Use of Artificial Neural Networks (ANNs) for the prediction of the Compressive Strength and Modulus of Elasticity of Concrete
The present investigation indicates the design of an artificial intelligence model based on artificial neural networks (RNA) that allows predicting the Compressive Strength (f'c) and Modulus of elasticity (Ec) of concrete. The methodology was carried out in three stages: The Delta Stage, where...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , |
| التنسيق: | article |
| اللغة: | spa |
| منشور في: |
2024
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://revistadigital.uce.edu.ec/index.php/INGENIO/article/view/5492 |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
إضافة وسم | الملخص: | The present investigation indicates the design of an artificial intelligence model based on artificial neural networks (RNA) that allows predicting the Compressive Strength (f'c) and Modulus of elasticity (Ec) of concrete. The methodology was carried out in three stages: The Delta Stage, where a database was formed consisting of the results of concrete designs (aggregate characterization, dosages, compressive strength and modulus of elasticity) made with GU type cement without additives and aggregates from quarries in the Metropolitan District of Quito, obtained from degree work from various universities in the country and from commercial tests carried out by the Materials and Models Testing Laboratory of the Faculty of Engineering and Applied Sciences. In the following Theta Stage, the design of the ANN was carried out using the Matlab software and the Neural Fitting tool (nftool) for training, validation and testing of the ANN through performance indicators such as the Pearson correlation coefficient (R) in the evaluation stage and the coefficient of determination (R2) to measure the efficiency of the ANN; Finally, in the Gamma stage, the predicted results of the ANN were verified with the actual (f'c) and (Ec) of the concrete obtained through tests carried out on 20 concrete cylinders, designed for resistances of 21, 24 and 28 MPa using aggregates from the Pifo quarry and Type GU cement. Establishing that the RNA satisfactorily predicts the compressive strength and modulus of elasticity of concrete, obtaining a value of R2 for (f'c) equal to 95.12% and for (Ec) of 92.20% between the predicted results with the actual results for mixtures of 21, 24 and 28 MPa; validating its use for the prediction of these properties in concrete. |
|---|