Papers by Keyword: Silicon Content

Abstract: The present study, which is based on the finite element method, investigates the alloy element content assessment of these Fe-Si alloy steel sheets from the view point of eddy current testing. The relative permeability and conductivity were preset parameters. Eddy current signals were gathered by a differential type cylindrical probe and subsequent nondestructive tests were carried out with the preparation of four Fe-Si alloy steel sheets with different silicon contents. The result of the numerical simulation demonstrated that the signal increased with the increase of the silicon content in the Si-Fe alloy steel sheets. And the measured result confirmed the rationality of the application of eddy current testing in the average silicon content assessment.

Abstract: Aluminum silicon alloys, the 4xxx group of the aluminum alloys, are mostly processed into cast parts. As for other cast designated alloys, their weldability is often considered as an unimportant feature. The interest, if any, is almost completely related to the repair welding.The present work analyses how silicon particle dimensions, which depends mostly on the silicon content into the metallic matrix, affects the weld properties. Also it is revealed the HAZ metallographic aspect for two frequently welding processes used on aluminum alloys, TIG and FSW. Friction welds resistance is better than that of the TIG welds, associated with a lack of symmetry (retreating side versus advancing side).Silicon particles diminish their dimensions for hypo eutectic or eutectic concentrations; for hyper eutectic compositions the effect has an almost unimportant effect.

Abstract: In this paper, a predicting model of silicon content in blast furnace hot metal is built based on artificial neural network and genetic algorithm, which optimizes the initialization weight values of neural network with genetic algorithm. This model can effectively improve the prediction accuracy and reduce the calculating time. Online application shows that the predicting model can effectively predict the silicon content in blast furnace hot metal and play an important role in production. when required absolute error was within ±0.03, the accuracy of model can reach 81.4%, and when absolute error was within ±0.04, the accuracy can reach 91.4%.

Abstract: A hierarchical fuzzy system model is presented based on data driving, and then, the model is used to predict the molten iron silicon content in BF. As input variables this model uses the control parameters of a current BF such as moisture, pulverized coal injection, oxygen addition, coke ratio etc. And variables employed to develop the model have been obtained from data collected online from Blast Furnace of Baotou Steel plant. This paper utilizes the fuzzy clustering algorithm combined nearest neighbor clustering and fuzzy c-means clustering to classify the input space. The simulation and error results show that the prediction based on hierarchical fuzzy model and data-driven method has good approximation and fit the output characteristics of the system. The most important point is that the number of fuzzy rules is greatly reduced.

Abstract: Polycarbosilane (PCS), an important precursor for manufacture of silicon carbide (SiC) ceramics, was prepared and analyzed to determine its chemical composition. The major elements of silicon (Si) and carbon (C) in Si-C backbones and side chains in PCS represent more than 80% with minor elements of oxygen and hydrogen being less than 20%. In this work, a conventional potassium silicofluoride volumetric method was explored and modified for establishing a standard routine procedure to evaluate Si content in PCS. The optimal conditions were investigated using an orthogonal designed four-factor-three-level normal experimental scheme. The suitable parameters and standard procedure to analyze Si in PCS were obtained.

Abstract: Various data-based soft sensor models have been established for online prediction of the silicon content in a pig iron. Actually, modeling data often contain many outliers and this can deteriorate the quality of models. However, little attention is paid to efficient outlier detection. Besides, most of traditional outlier detection methods are assumed that data are distributed (approximately) normally and thus they might be invalid for some situations. A new multivariate preprocessing method for outlier detection without any assumption of data distribution is proposed. This novel outlier detection method mainly utilizes a support vector clustering (SVC) strategy. After SVC-based preprocessing, a support vector regression soft sensor model is built. A comparative study for an industrial blast furnace is investigated and the results show its superiority.

Abstract: A new fuzzy prediction controller is established to prediction for silicon content in blast furnace hot metal. The forecasting process is only used the historical information of silicon content. This new algorithm consists of five steps: step 1 de-noises silicon content numerical data by wavelet analysis to smooth out noise; step 2 divides the input and output spaces of the de-noising numerical data into fuzzy regions; step 3 generates fuzzy rules from the de-noising data; step 4 assigns a degree to each of the generated rules for the purpose of resolving conflicts among the generated rules; step 5 determines a fuzzy prediction controller from input space to output space based on such rules. Simulation results show that: 84% hit rates of prediction in the range of [Si] 0.1% is attained using the prediction controller.