Spectral-Based Fault Diagnosis Methodology for Industrial Shot Blast Machinery Leveraging XGBoost and Feature Importance DOI Open Access
Joon-Hyuk Lee, Chibuzo Nwabufo Okwuosa,

Beak Cheon Shin

et al.

Published: Aug. 13, 2024

The optimal functionality and dependability of mechanical systems are important for the sustained productivity operational reliability industrial machinery which has direct impact on it’s longevity profitability. Therefore, failure a system or any it component would be detrimental to production continuity availability. Consequently,this study proposes robust diagnostic framework analyzing blade conditions shot blast machinery. involves spectral characteristics vibration signals generated by Industrial Shot Blast. Additionally, peak detection algorithms is introduced identify extract unique features present in magnitudes each signal spectrum. A feature importance algorithm then deployed as selection tool, these selected fed into 10 machine learning classifier, with Extreme gradient boosting (XGB) core classifier. Results show that XGB classifier achieved best accuracy 98.05%, cost-efficient computational cost 0.83 seconds. Other global assessment metrics were also implemented further validate model.

Language: Английский

Research on Outgoing Moisture Content Prediction Models of Corn Drying Process Based on Sensitive Variables DOI Creative Commons

Simin Xing,

Zimu Lin,

Xianglan Gao

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(13), P. 5680 - 5680

Published: June 28, 2024

Accurate prediction of outgoing moisture content is the key to achieving energy-saving and efficient technological transformation drying. This study relies on a grain drying simulation experiment system which combined counter current sections design corn kernel experiments. obtains 18 kinds temperature humidity variables during process uses Uninformative Variable Elimination (UVE) method screen sensitive affecting content. Subsequently, six models for were developed, innovatively incorporating Multiple Linear Regression (MLR), Extreme Learning Machine (ELM), Long Short-Term Memory (LSTM). The results show that eight have been screened predict corn. effectively reduced redundancy multicollinearity data in MLR model improved coefficient determination (R2) ELM LSTM by 0.02 0.05. established based full set has an R2 0.910 root-mean-square error (RMSE) 0.881%, while UVE-ELM UVE-LSTM achieve better fitting effect accuracy. with batch size 30, learning rate 0.01, 100 iterations. For training UVE-LSTM, value 0.931 RMSE 0.711%. model, sigmoid as activation function 14 neurons configured, runs fast best values validation are 0.943 0.946, respectively, RMSEs 0.544% 0.581%. proposed this provide reference technical support optimization automation control process.

Language: Английский

Citations

2
Overview of IoT Security Challenges and Sensors Specifications in PMSM for Elevator Applications DOI Creative Commons
Eleni Vlachou, Vasileios I. Vlachou, Dimitrios Efstathiou

et al.

Machines, Journal Year: 2024, Volume and Issue: 12(12), P. 839 - 839

Published: Nov. 22, 2024

The applications of the permanent magnet synchronous motor (PMSM) are most seen in elevator industry due to their high efficiency, low losses and potential for energy savings. Internet Things (IoT) is a modern technology which being incorporated various industrial applications, especially electrical machines as means control, monitoring preventive maintenance. This paper focused on reviewing use PMSM lift systems, application condition techniques real-time data collection using IoT technology. In addition, we focus different categories sensors, connectivity standards they should meet PMSMs used applications. Finally, analyze secure ways transmitting platforms so that transmission information takes into account possible unwanted instructions from exogenous factors.

Language: Английский

Citations

1
Research on Optimizing the Interactive Experience of English Learning for Digital Classrooms DOI Creative Commons
Shupeng Liu

Applied Mathematics and Nonlinear Sciences, Journal Year: 2024, Volume and Issue: 9(1)

Published: Jan. 1, 2024

Abstract The development of modern information technology is changing traditional classroom teaching facilities and modes, this paper investigates the experiences English learners in digital classroom. After clarifying composition classroom, 1,200 were selected for study, questionnaire scale was designed from three aspects, namely, internal factors individual learners, external environment learning, interactive experience learning. We distributed scales to collect relevant data then processed using statistical methods like independent samples t-test, correlation analysis, partial least squares regression, descriptive statistics. results study obtained as follows: mean values students’ learning oriented all range 3.5-4.0, which middle high level. = 3.085 + 0.288 attitudinal characteristics 0.031 self-efficacy 0.095 behavioral motivation 0.588 teacher influence 0.172 support technology. related many factors, construction classrooms can be optimized within students improve their perception experience.

Language: Английский

Citations

0
Research on Athlete Momentum Prediction Based on CNN-LSTM Model DOI
Jingyu Liu,

Yuhan Duan,

Shenghe Sun

et al.

Published: Oct. 18, 2024

Language: Английский

Citations

0
Spectral-Based Fault Diagnosis Methodology for Industrial Shot Blast Machinery Leveraging XGBoost and Feature Importance DOI Open Access
Joon-Hyuk Lee, Chibuzo Nwabufo Okwuosa,

Beak Cheon Shin

et al.

Published: Aug. 13, 2024

The optimal functionality and dependability of mechanical systems are important for the sustained productivity operational reliability industrial machinery which has direct impact on it’s longevity profitability. Therefore, failure a system or any it component would be detrimental to production continuity availability. Consequently,this study proposes robust diagnostic framework analyzing blade conditions shot blast machinery. involves spectral characteristics vibration signals generated by Industrial Shot Blast. Additionally, peak detection algorithms is introduced identify extract unique features present in magnitudes each signal spectrum. A feature importance algorithm then deployed as selection tool, these selected fed into 10 machine learning classifier, with Extreme gradient boosting (XGB) core classifier. Results show that XGB classifier achieved best accuracy 98.05%, cost-efficient computational cost 0.83 seconds. Other global assessment metrics were also implemented further validate model.

Language: Английский

Citations

0