基于约束多目标优化算法的盾构隧道设计方法

Constrained multi-objective optimization algorithm based design method of shield tunnel

当前盾构隧道的设计仍主要依托于经验确定设计参数,然后对其安全性进行校核,难以定量的考虑成本和控制指标(收敛变形等)的关系。以衬砌厚度、截面配筋率、横向接头螺栓直径为设计参数,以成本及结构的水平收敛变形为优化目标,结合多目标优化的算法,进行了盾构隧道的横断面设计。采用非支配原理,通过引入约束违反函数,实现了基于NSGA-Ⅱ算法的复杂约束条件处理。最后通过具体算例,完成了满足截面安全性要求的限制条件下,完整Pareto前沿面的获取工作,并与无约束优化结果进行了分析对比,说明了迭代过程的收敛性,阐述了所得Pareto前沿面的价值意义,得到了不同设计条件下的最优解。

The design of shield tunnels still mainly depends on experience to choose design parameters,then check the structure safety,which results in the difficulty to consider the relationship between cost and control indicators(tunnel convergence deformation,etc.)quantitatively.Takes lining thickness,steel reinforcement ratio,bolt diameter of circumferential joints as design parameters,costs and lateral convergence deformation of the tunnel as optimization targets,and combines the principle algorithm of multi-objective optimization to design the shield tunnel.Based on the NSGA-Ⅱalgorithm,the non-dominated principle is used to deal with the complex constraints,which is realized by adopting a constraint violation function.Finally,an illustrative example is introduced and the complete Pareto front is obtained while the safety constraints are satisfied.Then the difference of the result between constraint and unconstraint situations is analyzed,the convergence of the iterative process is expressed,then the value of Pareto front is discussed and the optimal solutions in different conditions are listed and compared.