Study Of Intelligent Diagnostic Mode Of The Composite Structure

Through the nonlinear analysis of loading properties about continuous composite constructs under two-stage, the author obtained that this composite construct's cases of stress excess of tensile steel are more serious, and the amounts of strain lag of added concrete are smaller than single supported composite beam's .These cases cause faster increase of curvature, crack and deflection under fist-stage. Additional with using method of finite element this paper shows that the continuous composite constructs have these properties of moment excess in middle span and stress lag of negative steel in abutment 0 These properties result in plastic redistribution of section internal force between middle span and abutment that enhance flexure strength, crack-resistance and ductility. After acquaint these theory, the author Applied theory of gray correlation analysis to analysis some factors that influence the internal force transfusion behavior for composite construct such as proportion of valid height under two-stage hoi/ho. moment ration in first-stage MI/MIU, moment ration in second-stage M2/M2u as well as proportion of steel areas P/P玿o The analysis result shows that the most important correlation factor is proportion of valid height under two-stage hoi/ho, the least influence factor is proportion of steel areas P / P ^,0 On the other hand Considering properties of composite constructs under two-stage loading that its amounts of curvature, crack and deflection are more excessive than reinforced concrete and stressed concrete's at the same parameters, The author applied neural network to estimate durability of composite construct. The result has indicated that this method can be converged in fast speed by mean of dynamic learning algorithm and excellence in aping, foreseeing accurately crack and deflection of composite construct.