Prediction of Strip Width in Finishing-Mill Group Based on PCA-PSO-LightGBM DOI
Jialiang Wang,

Runyun Yao,

Jingcheng Wang

и другие.

Опубликована: Сен. 23, 2023

Aiming to improve the strip free width hitting rate in finishing rolling, this paper proposes a Light Gradient Boosting Machine (LightGBM) prediction algorithm based on particle Swarm Optimization (PSO) with Principal Component Analysis (PCA). First, raw data are effectively dimensionally reduced using PCA after preprocessing, and input into LightGBM establish model, followed by optimization of key parameters PSO. Finally, design experiments simulations carried out actual production line steel plant Shanghai, results show that PCA-PSO-LightGBM model proposed has good accuracy robustness, meets requirements practical applications.

Язык: Английский

Dynamic Operation Optimization of Complex Industries Based on a Data-Driven Strategy DOI Open Access
Huixin Tian,

Chenning Zhao,

Jueping Xie

и другие.

Processes, Год журнала: 2024, Номер 12(1), С. 189 - 189

Опубликована: Янв. 15, 2024

As industrial practices continue to evolve, complex process industries often exhibit characteristics such as multivariate correlation, dynamism, and nonlinearity, making traditional mechanism modeling inadequate in terms of addressing the intricacies problems. In recent years, with advancements control theory practices, there has been a substantial increase volume data. Data-driven dynamic operation optimization techniques have emerged effective solutions for handling processes. By responding environmental changes utilizing advanced algorithms, it is possible achieve operational processes, thereby reducing costs emissions, improving efficiency, increasing productivity. This correlates nicely goals set forth by conventional theories. Nowadays, this dynamic, data-driven strategy shown significant potential characterized correlations nonlinear behavior. paper approaches subject from perspective establishing models reviewing state-of-the-art time series forecasting cope changing objective functions over time. Meanwhile, aiming at problem concept drift series, summarizes new detection methods introduces model update solve challenge. solving multi-objective problems, reviews developments change response while summarizing commonly used well latest performance measures conclusion, discussion research progress challenges relevant domains undertaken, followed proposal directions future research. review will help deeply understand importance application prospects fields.

Язык: Английский

Процитировано

3
High-Precision machining energy consumption prediction based on multi-sensor data fusion and Ns-Transformer network DOI
Meihang Zhang, Hua Zhang, Wei Yan

и другие.

Expert Systems with Applications, Год журнала: 2025, Номер unknown, С. 126903 - 126903

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

0
An interpretable and reliable framework for alloy discovery in thermomechanical processing DOI Creative Commons
Sushant Sinha,

Xiaoping Ma,

Kashif Rehman

и другие.

Materials Today Communications, Год журнала: 2025, Номер unknown, С. 112134 - 112134

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
A novel paradigm for predicting and interpreting uneven roll wear in the hot rolling steel industry DOI
Wen Peng,

Cheng-yan Ding,

Yü Liu

и другие.

Computers in Industry, Год журнала: 2025, Номер 170, С. 104318 - 104318

Опубликована: Май 21, 2025

Язык: Английский

Процитировано

0
Multi-objective optimization enabling CFRP energy-efficient milling based on deep reinforcement learning DOI
Meihang Zhang, Hua Zhang, Wei Yan

и другие.

Applied Intelligence, Год журнала: 2024, Номер unknown

Опубликована: Сен. 18, 2024

Язык: Английский

Процитировано

2
Prediction of Strip Width in Finishing-Mill Group Based on PCA-PSO-LightGBM DOI
Jialiang Wang,

Runyun Yao,

Jingcheng Wang

и другие.

Опубликована: Сен. 23, 2023

Aiming to improve the strip free width hitting rate in finishing rolling, this paper proposes a Light Gradient Boosting Machine (LightGBM) prediction algorithm based on particle Swarm Optimization (PSO) with Principal Component Analysis (PCA). First, raw data are effectively dimensionally reduced using PCA after preprocessing, and input into LightGBM establish model, followed by optimization of key parameters PSO. Finally, design experiments simulations carried out actual production line steel plant Shanghai, results show that PCA-PSO-LightGBM model proposed has good accuracy robustness, meets requirements practical applications.

Язык: Английский

Процитировано

0