Article
Multidisciplinary Sciences
Hongmei Shi, Lin Wei, Cui Wang, Shuai Wang, Yufu Ning
Summary: The traditional combination forecasting model requires precise historical data and cannot handle uncertainty or imprecise data. This paper proposes two uncertain combination forecasting models based on uncertain least squares estimation, which can better deal with the forecasting problem of imprecise data. Experimental results show that the proposed model outperforms the existing models in terms of forecasting accuracy.
Article
Thermodynamics
Mohamed Maaouane, Smail Zouggar, Goran Krajacic, Hassan Zahboune
Summary: Forecasting energy demand for the industrial sector is complex due to specific energy demand differences in each sub-sector. A model using multiple linear regression method and macroeconomic variables was developed and tested in Morocco, revealing a need for 8.27 MToe in 2050 and potential energy savings through efficiency measures. The study also showed that biogas could only replace 36.4% of LPG demand.
Article
Engineering, Chemical
Pawel Wozniak, Agata Bienczak, Stanislaw Nosal, Joanna Piepiorka-Stepuk, Monika Sterczynska
Summary: This study investigates the relationship between the efficiency of onion peeling and the amount of waste generated. The results show that the peeling efficiency of the onion and waste of peel mass depend on the compressed air pressure, while the hardness and size of the onion are irrelevant to the process.
Article
Mathematics
Tao Hu, Baosheng Liang
Summary: Motivated by the relative loss estimator of the median, a new class of estimators for linear quantile models is proposed using a general relative loss function. The proposed estimator is shown to have smaller variance and be more efficient than traditional linear quantile estimator. Simulation studies and application in a prostate cancer study demonstrate good performance of the proposed method.
Article
Engineering, Mechanical
Yanjuan Hu, Wenjun Lv, Zhanli Wang, Liang Liu, Hongliang Liu
Summary: This paper proposes a method of compensating for brake disc balance error by using a machine learning algorithm. The random forest model shows higher prediction accuracy in compensating for the errors during the calibration process, improving the balance accuracy.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Chemistry, Multidisciplinary
Zhike Zhao, Caizhang Wu
Summary: A wheat inventory monitoring method based on inventory measurement and support vector machine regression (SVR) prediction model is proposed to address the challenges in evaluating the storage quantity of wheat. The experimental results demonstrated that the SVR prediction model had a high prediction accuracy, meeting the requirements of wheat quantity monitoring in grain warehouses.
APPLIED SCIENCES-BASEL
(2023)
Article
Green & Sustainable Science & Technology
Shengming Tang, Tiantian Li, Yun Guo, Rong Zhu, Hongya Qu
Summary: This study conducted experimental tests on DWL at two national meteorological observatories in China and proposed a correction methodology based on multiple linear regression model to eliminate measurement errors induced by environmental conditions. The corrected DWL data showed significant improvements in accuracy and reliability.
Article
Computer Science, Interdisciplinary Applications
Caiyun Fan, Wenbin Lu, Yong Zhou
Summary: The study introduces an omnibus test for examining error heterogeneity in censored linear regression, based on testing variance components in a working kernel machine regression model. The empirical performance of the proposed tests is evaluated through simulations and real data sets.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2021)
Article
Materials Science, Multidisciplinary
He Fei, Chai Xianyi, Zhu Zhenghai
Summary: This study proposes prediction models based on BPNN-IL for total blow oxygen volume and second blow oxygen volume in the dynamic BOF steelmaking process. The results show that the BPNN-IL models provide the most accurate prediction, and the introduction of incremental learning further improves the predictive accuracy.
HIGH TEMPERATURE MATERIALS AND PROCESSES
(2022)
Article
Metallurgy & Metallurgical Engineering
Miao Wang, Chuang Gao, Xingang Ai, Baopeng Zhai, Shengli Li
Summary: This paper proposes a hybrid end-point static control model for accurate control of the basic oxygen furnace (BOF) end-point in steelmaking. By establishing a prediction model and optimizing the objective function, the optimal values for oxygen blowing volume and lime weight are calculated, meeting the requirements of actual field production.
TRANSACTIONS OF THE INDIAN INSTITUTE OF METALS
(2022)
Article
Thermodynamics
Tun-Ping Teng, Wei-Jen Chen
Summary: Households in humid climates commonly use portable refrigeration dehumidifiers, but internal heat can lead to RH measurement deviations. This study proposes solutions using an external temperature sensor and multiple linear regression algorithm, which significantly reduce the deviations.
CASE STUDIES IN THERMAL ENGINEERING
(2022)
Article
Water Resources
Tessa Bermarija, Lindsay Johnston, Chris Greene, Barret Kurylyk, Rob Jamieson
Summary: Study region: Halifax, Nova Scotia, Canada. Study focus: Many lakes in the Halifax region are facing the issue of elevated chloride (Cl-) levels due to deicing salt application. In this study, geospatial analysis and linear regression methods were used to identify the main contributing factors to high Cl- concentrations in lakes. A mass balance model was also developed to estimate Cl- loading rates for different land use categories. The findings provide new hydrological insights for the region and support predictive modeling for future development impacts.
JOURNAL OF HYDROLOGY-REGIONAL STUDIES
(2023)
Article
Chemistry, Analytical
Zhuang Lv, Kaifeng Zhu, Xin He, Lei Zhang, Jiawei He, Zhiya Mu, Jun Wang, Xin Zhang, Ruidong Hao
Summary: In this study, a new method for detecting and correcting phase unwrapping errors (PUE) in fringe projection profilometry is proposed. The method uses multiple linear regression analysis and an improved median filter to detect and correct errors in the unwrapped phase map. Experimental results show that the method is effective and robust, and it can handle highly abrupt or discontinuous regions.
Article
Automation & Control Systems
Dongdong Chen, Pin Lv, Lei Xue, Hongwen Xing, Lixin Lu, Dongdong Kong
Summary: This paper proposes a positional error compensation method based on Bayesian linear regression to improve the positional accuracy of aviation drilling robots. By constructing a stochastic model for robot positional error and utilizing Bayesian linear regression, predicted positional errors and confidence intervals can be provided for error compensation. Experimental results show that this method significantly reduces the positional error of aviation drilling robots.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Article
Mathematics, Applied
Hao Ming, Huilan Liu, Hu Yang
Summary: This study examines the multiplicative additive models based on the LPRE criterion proposed by Chen et al. (2016), utilizing B-spline basis functions for nonparametric function estimation and SCAD penalty function for variable selection. The research demonstrates optimal convergence rate and variable selection consistency, with simulation results and case analysis showing superior performance compared to existing methods.
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS
(2022)