A method for predicting concrete durability based on data mining and artificial intelligence algorithm

摘要

#$%^&*AU2020101854A420200924.pdf#####ABSTRACT The invention belongs to the technical field of concrete frost resistance prediction, and discloses a method for predicting concrete frost resistance based on a random forest least squares support vector machine, which mainly comprises the following steps: step 1, based on concrete materials and mixture ratio, selecting main factors affecting concrete frost resistance, constructing an index system, and collecting corresponding sample data as input of a random forest model; Step 2: Order the importance of variables based on random forest regression model, select the characteristics of influencing factors, and select the optimal characteristic variable set to reduce dimensions; Step 3: Input the optimal feature set, use the least squares support vector machine (LLSVM) after parameter optimization to model, train sample data, output the prediction result of concrete relative dynamic elastic modulus, and verify the prediction result of the model by using the test set; Step 4, error analysis is carried out on the prediction results, support vector machines and artificial neural networks without feature selection are selected for modeling, and the same error index is used for comparative analysis to verify the applicability and superiority of the model. According to the method, the random forest feature selection is combined with the least square support vector machine, so that the key features can be extracted on the premise of ensuring the accuracy of the results, the precision of the prediction model is improved, the prediction result is more accurate and stable, and the method can be used as an effective tool for quickly predicting the frost resistance of concrete.