Abstract:

For the prefabricated components of cable-stayed bridges manufactured according to established procedures, how to achieve coordination in assembly under adjusted new processes is often a challenge faced when making decisions based on overall progress. By integrating the stress-free state method and tolerance allocation method, this paper first constructs an inverse method for the assembly tolerance interval of cable-stayed bridges based on machine learning, and inversely calculates the passive assembly tolerance intervals for each component of a certain cable-stayed bridge based on prior error reserves. Integrating the actual measurement data from previous engineering sections, a framework for active fault-tolerant assembly tolerance design of cable-stayed bridges is constructed. The results show that the interval inversion method can effectively improve the operability of the construction site while ensuring structural safety and design optimality. Compared with passive tolerance design, the active tolerance design method can increase the main beam splicing angle tolerance interval by 1.4 times (G2 beam section) and the stress-free cable length tolerance interval by 2.1 times (cable S16). The active tolerance analysis framework can adaptively adjust the tolerance range of subsequent components to assemble failure scenarios, reducing the occurrence of work stoppages and reworks.

Key words: cable-stayed bridge, proactive fault tolerance, assembly tolerance, construction control, adaptive adjustment