optimization algorithms

Joints are considered the weakest part of an engineering structure and failure usually occurs in this region, firstly. One of the main factors in the rupture of adhesive joints is the normal stresses between the layers created by the presence of an out-of-center load and bending moment. The present research work has focused on the influence of parameters including the adhesive zone length, adhesive and adherend layer thickness on reducing the amount of normal interlayer stress in a single-lap adhesive joint. Optimization of parameters have been done using BA and PSO optimization algorithm. The distribution of normal and shear stresses are based on two-dimensional elasticity theory that includes the complete stress-strain and strain-displacement relations for the adhesive and adherends. The results obtained from current research revealed that by optimization of mentioned parameters, the value of peeling stress is significantly reduced. Although increasing in Young’s modulus of adhesive layer leads to an increase in normal stress of the joint, it creates a more uniform stress distribution at the edges. The outcomes also revealed that increasing the length of the joint zone and the thickness of adherends can improve the interlayer normal stress in the adhesive joint.

In this paper, a new approach to the use of genetic algorithms and the predictive control method, for goal tracking is presented. A hypothetical rocket is modelled for the analyses. Rocket guidance algorithm is developed to achieve a desired mission goal according to some performance criteria and the imposed constraints. Given that goals can be fixed or moving, we have focused and expanded on this issue in this study and also the dynamic modelling of flying objects with six-degrees-of-freedom (DOF) is used to make the design more similar to the actual model. The predictive control method is used to predict the next step of rocket and aim movement. At each step of the problem, the rocket distance to the aim is obtained, and a trajectory is predicted to move the rocket towards the purpose. The objective function of this problem, in addition to the distance from the rocket position to the target, are also parameters of the dynamic model of the rocket. Therefore, these parameters are optimized at each step of the problem solving. Ultimately, the rocket strikes the intended aim by following this optimal path. Finally, for the validation of the model, numerical results are obtained for both Genetic Algorithms (GA) and Particle Swarm Optimization (PSO). Simulation results demonstrate the effectiveness and feasibility of the proposed optimization technique.

In this study, five different heat exchangers (HE) including, plate-fin (PFHE), fin tube (FTHE), rotary regenerator (RR), shell and tube (STHE) and gasket plate (GPHE) are optimized using four different algorithms including the binary genetic algorithm (BGA), real parameter genetic algorithm (RGA), particle swarm optimization (PSO) algorithm and the differential evolution (DE) algorithm. Verified codes are used for all heat exchangers and total annual cost (TAC) is considered as the objective function and heat exchanger configuration parameters are chosen as design parameters in all studied exchangers. RGA has the lowest insensitivity to the algorithm input parameters, or lowest relative standard deviation (RSD), for all studied heat exchangers. The best TAC in the GPHE, FTHE, PFHE, RR, STHE can be achieved in the points <C1, C2> = (0,0.6), (0, 1.95), (0, 1.5), (0, 2.1), (0, 1.65) and < , > = (2.4,2.4), (1, 2.4), (3.25, 3.75), (3.15, 3), (2.6, 2.8) where  the lowest run- time and RSD are our basic requirements, respectively. The results also reveal that DE has the worst result in the case of RSD and GA has the worst result in the case of run-time. Finally, RGA is recommended for the optimization of different types of heat exchangers.