Rotating Surface Conformal Array Optimization Algorithm

Conformal arrays antenna can be formed by installing the antenna element on the aircraft surface. Using conformal arrays antenna can achieve greater aerodynamic performance and improve weapons'performance. The aperture of conformal array should be big enough to keep high-resolution, and fewer elements should be needed to reduce the cost of antenna system. Sparse array study how to make elements'layout as few as possible with big aperture to let the performance of array antenna be better.The peak side lobe level (PSLL) is a key parameter of an antenna arrays, and it's a complicated and nonlinear function concering the position of elements. The optimization of the element positions of the sparse antenna arrays is a nonlinear optimization, and it's hardly to find the best position of elements using traditional numerical optimization methods. Simulated annealing algorithm(SA) and genetic algorithm(GA) are global random optimized methods, and they only need the fitness function's imformation. They adapt to muti-dimensional problems which are incontinuous, and have many locals.Conformal array is combined with sparse array in this paper. To aim at some complicated conformal array models——conical array, cylindrical array and concentric rings array, the position of elements is optimized to reduce the PSLL with GA and SA. The main details and contributions of this dissertation include the following several aspects:1. Introduce the genetic algorithm and simulated annealing algorithm's basic concept, principle, algorithm flow and its application.2. Pattern functions of arbitrary arrays are deduced. A fitness function evaluating the PSLL on the whole plane is given base on the the pattern function of conical array. Using SA makes a thinning layout of conical array to reduce its PSLL.3. SA is used in the synthesis problem of sparse cylindrical antenna arrays. There are multiple constraints including the aperture of array, the number of elements and the minimum element spacing. Cylindrical coordinate system is used in the optimization process to reduce computation burden. 4. To aim at the plane array with circular boundary, a new sparse method is given. Elements can be distributed non-uniformly using this method, and it can make the layout of elements more free. Lower PSLL can be obtained by finding the best element position using modified GA.