Homology Design On Reflector And Electromechanical Optimization Of Large Antenna

This work was supported by a grant from the National Program on Key BasicResearch Project (973Program). On the basis of the predecessor work, this thesis ismainly concerned with the homology design and electromechanical syntheticoptimization of large radar and antenna. The main research work can be described asfollows.1. On the degradation of electrical performance due to the main reflectordeformation of large shaped Cassegrain antennas, a method for compensation bymoving sub-reflector is presented. First of all, a group of best-fit paraboloids are foundby least-square fitting to the theoretical discrete data. Second, the group of paraboloidsis used to fit the deformed main reflector, with the constraint of all these focuses beingin line. At last, the best-fit parameters are optimized and the adjustments of sub-reflectorare derived with the ratio of main reflector and sub-reflector.2. Based on the field-coupling model about the electromagnetic field and thestructural displacement field, a topology electromechanical synthetic optimization ofreflector antenna is studied. The selection criterion of ground structure is proposed forthe topology optimization of antenna structure. This chapter discusses the method andprocess of the field-coupling model in detail, and points out the problem of adding thetopology variables into the optimization model. The main contribution of this chapter isto establish the optimization design model treating the antenna electrical properties astarget, taking the size, shape, topology of antenna structure and compensation quantityof sub-reflector as design variables. At the same time, the constraints of weight, strengthis satisfied.3. Simulation results of65m antenna show that electrical performances areimproved by the electro-mechanical integrated optimization model. The greatest benefitafter the topology change is to flat the phase error of aperture, thus the side-lobes arereduced, so as to reach the engineering design requirements. At the same time, the gainof antenna is satisfied.4. The intersection elements exist in the result of topology optimizationfrequently. It is not reasonable to delete the intersection elements. It will give thestructure of manufacturing and assembly inconvenience. Generally speaking, theintersection is avoided for engineering design. The contribution is that elementintersection is described in terms of a continuous intersection factor. A Heaviside function is used to map element cross-section area to intersection properties. Therefore,the intersection feature is described by a continuous and differentiable function.According to this, the intersection constraint is added to the topology optimization couldbe calculated by mature mathematical programming method.5. Considering the inadequacies of the method about multi-discrete structuraloptimization at present. This part constructs a new mathematical model of discretevariables. The ultimate optimal results fall near the discrete values through the mappingof the continuous variables. A piecewise function is constructed to punish thecontinuous variables on each interval of all adjacent discrete variables. The solvingscale is smaller compared to the branch and bound method. Finally, the problem of amulti-variable topology optimization is transformed into a general nonlinearoptimization problem. Finally, case verification is conducted on the largest phased arrayradar and a better optimization result is obtained. The design will be applied toengineering.