Parameter Analysis Of Improved Ant Colony Algorithms And Their Application In Optimal Configuration Of Dampers

As a kind of modern intelligent optimization algorithm,the essence of ant colony algorithm is a kind of pseudo random search and parallel algorithm.It doesn't depend on gradient information of objective function to solve problems,which improve people's ability to deal with the challenge of practical engineering optimization through heuristic function and the information positive feedback mechanism of the algorithm to guide the the whole optimization process of ant colony,getting a better solution in the problems of complex and difficult to use conventional optimization methods for processing.It is well known that the value of key parameters in the ant colony algorithm has a crucial influence on the optimization performance of the algorithm.Good parameter settings can effectively enhance the algorithm's global search capabilities,thereby increasing the convergence speed and reducing the calculation time.At present,there is no strict theoretical basis for parameter setting in ant colony algorithm,and it is difficult to determine the optimal combination of parameter setting by analytical method.Aiming at the specific engineering optimization problems,the key parameters of ant colony algorithm are analyzed and applied for solving them,which is a current research hotspot.Ant colony algorithm can be used to solve the problem of damper optimization in high-rise building structure.However,when the layers of building is higher and the design space is getting bigger,the number of dampers needed to be installed also increase.The deficiency of low optimization efficiency,non-convergence and trapping into local optimum may appear when the basic ant colony algorithm is used to solve the problems.Therefore,the key parameters such as the number of elite ants?,weight coefficient of?,pheromone smoothing coefficient?of the elite strategy of ant system(AS_elite),sorting weighting of ant colony algorithm(ACO_rw),and maximum minimum ant colony algorithm?MMAS?are analysed for multiple sets of numerical simulation in this thesis so as to determine the value range of key parameters and propose the scheme to improve the solution through parameter setting,then further apply it to the optimization arrangement of dampers in high-rise building structures.In order to select the key parameters of ant colony algorithms,and compare optimization performance among the improved ant colony algorithms for the damper layout.Parameter performance analysis was studied for the basic ant colony algorithm at first in this thesis,and applied it to solve the optimal placement of the dampers.The results show that the improved ant colony algorithms would speed up the convergence of the algorithm and improve overall quality of the solution when analyzing these optimization problem.The main tasks of this thesis are as follows:Firstly,the performance,role and selection principle of of the relevantvparameters were analyzed based on the basic ant colony algorithm for the key parameters selecting and numerical verification experiments to obtain reasonable values of key parameters in the basic ant colony algorithm.Then,the basic ant colony algorithm was used to solve the damper layout problem of a 20-story frame structure.The numerical example showed that the algorithm and parameter values were feasible and effective.Secondly,the key parameters of the three improved ant colony algorithms such as ASelite,ACOrw,and MMAS were simulated and tested to get reasonable range value.Then they were applied to a 22-layer frame structure respectively,and compared the optimization results of the three improved ant colony algorithms.In order to compare optimization performance of the improved ant colony algorithms,taking into account the effects of the Northridge-ew,EL Centro,and TANGSHAN-ew seismic waves,the basic ant colony algorithm and three improved ant colony algorithms were applied to the optimal placement of dampers in 12-story,16-story 18-story frame structures based on the same parameters.The results show that the improved ant colony algorithms can make the best use of its own advantages and attract more ants to move towards better paths,so as to find the optimal solution faster and improve overall quality of the solution.Lastly,considering the requirements of safety and comfort of the structure,the combination of three-level response index such as structural layer displacement,interlayer displacement and acceleration under different weighting coefficients were used as the objective function for the optimization problem,and the corresponding heuristic function was constructed.Taking into account the Northridge-ew,EL Centro,and TANGSHAN-ew seismic wave effects,the dampers optimal placement of the 16-story and 20-story frame structures was performed through MMAS.The optimization results based on the objective functions under different weighting coefficients combinations were also discussed.In order to make the optimization objectives comparable,the objective function was normalized,and an optimized layout scheme for high-rise building structure dampers that met the practical engineering significance was obtained.In this thesis,the reasonable range of key parameters was obtained through the parameters optimization analysis of ant colony algorithms.The ant colony algorithms were applied to solve the optimization problem of dampers in high-rise building structures.The results of these examples show that the ant colony algorithms were feasible and effective for the solusion of damper optimization layout problem,which has a good practical value in engineering.