AE-XGBoost: A Novel Approach for Machine Tool Machining Size Prediction Combining XGBoost, AE and SHAP

Mathematics, Journal Year: 2025, Volume and Issue: 13(5), P. 835 - 835

Published: March 2, 2025

To achieve intelligent manufacturing and improve the machining quality of machine tools, this paper proposes an interpretable size prediction model combining eXtreme Gradient Boosting (XGBoost), autoencoder (AE), Shapley additive explanation (SHAP) analysis. In study, XGBoost was used to establish evaluation system for actual computer numerical control (CNC) tools. The combined with SHAP approximation effectively capture local global features in data using autoencoders transform preprocessed into more representative feature vectors. Grey correlation analysis (GRA) principal component (PCA) were reduce dimensions original features, synthetic minority overstimulation technique Gaussian noise regression (SMOGN) method deal problem imbalance. Taking tool as response parameter, based on parameters milling process CNC tool, effectiveness is verified. experimental results show that proposed AE-XGBoost superior traditional method, accuracy 7.11% higher than method. subsequent reveals importance interrelationship provides a reliable decision support processing personnel, helping manufacturing.

Language: Английский

---

A Machine Learning Predictive Model for the Charging Capacity of Stationary Lithium-Ion Batteries Connected to Renewable Energy Sources at Remote Oil and Gas Fields

Published: April 21, 2025

Abstract As the oil and gas industry strives to meet its energy transition goals while ensuring reliable uninterruptable electricity generation operational needs, where grid access is often limited, Renewable Energy Sources (RES), particularly wind solar power, have become increasingly vital. Stationary lithium-ion batteries (LIBs) play a pivotal role in needs at sites, providing backup power for critical systems, enabling microgrid stability, reducing reliance on fossil-fuel-powered generators. This paper addresses challenge of limited useful life LIBs—a key storage solution integrating renewable sources applications—by leveraging advanced AI machine learning techniques. A Machine Learning predictive model developed using Long Short-Term Memory (LSTM) algorithm enhance accuracy reliability battery management systems. The trained an extensive dataset grid-scale LIBs which includes measurements Voltage, Current, Cell Temperature, Environment Temperature. Rigorous data pre-processing steps were undertaken ensure integrity performance. sequential time-based splitting technique preserved temporal dependencies crucial time-series analysis. Feature engineering identified Voltage Current as predictors, supplemented by timestamps temperature features. Data augmentation was employed generate synthetic time series capacity, further enhancing accuracy, generalization. LSTM effectively predicts annual degradation charging capacity stationary Lithium-ion subjected sporadic currents generated RES, achieving Root Mean Square Error 0.0126 Absolute 0.0107, proving adequacy handling complex, non-linear dynamics inherent performance metrics. study significant gap research, focusing predicting offshore remote locations experience current, opposed constant current voltage conditions typically studied. It underscores potential learning-driven advancements technology, enhanced efficiency consequently more sustainable operations.

Language: Английский

---

Imbalanced regression pipeline recommendation

Machine Learning, Journal Year: 2025, Volume and Issue: 114(6)

Published: April 29, 2025

---

CSIML: a cost-sensitive and iterative machine-learning method for small and imbalanced materials data sets

Chemistry Letters, Journal Year: 2024, Volume and Issue: 53(5)

Published: May 1, 2024

Abstract Materials science research benefits from the powerful machine-learning (ML) surrogate models, but it is also limited by implicit requirement for sufficiently big and balanced data distribution ML. In this paper, we propose a model to obtain more credible results small imbalanced materials sets as well chemical knowledge. Taking 2 bandgaps instances, demonstrate usability performance of our compared with common ML models normal sampling resampling methods.

Language: Английский

---

The relationship between prognostic factors and patient satisfaction with performance of self-identified goals following interdisciplinary mild traumatic brain injury rehabilitation

Physiotherapy Theory and Practice, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 9

Published: Sept. 11, 2024

Individuals with persistent symptoms following a mild traumatic brain injury (mTBI) demonstrate improved satisfaction their performance of self-identified rehabilitation goals after completing combined occupational therapy and physiotherapy group intervention. However, the relationship between factors associated developing an mTBI intervention are unknown.

Language: Английский

---