Genetic Algorithms in Optimizing Transient Suppression Devices

Analysis and control of pressure transients in pipelines continues to be a tedious process, despite availability of sophisticated modeling packages. The complexity is magnified when the transient suppression devices, such as surge tanks, need to be designed for minimum cost. The paper describes a surge tank design model that obtains an optimal set of decision variables while satisfying a specified set of pressure constraints. The model was developed based on a bi-level optimization framework and it employs a genetic algorithm in minimizing a nonlinear objective function. The model was applied to optimize sizes of two surge tanks on an example water distribution network. Enumeration of the entire solution space demonstrated that the solution obtained by the genetic optimization model is close to the global optimal solution, while requiring only a fraction of computational time that was necessary for enumeration.