Comparison of Particle Swarm Optimization and Simulated Annealing for Weight Optimization of Composite Leaf Spring

Nowadays weight optimization is increasingly becoming an important tool for manufacturing and mechanical design. A formulation and solution technique using Particle Swarm Optimization (PSO) and Simulated Annealing (SA) for design optimization of Composite Leaf Springs is presented in this work. Leaf Springs are long and narrow plates attached to the frame of a trailer that rest above or below the trailer's axle. This paper aims at minimizing the weight of leaf spring subjected to certain constraints. The dimensions of an existing conventional leaf spring of a light commercial vehicle are used to design mono composite (E- Glass epoxy) leaf spring which is of great interest to the transportation industry. The constant cross-section design is used due to its capability for mass production and to accommodate continuous reinforcement of fibres. The design constraints are bending stresses and deflection. Compared to the steel spring, the composite spring has stresses and deflection that are much lower, and the spring weight is nearly 85.02% lower using Particle Swarm Optimization and 78.87% lower using Simulated Annealing. From the results, it is observed that the composite leaf spring is lighter and more economical than the conventional steel spring with similar design specifications and two techniques are compared which shows that Particle Swarm Optimization has outperformed Simulated Annealing.