Prediction of Rolling Force in the Hot Strip Rolling Using Support Vector Regression with Principal Components Analysis

Fontes alternativas de energia estão se tornando cada vez mais frequentes, tendo como objetivo reduzir a poluição ambiental, além de serem ideais para superar a crise energética, logo, neste contexto, a energia solar se destaca por ser abundante. Devido ao alto nível de incerteza dos fatores que interferem diretamente na geração de energia solar, como temperatura e radiação solar, realizar previsões de energia solar com alta precisão é um desafio. Assim, o objetivo deste artigo é desenvolver um modelo de previsão por meio de séries temporais que possibilite prever a produção de energia solar, para 1, 3 e 6 passos à frente, enfatizando a potencialidade da rede neural, utilizando um banco de dados de uma usina fotovoltaica localizada no Uruguai. Para o desenvolvimento da proposta, técnicas de pré-processamento e os métodos de previsão regressão de vetores de suporte (Support Vector Regression, SVR), rede neural perceptron multicamadas com regularização bayesiana (Bayesian Regularized Neural Network, BRNN) e modelo linear generalizado (Generalized Linear Model, GLM) foram combinados. Por fim, tais combinações foram comparadas usando medidas de desempenho. Notou-se que a combinação da análise de componentes principais (Principal Components Analysis - PCA) e a Rede Neural Perceptron Multicamadas com Regularização Bayesiana obteve os melhores resultados, utilizando as três medidas de desempenho.

Download Full-text

Prediction of Rolling Force for Hot Strip Rolling Based on RFR

10.1109/ccdc52312.2021.9602823 ◽

2021 ◽

Author(s):

Ren Yan ◽

Su Nan ◽

Yang Jing ◽

Gao Xiaowen ◽

Wang Huimin ◽

...

Keyword(s):

Rolling Force ◽

Hot Strip Rolling ◽

Strip Rolling ◽

Hot Strip

Download Full-text

Improvement of Methods of Calculation of Forces in Hot Strip Rolling

Solid State Phenomena ◽

10.4028/www.scientific.net/ssp.299.577 ◽

2020 ◽

Vol 299 ◽

pp. 577-581

Author(s):

Georgy L. Baranov

Keyword(s):

Rolling Force ◽

Contact Stresses ◽

Plasticity Condition ◽

Neutral Position ◽

Hot Strip Rolling ◽

Strip Rolling ◽

Normal Contact ◽

Primary Stress ◽

Hot Strip ◽

Frictional Coefficients

A new solution Karman’s equation with the Mises plasticity condition is proposed for determining contact stresses in the slip zones for hot strip rolling. Replacement of the precise plasticity condition by an approximate condition in terms of primary stress leads to a substantial decrease in the length of slip zones and to increase of the rolling force. It was shown that, even at high frictional coefficients, the length of slip zones forms a significant part of the length of deformation region. On the basis of the obtained solutions the techniques for plotting curves of the normal contact stresses, determining the length of the slip zones, the neutral position of the cross section and rolling force refined.

Download Full-text

Application of support vector machine for the prediction of strip crown in hot strip rolling

10.21203/rs.3.rs-31633/v1 ◽

2020 ◽

Author(s):

Ya-feng Ji ◽

Le-Bao Song ◽

Hao Yuan ◽

Wen Peng ◽

Hua-Ying Li ◽

...

Keyword(s):

Support Vector Machine ◽

Prediction Accuracy ◽

Prediction Models ◽

Mean Squared Error ◽

Prediction Performance ◽

Support Vector ◽

Hot Strip Rolling ◽

Strip Rolling ◽

Hot Strip ◽

Strip Crown

Abstract In order to enhance the prediction accuracy of the strip crown and improve the quality of final product in the hot strip rolling, an optimized model based upon support vector machine (SVM) is proposed firstly. Meanwhile, for purposes of enriching data information and ensuring data quality, the actual data from a hot-rolled plant are collected to establish prediction model, as well as the prediction performance of models was evaluated by using multiple indicators. Besides, the traditional SVM model and the combined prediction models with the particle swarm optimization (PSO) and the cuckoo search (CS) optimization algorithm are also proposed. Furthermore, the prediction performance comparisons of the three different methods are discussed and validated. The results show that the CS-SVM has the highest prediction accuracy compared to the other two methods, and the root mean squared error (RMSE) of the proposed CS-SVM is 2.05µm, and 98.11% of prediction data have an absolute error below 4.5μm. In addition, the results also demonstrated that the CS-SVM not only with faster convergence speed and higher prediction accuracy but can be well applied to the actual hot strip rolling production.

Download Full-text

A Statistical Machine Learning Model to Predict Equivalent Circulation Density ECD while Drilling, Based on Principal Components Analysis PCA

10.2118/202101-ms ◽

2021 ◽

Author(s):

Ahmed AlSaihati ◽

Salaheldin Elkatatny ◽

Hani Gamal ◽

Abdulazeez Abdulraheem

Keyword(s):

Machine Learning ◽

Principal Components Analysis ◽

Principal Components ◽

Regularization Parameter ◽

Coefficient Of Determination ◽

Support Vector ◽

Linear Kernel ◽

Svm Model ◽

Components Analysis ◽

Testing Set

Abstract Mathematical equations, based on conservation of mass and momentum, are used to determine the ECD at different depths in the wellbore. However, such equations do not consider important factors that have a influence on the ECD such as: (i) bottom hole temperature, (ii) pipe rotation and eccentricity, and (iii) wellbore roughness. Thus, discrepancy between the calculated ECDs and actual ones has been reported in the literature. This paper aims to explore how artificial intelligence (AI) and machine learning (ML) could provide real-time accurate prediction of the ECD, to have more insight and management of wellbore downhole conditions. For this purpose, a supervised ML algorithm, support vector machine (SVM), based on principal components analysis (PCA), was developed. Actual field data of Well-1 including drilling surface parameters and ECDs, measured by downhole sensors, were collected to develop a classical SVM model. The dataset was split with an 80/20 training-testing data ratio. Sensitivity analysis with different SVM parameters such as regularization parameter C, gamma, kernel type (linear, radial basis function "RBF") was performed. The performance of the model was assessed in terms of root mean square error (RMSE) and coefficient of determination (R2). Afterward, PCA was applied to the dataset of Well-1 to develop an SVM model using the transformed dataset in PCA space. The performance of the model while using different numbers of principal components was evaluated. The results showed that the classical SVM with the linear kernel predicted the ECD with RMSE of 0.53 and R2 of 0.97 in the training set, while RMSE and R2 were 0.56 and 0.97 respectively in the testing set. The PCA-based SVM model, with the linear kernel and four principal components (93.53% variation of the dataset), predicted the ECD with RMSE 0.79 and R2 of 0.95 in the testing set.

Download Full-text