Failures Prediction in the Cold Forging Process Using Machine Learning Methods

Żabiński, Tomasz; Mączka, Tomasz; Kluska, Jacek; Kusy, Maciej; Hajduk, Z.; Prucnal, S.

doi:10.1007/978-3-319-07173-2_53

Cited by 10 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [22] the LSTM neural network performing a classification task for failure detection in the cold forging process [23], targeted for FPGA implementation, has been presented. The network consisted of two layers.…”

Section: Impact Of the Activation Function Accuracymentioning

confidence: 99%

“…The impact of the activation function accuracy has also been examined for the auto-associative neural network performing novelty detection in the milling process [25]. Another auto-associative neural network has also been developed and examined for the cold forging process [23]. Contrary to the previous findings, in both cases it turned out that the activation function accuracy is virtually negligible due to the fact that the arguments of the hyperbolic tangent function took values for which the function value was limited to 1.0 or -1.0.…”

Section: Impact Of the Activation Function Accuracymentioning

confidence: 99%

See 1 more Smart Citation

Very High Accuracy Hyperbolic Tangent Function Implementation in FPGAs

Hajduk¹,

Dec²

2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

The paper presents in detail a relatively simple implementation method of the hyperbolic tangent function, particularly targeted for FPGAs. The research goal of the proposed method was to examine the usage of the approximation of ordinary or Chebyshev polynomials for the implementation of the function. Several miscellaneous implementation versions have been considered. They differ in the polynomial degree, number of intervals for which the domain of the function is divided, etc. Both floatingpoint and fixed-point implementations have been presented. An impact on the FPGA resources utilization and calculations time for the implementation versions has also been briefly analyzed. Special attention has been paid to the accuracy of the calculations of the function. It turned out that applying the proposed method, a very high calculations accuracy can be achieved, while simultaneously maintaining reasonable resources utilization and short calculations time. The proposed method can be an effective alternative to other encountered implementation methods such as CORDIC. Additionally, the presented hardware architecture is more versatile and can be easily adapted for the implementation of other mathematical functions.

show abstract

Section: Impact Of the Activation Function Accuracymentioning

confidence: 99%

Section: Impact Of the Activation Function Accuracymentioning

confidence: 99%

Very High Accuracy Hyperbolic Tangent Function Implementation in FPGAs

Hajduk¹,

Dec²

2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…4. From the authors' experience and other works, it is clear that signal features in the frequency domain instead of features in the time domain should be considered for building a classifier [29], [30], [25], [24]. Thus, the vector Z = [Z (k) , .…”

Section: Input Data Preparation For Aann Implementationmentioning

confidence: 99%

FPGA-Embedded Anomaly Detection System for Milling Process

et al. 2021

Self Cite

View full text Add to dashboard Cite

The main goal of this work is to design a supervising controller able to detect an anomaly in the milling process and implement the soultion in Field Programmable Gate Array (FPGA) chip. Executing this task, the controller continuously monitors the vibration signal coming from the acceleration sensor, installed on the milling machine, and striving to isolate new vibration patterns which are different from typical patterns recorded for the correct milling process. The detection method relies on determining selected signal features in the frequency domain and applying an auto-associative neural network (AANN) for novelty detection. It has been shown that by exercising the frequency spectrum of the vibration signal, extracting specific features of the signal's spectrum, and using an auto-associative neural network, it is possible to detect anomalies in a milling process with relatively high efficiency. The accuracy, sensitivity, specificity, precision, and false alarm rate are equal to 94.3, 100, 91.2, 88.9, and 8.8 percent. All necessary calculations can be accomplished by the developed single-chip FPGA embedded supervising controller. The controller allows high-speed calculations under low power consumption. It characterizes high reliability and low price compared to typically encountered solutions.

show abstract

“…This article falls within current research trends for condition monitoring of machines and tools. Typical problems include: spindle bearing condition monitoring [3], detection of damage tool of CNC machines or robots [4,5,6,7,8], detection of irregularities during the performance of industrial processes [4,5,9,10].…”

Section: Introductionmentioning

confidence: 99%

The Manipulator Tool Fault Diagnostics Based on Vibration Analysis in the Frequency Domain

Gierlak

Szuster

2016

AMM

View full text Add to dashboard Cite

The issues presented in the article, relate to the detection of damage of a cutting tool used in robotised machining. Due to the time saving requirements, it is desirable to carry out the current control of the tool mounted in the holder of the robotic manipulator. The tool is a ceramic fiber brush used for grinding. A typical damage of the brush is fiber breakage, which leads to an unbalance of the tool and vibrations. The phenomenon of vibrations and parameters of the vibratory motion of the tool have been used as a carrier of information about the state of the tool. On the basis of the measurement data, obtained during tests of tools with varying degrees of damage, classifiers of the tool state were built. Two types of classifiers were tested: decision trees and artificial neural networks. The results confirm that it is possible to build a classifier of the tool state with high effectiveness reaching up to 99,875%.

show abstract

Failures Prediction in the Cold Forging Process Using Machine Learning Methods

Cited by 10 publications

References 9 publications

Very High Accuracy Hyperbolic Tangent Function Implementation in FPGAs

Very High Accuracy Hyperbolic Tangent Function Implementation in FPGAs

FPGA-Embedded Anomaly Detection System for Milling Process

The Manipulator Tool Fault Diagnostics Based on Vibration Analysis in the Frequency Domain

Contact Info

Product

Resources

About