克里金
经验模型
决定系数
超参数
剪切减薄
极限学习机
人工智能
数学
机器学习
阿卡克信息准则
计算机科学
人工神经网络
流变学
统计
材料科学
模拟
复合材料
作者
Siddharth Vishwakarma,Shubham Mandliya,Chandrakant Genu Dalbhagat,Pallavi Kumari Singh,Hari Niwas Mishra
摘要
Abstract The rheological properties (viscosity) of marjoram fortified rice dough (MFRD) inside extruder are helpful for product development, process control, and product quality. This study aims to predict viscosity of MFRD through regression analysis of existing empirical models and machine learning (ML) techniques. At first, MFRD was subjected to steady shear analysis (SSA) at various temperatures (60–100°C) and shear rates (1–50 s −1 ). The SSA data was split into training and validation sets in 70:30 ratio. Rheological modeling was conducted using various empirical models. Hyperparameter tuning was performed using existing MATLAB functions to develop ML models. The SSA revealed that MFRD exhibited pseudoplastic behavior, with viscosity decreasing with increasing temperature up to 70°C and then increasing again, most probably due to starch gelatinization. The regression analysis indicated satisfactory results for Power Law, Carreau, Cross, Sisko, Carreau‐Yasuda model ( R 2 : 0.9849–0.9992) was deemed the best based on Akaike information criterion. A dual variable model (DVM) was then developed using this model, and its coefficient was calculated more efficiently through prepareSurfaceData ( R 2 : 0.8865) MATLAB function than lsqcurvefit ( R 2 : 0.8280) function. In ML model development, Levenberg–Marquardt was the most effective for artificial neural networks (ANNs), grid search for support vector, and Bayesian optimization for decision tree, ensemble, and gaussian process regression (GPR). Only ANN and GPR models had R 2 (testing) of 1 during training, but GPR model (Nash‐Sutcliffe efficiency: 0.9995) outperformed others during validation. Therefore, GPR model can be used to accurately predict MFRD viscosity and DVM for simulation study. Practical applications Shear profile and temperature within the extruder change frequently, affecting the viscosity of MFRD and the extrudate quality. The present study develops a dual variable viscosity prediction model using regression, which can be utilized in simulating extrusion process of MFRD. Using ML approach, the GPR model was found to be better than other models in accurately predicting MFRD viscosity at a range of shear and temperature conditions between 60–100°C and 3–50 s −1 , respectively.
科研通智能强力驱动
Strongly Powered by AbleSci AI