Prediction Of Early Mechanical Properties Of Bag Concrete Based On BP Neural Network

The mechanical properties of early-age concrete bag is one of the key factors that affect the structural safety during the construction period,and it is also the restriction factor of the project progress.In this study,the solid waste from the jade factory and power plant near Bayannur City,Inner Mongolia,was introduced into the form bag concrete to replace some cement,and the Mechanical Properties and Nuclear Magnetic Resonance(N)were studied.In this paper,the early mechanical properties and micro-pore structure of bag concrete with different mineral admixtures are studied.On this basis,grey relational grade analysis and BP neural network theory are introduced to study the relationship between the early mechanical properties of concrete with different proportion of admixtures and the mixture ratio and pore structure,a prediction model of early compressive strength of bag concrete based on BP neural network is established.The reasonable use of solid waste,reduce the amount of cement,reduce environmental pollution,but also to optimize the mixture ratio of bag concrete in Hetao area to provide theoretical support.The main research results are as follows:(1)Macroscopical early mechanical experiment results show that reasonable addition of fly ash and Silica fume instead of cement can improve the early compressive strength and elastic modulus of bag concrete.The compressive strength decreases with the increase of the content of Silica fume,and increases at first and then decreases with the increase of the content of fly ash The early compressive strength of F15S4 is higher than that of other groups,and the early compressive strength of F15S4 is higher than that of other groups.The Stress-strain curve shows that F15S4 has a higher modulus of elasticity.(2)Ntest shows that proper amount of fly ash and silica fume can improve the pore structure of concrete.After the reasonable addition of fly ash and silica fume instead of cement,the pore area and porosity of the mould bag concrete decrease.The experimental results show that the porosity and Pore area of F15S4 group are lower than those of other groups.With the increase of age,the GEL holes in each group were gradually transformed into capillary and non-capillary.The Gel holes and capillary holes in the double-doped group were higher than those in the single-doped fly Ash Group,and the GEL holes of F15S4 were the most.(3)The correlation coefficient between the mixture ratio and the compressive strength,the correlation Coefficient between the Pore Structure Parameters and the compressive strength are obtained by the grey correlation analysis.The maximum correlation between cement content and early compressive strength of concrete is porosity,and that between 7D and 14d compressive strength is porosity,and that between 28d compressive strength is pore radius,which is between 0.1 m and 5 M.(4)Combined with BP neural network theory,the prediction model of early compressive strength of concrete with mould bag is established.The prediction models of early compressive strength of concrete with mixed ratio and pore structure parameters are established respectively,and it is found that the precision of prediction models is high.