Optimal Flexural Design of FRP-Reinforced Concrete Beams Using a Particle Swarm Optimizer
This addresses the problem of efficient and cost-effective design for engineers working with FRP-reinforced concrete beams, but it is incremental as it applies an existing AI technique to a specific domain.
The paper tackled the iterative design of FRP-reinforced concrete beams by developing a least-cost section model based on ACI 440.1 R-06 recommendations and using particle swarm optimization (PSO) to handle optimization tasks, resulting in a method to find suitable solutions among infinite possibilities.
The design of the cross-section of an FRP-reinforced concrete beam is an iterative process of estimating both its dimensions and the reinforcement ratio, followed by the check of the compliance of a number of strength and serviceability constraints. The process continues until a suitable solution is found. Since there are infinite solutions to the problem, it appears convenient to define some optimality criteria so as to measure the relative goodness of the different solutions. This paper intends to develop a preliminary least-cost section design model that follows the recommendations in the ACI 440.1 R-06, and uses a relatively new artificial intelligence technique called particle swarm optimization (PSO) to handle the optimization tasks. The latter is based on the intelligence that emerges from the low-level interactions among a number of relatively non-intelligent individuals within a population.