Research On Multi-objective Morphological Optimization Control Of Adaptive Ribbed Cable Dome Structure

With the increasing maturity of artificial intelligence theories and technologies,the concept of intelligent structures has begun to emerge and develop in the field of civil engineering.The characteristics of the cable dome structure,such as light weight and easy shape control,were the key attributes necessary for intelligent structure.Thus,it was of great significance to study the adaptive cable dome structure which introduced the concept of intelligent structure.Actuators which could vary members’ length actively were introduced into ribbed cable dome with diameter of 120 m for shape adjustment and reducing the interior forces.Starting from the basic theory of the cable dome structure,force optimization control,shape optimization control,multi-variable morphological optimization control and morphological multi-objective synchronization optimization control under load were systematically studied.Firstly,based on the nonlinear finite element analysis theory of the cable dome structure,a MATLAB calculation program for iteratively solving the structural response under structural element length changes was written.After that,using the finite element software ANSYS to verify the program results.Draws a conclusion that the program results are valid and accurate.Then,through the calculation of the examples,the rule of structural response and the length change of rod element and cable element were summarized.Secondly,based on multi-population genetic algorithm and particle swarm optimization algorithm,force and shape optimization control were studied.Aiming at the optimization control of force and shape under load,and using the momentum of actuator as independent variable,the mathematical model of force and shape optimization control was established.The solution strategy based on these two algorithms is proposed and the corresponding program was compiled.Besides,through an example,the advantages and disadvantages of the two algorithms were compared,and the optimal actuator layout in the force and shape optimization control was obtained.In addition,it summarized the influence of the number of control units and the position of control units on the optimization effect,which provided theoretical guidance for designers to select control units.Thirdly,based on the force and shape optimization control,the position variable of the control unit was introduced as the independent variable,and the multi-variable morphological optimization control research was carried out.In addition to considering the necessary constraints,the morphological control requirements were added as constraints,a mathematical model was established.And a loop traversal solution strategy based on particle swarm optimization was proposed to verify the applicability of the algorithm.The force optimization was specifically used as the optimization goal,and the morphological control requirements were set to perform an example analysis,where the optimal solution that considers all possible actuator layout schemes in the three cases of one type,two types,and three types of unit regulation was obtained respectively.That was,the inner ring notch,inner ring notch + middle ring stay cable,middle ring notch + middle ring stay cable + outer ring notch,respectively,which could reduce the working state coefficient of the structure by 17.36%,21.45%,30.29% compared with the load state.Finally,the influence of the number of control units and the position of control units on the results was discussed.The multi-objective mathematical model was established which Taking internal force optimization control and shape optimization control as two optimization objectives for simultaneous optimization.And the solution strategy of NSGA-Ⅱbased on Pareto optimal solution mechanism was proposed.Based on that,the discrete variable-control unit position vector was added to the mathematical model,with the establishment of a multi-variable morphological multi-objective simultaneous optimization mathematical model,and an outer traversal cycle was proposed.The inner multi-objective optimization and the final nondominated sorting strategy were based on the improved NSGA-Ⅱ algorithm.The practical example verified the practicability and accuracy of the two solving strategies.What’s more,the Pareto optimal solution set under the two optimization models were obtained respectively.After comparing and analyzing the results,the influence of the number of control units on the frontier of Pareto was summarized,which provided theoretical support for decision makers to make reasonable choices according to their own needs.