Research on Debonding Degree of Building Exterior Wall Bricks Based on ISSA-KELM Algorithm

In recent years, the building exterior wall tiles falling off caused a lot of damage to property and injuries, in order to achieve more accurate off, put forward using percussion method to extract the sound signal characteristics combined with pattern recognition method for building exterior wall tiles debonding degree were studied. In order to improve the prediction accuracy of the debonding degree of building exterior brick, a prediction method of the debonding degree of building exterior brick based on the improved sparrow search algorithm (ISSA) and optimized kernel extreme learning machine (KELM) was proposed. In order to overcome the disadvantages of the standard sparrow search algorithm(SSA), such as uneven population distribution and easy to fall into local optimum, the algorithm firstly uses Circle chaotic map to generate the initial population with uniform distribution, and then uses the elite opposition-based(EOBL) learning strategy to broaden the search domain of the algorithm, obtain elite sparrows, avoid algorithms falling into local optimality. The KELM parameters were optimized by ISSA algorithm, and a prediction model based on the degree of brick debonding outside KELM was established. The characteristic data of sound signal collected from building exterior brick of different degree of debonding were used to study. The results indicate that the proposed method is superior KELM, PSO-KELM, GWO-KELM and SSA-KELM algorithms, and has certain applicability to detect the degree of debonding of building exterior brick.