Assessment of concrete compressive strength after fire based on evolutionary neural network

Assessment of concrete compressive strength is one of the most essential tasks in the damage degree and bearing capacity diagnosis and identification of concrete structure damaged by fire. An evolutionary algorithm radial basis function neural network model (EARBFNN) optimized was introduced to assessing concrete compressive strength, and an ultrasonic and rebound combined method is adopted to collect original experiment data for concrete component after fire. At last, a regressive calculation is applied to comparing the assessment effect with the EARBFNN method, and the experimental test and simulation analysis result has proven that EARBFNN has higher precision than that of regressive calculation.