Research And Application About Intelligent Algorithms For Predicting Bearing Capacity Of RC Beams Strengthened By Carbon Fiber Sheets

While construction and technology develops continually, the construction industry is faced with the maintenance and reform about existing buildings. As a new reinforced method, the technique about reinforcing by carbon fiber reinforced plastic (CFRP) has become a hot topic in the engineering fields, domestic and foreign scholars have conducted extensive research on it. However, there are many influencing factors about flexural capacity of RC beams strengthened by carbon fiber sheets, and the relationships between them are very complicated, while there are lots of limitations by using the existing methods of calculating and analyzing, so we can study the relationships between the influencing factors by using the intelligent algorithms.First of all, we confirm the main influencing factors (?)bout flexural capacity of RC beams strengthened by carbon fiber sheets by researching and analyzing the relevant experiments of RC beams reinforced by CFRP, then establish forecasting(?)models about flexural capacity of RC beams reinforced by CFRP which are based on the experimental data with using different intelligent algorithms, such as BP neural network, RBF n(?)ural network, genetic algorithm and least squares support vector machine, and compare the ad(?) antages and disadvantages of them. By analyzing and verifying the results, it proves that the (?)olutions of structural engineering problems by using intelligent algorithms is feasible and effective, particularly the RBF neural network and least squares support vector machine are (?)ot only computing speedy and generalizing well, but also have high accuracy of forecasting and good prospects about engineering application.At last, the main contributions and conclusions of this dissertation are summarized and some problems to be addressed about the research are set forwar...