Improved Simulated Annealing Algorithm And Its Application In Seismic Control Performance Optimization Of Nonlinear Energy Sinks

The improved Simulated Annealing(Simulated Annealing,SA)optimization algorithm has put forward for looking for the key design parameter of Nonlinear Energy Sink(NES).The initial temperature and annealing factor of the cooling schedule in SA algorithm are trained by the two-stage algorithm process,and the cooling schedule of SA algorithm is better suited to the optimization of NES's parameters.The two-stage algorithm process can solve the problem that the iterative convergence is slow and easy to get into local optimum in the process of large space optimization with NES design parameters.Firstly,the fast interpolation get the optimal parameters of the small design parameter space for obtaining preliminary character of solution space and initial parameters in the second stage of large space SA.It can train more efficient Cooling schedule parameters and Set the control threshold lead to greatly reduce the risk of local optimum,and obtain the optimal solution of the whole design space with high efficiency and stability.Based on the improved SA algorithm,the following several kinds of NES damping control system are optimized.(1)In the first place,under the excitation of the pulse,the single degree-of-freedom system with the negative stiffness NES control system is optimized by using the improved SA algorithm for its key parameters.Compared with the traditional SA algorithm,the improved SA algorithm is better than the traditional SA algorithm in the convergence performance and the global optimal solution.Secondly,on the basis of optimization and further analysis of the in pulse and seismic load vibration reduction performance of two kinds of different incentive,compared with type I NES,TMD,numerical display the negative stiffness NES overall better damping performance.Especially when the main structural characteristics changed,its control effect is more obvious;In the end,the vibration reduction performance of the two degrees of freedom under seismic acceleration wave excitation is analyzed,and similar results are obtained.(2)For the 20 floors Benchmark model,the negative stiffness NES has a good performance in controlling the main structure when the stiffness of main structure changes and to TMD,when acceleration seismic wave peak value change,its control effect will not be affected.To combine the advantages of two kinds of control devices,a combined control device with negative stiffness NES and TMD is proposed.For this parallel control device,more parameters need to be optimized,and the optimization solution space is larger.The improved SA algorithm is used to optimize the parameters.The parallel control strategy is revealed and has the advantage of NES and TMD.It has good control performance both in stiffness robustness and seismic peak value robustness,and it also verifies the effectiveness of the improved SA algorithm.(3)For the Vibro-Impact NES model,an improved NES form is proposed,and then the numerical modeling of the mechanical model,the optimization parameters and the design of the seismic excitation wave are carried out.The improved SA is used to optimize the key parameters of the improved Vibro-Impact NES.The improved Vibro-Impact NES with different mass ratios is arranged on the top layer of the four storey frame main structure.The analysis of the evaluation index shows that the improved Vibro-Impact NES has a certain control effect on the main structure,and reveals that the improved Vibro-Impact NES is better than that before the improvement of Vibro-Impact NES.