Study On Physical-Based Simulation Of Antenna Reflector Assembly Optimization Technology

Large reflector antenna has large radius,numbers of parts and the significant impact of gravity,which make the antenna relfector easy to be distorted and impair its electrical performance.Due to the above reasons,the development will be more difficult and the longer development cycle will be required to ensure the performance of large reflector antenna.In engineering,the development of antenna usually focuses on its design stage,which will improve the lack of advanced manufacturing technologies and the assembly deformation.It only can be improved through adjusting the panels when the reflector is insufficient precision because of being deformed seriously during the assembly stage,which will extend the antenna assembly cycle and increase of assembly manual consumption.Based on the observation and analysis of the assembly process of large reflector antenna in the actual engineering,the deformation of the reflector surface in the assembly process and the possible factors leading to the deformation are deduced.On the basis of the above researches,the finite element method is applied to simulate the assembly of the reflector before the assembly stage.And the assembly of the reflector is further optimized by using the assembly sequence planning method.The main content of this dissertation is summarized as follows.1.The idea of the reflector assembly optimization considering finite element method(FEM),intelligent algorithms(IAs)and assembly sequence planning(ASP)is established based on the description of the research objectives and research thinking of physics-based simulation.The basic architecture of the simulation model of reflector assembly is given,the proper intelligent algorithms is introduced to the ASP of reflector.And the key technologies involved are briefly described.2.Through the observation,comparison and analysis of the existing large-scale reflector assembly process,a finite element dynamic simulation method is proposed,which is verified by the simulation of the single panel and the whole reflector.The finite element simulation method of the reflector surface assembly is established,and the main factors influencing the deformation of the reflector surface are considered.The “life-death element method” is introduced to optimize the simulation model of reflector surface assembly which can achieve the dynamic simulation of reflector assembly with smaller modeling work and computerized calculation.Compared with the traditional finite element simulation method,this simulation model is more accurately,and a more accurate and detailed simulation results of the assembly surface deformation can be obtained.3.Based on the understanding of the actual engineering reflector assembly method,the ASP method of the reflector based on genetic algorithm(GA)is proposed,and the example is calculated and the corresponding finite element simulation is used to verify the correctness and validity of the algorithm.The GA of the reflector assembly is selected to sovel the ASP problems of reflector after the comparison of several intelligent algorithms.According to the characteristics of the reflector panels and the factors influencing the panel assembly deformation,the genome and the evaluation system,etc.of the genetic algorithm are established.The parent generations is selected by “gambling roulette” method.Crossover coefficient and mutation coefficient are calculated by the equations created according to the characteristics of reflector assembly.the iteration termination conditions are determined.The mathematical model of the genetic algorithm for ASP problems of the reflector is given.and the assembly sequence planning method is obtained.4.Based on the GA of the reflector assembly,an algorithm for ASP of reflector based on ant colony algorithm(ACA),GA and FEM is eatablished and this hybrid algorithm is verified more smart and perfect.Considering computational characteristics of the GA,the ACA with parallelism,positive feedback mechanism and high efficiency is introduced in the GA to create a hybrid algorihm-ant colony and genetic algorithm(ACGA),which makes the ASP of reflector assembly further optimized.Based on the GA,the ant colony algorithm is used to obtain the initial population of GA,which improves the diversity of the initial population and the efficiency of the late search.Each ant chooses the initial feasible assembly node to start the tour.The initial solution is obtained after completing a travel,which can be as the initial population of the GA.the optimal solution can be transformed into the initial pheromone distribution of the ant colony algorithm when the iteration termination condition is not satisfied.The ant colony algorithm will process the pheromone and start the next tour.The ACGA algorithm is applied to the ASP of reflector.The optimal solution of ACGA is employed the finite element simulation model of the antenna reflector.The simulation results show that the performance of the algorithm is superior to the genetic algorithm and the ant colony algorithm,and also avoids the impact of the overall performance due to the premature or late finish of the GA.5.Based on the above ideas,a finite element simulation model of reflector is established by taking a reflector antenna as an example.Through simulation calculation,the reflector deformation influenced by the factors including pre-tightening forces of screws and assembly sequences is obtained.The mathematic model of the GA and the ACGA for the ASP of the reflector are established.The optimal solutions obtained from the two algorithms are conducted to the finite element simulate model,which can verify that the ACGA is superior to the GA.The optimal solution of ACGA is compared with the actual engineering measurement of the reflector deformation,which verify that the ASP of the reflector can improve the deformation of the reflector and ensure the precision of the reflector.