基于稀疏采样的FDTD/TDPO混合优化算法

      Hybrid optimization algorithm of FDTD/TDPO based on sparse sampling

      • 摘要: 基于时域有限差分(finite-difference time-domain, FDTD)和时域物理光学(time-domain physical optics, TDPO)的混合算法提出用稀疏采样优化近场外推的方法,解决了大口径反射面天线电磁计算用时过长、计算效率低的问题。传统混合算法中馈源的近场外推到反射面天线的步骤计算量较大,稀疏采样法从空间和时间两个维度进行间隔采样,减少了此步骤的计算量,从而节省了用时。给出了馈源为喇叭天线、抛物面口径为100个波长的抛物面天线算例,所得计算结果与传统方法相比最大相对误差仅有−36.03 dB,用时最快可缩短为原来的2.51%,与CST软件计算结果对比吻合度较高,用时缩减了93.18%,验证了本文优化算法可在不影响计算精度的前提下提高计算效率。

         

        Abstract: Based on hybrid method of finite-difference time-domain (FDTD) and time-domain physical optics (TDPO), this study applies sparse sampling to near- and far-field calculations to improve the efficiency of large aperture reflector antennas computation. In the conventional hybrid method, near-field of feed to far-field of reflector antenna transformation has largest calculation amount, and it will be reduced by sparse sampling optimization method. This method samples at intervals on time domain and space which is much faster. The example of large parabolic antenna whose aperture size is 100 wave length fed by horn is given. Comparing with the results of traditional method, the maximum relative error is only −36.03 dB, while the time is reduced to 2.51%. The computing results are in good agreement with those by CST software, the time is reduced by 93.18%, which demonstrates that the proposed method has small error and high computational efficiency.

         

      /

      返回文章
      返回