| Reinforced concrete structure is widely used because it bands concrete's and steel bar's advantage, and its cost are not very high. Concrete's durability is the characteristic of itself working ability in usual circumstance, using circumstance and inner factors, or ability of resisting outer circumstance and destruction of inner action.This issue is analyzed the reliability of reinforced concrete component by using neural networks method and referencing the degree of steel corrosion in concrete. The dissertation is divided into three parts. The first part is on the analysis of concrete's durability; in the second part, reinforced concrete structure's reliability is studied on reliability theory; and in the third part, steel corrosion is predicted based on BP neural networks.The main content of the paper is as follows:(1) Concrete durability's background and research situation is introduced, and is summarized from material (freeze-thaw, carbonation, steel corrosion and so on), component and structure, the directions of further developing is also discussed.(2) The structural reliability theory and the calculation methods of degree of reliability are introduced; the reliability of reinforced concrete structure in different time is analyzed. The durability and service life prediction of existing structures are studied; it not only provides the foundation in maintenance and reinforcement, but also its result can be used directly.(3) The development of neural networks and its basic theory are introduced, and the BP neural networks and its defect are mainly studied. The quantity of steel corrosion is predicted based on BP neural networks, and it provides foundation for structural durability prediction. |
