Predictive Modeling and Optimization of TBM Operations: Advanced Techniques Applied to the Jakarta MRT Project DOI

Chairul SalamM,

Orhan Kural

Published: Dec. 5, 2024

Abstract The effectiveness of Earth Pressure Balance (EPB) Tunnel Boring Machines (TBMs) in urban underground construction relies on understanding and optimizing their performance under variable geotechnical conditions. This study investigates the key parameters impacting TBM efficiency during Jakarta Mass Rapid Transit (MRT) Underground Section CP106. Data from operation were analyzed using statistical machine learning techniques, including Mutual Information (MI), Partial Dependence Plots (PDP), Analysis Variance (ANOVA), to identify influential such as Tensile Strength, Uniaxial Spacing, Penetration. Predictive models, Gradient Boosting Regressor, Random Forest Linear Regression, evaluated based error metrics R-squared values, with Regressor showing highest predictive accuracy. Clustering analyses K-Means Principal Component (PCA) further classified operational states, identifying conditions that optimize energy reduce mechanical wear. findings suggest configurations lower Specific Energy, Normal Force, Rolling Force contribute more efficient, less force-intensive tunneling. These insights provide a basis for refining operations modeling tunneling projects.

Language: Английский

Improvement of Principal Component Analysis Algorithm and Its Simulation Experiment DOI
Ling Zhang

Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 194 - 207

Published: Jan. 1, 2025

Language: Английский

Citations

0
Characteristic Canonical Analysis-Based Attack Detection of Industrial Control Systems in the Geological Drilling Process DOI Open Access

Mingdi Xu,

Zhaoyang Jin,

Shengjie Ye

et al.

Processes, Journal Year: 2024, Volume and Issue: 12(9), P. 2053 - 2053

Published: Sept. 23, 2024

Modern industrial control systems (ICSs), which consist of sensor nodes, actuators, and buses, contribute significantly to the enhancement production efficiency. Massive node arrangements, security vulnerabilities, complex operating status characterize ICSs, lead a threat processes’ stability. In this work, condition-monitoring method for ICSs based on canonical variate analysis with probabilistic principal component is proposed. This considers essential information data. Firstly, one-way variance utilized select major variables that affect performance. Then, concurrent monitoring model established both serially correlated subspace its residual subspace, divided by analysis. After that, statistics limits are constructed. Finally, effectiveness superiority proposed validated through comparisons actual drilling operations. The has better sensitivity than traditional methods. experimental result reveals can effectively monitor performance in process highest accuracy 92.31% minimum delay 11 s. achieves much real-world scenarios due distributed structural division characteristic conducted paper.

Language: Английский

Citations

0
Predictive Modeling and Optimization of TBM Operations: Advanced Techniques Applied to the Jakarta MRT Project DOI

Chairul SalamM,

Orhan Kural

Published: Dec. 5, 2024

Abstract The effectiveness of Earth Pressure Balance (EPB) Tunnel Boring Machines (TBMs) in urban underground construction relies on understanding and optimizing their performance under variable geotechnical conditions. This study investigates the key parameters impacting TBM efficiency during Jakarta Mass Rapid Transit (MRT) Underground Section CP106. Data from operation were analyzed using statistical machine learning techniques, including Mutual Information (MI), Partial Dependence Plots (PDP), Analysis Variance (ANOVA), to identify influential such as Tensile Strength, Uniaxial Spacing, Penetration. Predictive models, Gradient Boosting Regressor, Random Forest Linear Regression, evaluated based error metrics R-squared values, with Regressor showing highest predictive accuracy. Clustering analyses K-Means Principal Component (PCA) further classified operational states, identifying conditions that optimize energy reduce mechanical wear. findings suggest configurations lower Specific Energy, Normal Force, Rolling Force contribute more efficient, less force-intensive tunneling. These insights provide a basis for refining operations modeling tunneling projects.

Language: Английский

Citations

0