameter optimization function module, it provides a new idea for the self-development of CNC machining system.

show abstract

“…However, most of these data-driven techniques lack the ability to process unstructured data and hence cannot be easily generalized. Further, the large amount of multivariate data generated in manufacturing processes combined with the high correlation and high dimensional characteristics often require automatic feature extraction and representation capabilities [21,22]. Deep learning algorithms that take into account the temporal correlations of manufacturing data can achieve these capabilities.In Milling cutting, Speed, Feed, and Depth of cut are the major three input parameters.…”

Section: Introduction and Overviewmentioning

confidence: 99%

Tool Wear Prediction in Milling: A Comparative Analysis Based on Machine Learning and Deep Learning Approaches

Kamat¹,

Nargund²,

Kumar³

et al. 2022

IJCDS

View full text Add to dashboard Cite

The milling machine's cutting tool is a vital asset; its breakdown results in unplanned downtime, which reduces industrial efficiency.Tool-Wear Monitoring(TWM) is one of the primary goals of the manufacturing industry due to the manifold benefits it provides, such as optimizing production efficiency, improving performance, and increasing the life of the tool. Most of the work carried out in this domain involves statistical-based techniques, which require expert domain knowledge in formulating degradation models of the tools. Data-driven machine learning and deep learning models have recently been used to analyze tool wear data and make efficient predictions about its remaining useful life. This paper presents a comparative approach to tool wear monitoring using the clustering machine learning technique of K-Nearest Neighbour (k-NN) and deep learning technique of Convolutional Neural Network (CNN) and hybrid Autoencoder-LSTM (AE-LSTM) models. The CNN and AE-LSTM techniques out-perform k-NN by achieving a higher degree of separability of around 93% and 87%, respectively, as per the ROC-AUC values. The techniques provide improved outcomes in terms of precision, recall, and f1-score, indicating that the models are more accurate at detecting false positives.

show abstract

A feature selection method based on an improved fruit fly optimization algorithm in the process of numerical control milling

Cited by 7 publications

References 15 publications

Tool wear condition monitoring method based on Relevance Vector Machine

Tool wear condition monitoring method based on Relevance Vector Machine

Application of combined optimization strategy and system of CNC machining parameter

Tool Wear Prediction in Milling: A Comparative Analysis Based on Machine Learning and Deep Learning Approaches

Contact Info

Product

Resources

About