[数据处理]CST MWS仿真方向图每调整一次频率点都要重新仿真一次，这也太麻烦了吧。

如题。用CST仿真方向图的时候，设置了几个频率点。后来想要仿真增益，觉得频率点太稀疏了，增加了几个频率点，又要重新仿真才能得到结果。HFSS只要仿真完毕，想要得到方向图和增益都只要在results->creat report中设置就行了，不需要重新计算。请问是不是我对CST的设置有问题。我用的是CST2010。有没有简单的办法，用CST计算一次，以后就 .. ZMSP8(V  
q@;z((45  

未注册仅能浏览部分内容，查看全部内容及附件请先 登录 或 注册

楼主确定HFSS的results-creat report是一个准确的输出结果的方式？即使没有在一开始仿真的时候设置监视器？我的感觉是这个结果是否是通过已有场数据经过数值拟合计算得来的呢？ vB, X)  
8Pklw^k  
CST确实是这样，就是如果增加了频点，必须得重新计算。这个跟算法有关，跟计算时储存的结果有关。 RRy3N )HR  
t#mW`rGE_  
但是话说回来，我觉得CST应该有那种插值拟合场结果的算法，只是未必准确。。。

用时域不行吗？你可以用时域设置个宽带的远场监视器试试，插值的精度可以定义的，1e-5够了吧?

的确没有听说CST有插值计算场的方法 ?7]G)8G6  
O/U?Wq  
HSWki';G  
话说回来： OwT_W)$  
1.CST是时域算法，增加监视器数量对其仿真时间并不影响或者影响非常小（影响的主要是存贮空间），因而你如果需要比较多的频点预先多设几个Monitor即可。你仿真前不会连自己需要哪些频点的Monitor都不知道吧 UEx13!iFo  
Y7zs)W8xTT  
2.内插法毕竟存在误差，考虑到前面那一条，对CST来说好像真的没那么必要

有宽带的远场监视器啊。

OK。学习了，谢谢f1381584904童鞋~~ p t{/|P  
j|% C?N  
根据Help文件 Solve——>Monitors，Broadband Farfield Monitor即为楼主所需的。其采用球面波展开的方法计算远场，Accuracy决定了其球面波展开式的阶数，而Freq。Sample决定了瞬态求解远场采样点的数量，其余频点远场将通过内插得到~~ .9bP8u2B{  
(P:.@P~  
Broadband: If a farfield monitor was selected in the type frame a broadband farfield can be calculated during the transient solver run. The broadband farfield monitor is based on an expansion of the farfield in terms of spherical waves. The origin of the expansion is the center of the bounding box. It allows to obtain the farfield at any frequency in the specified range as well as at any time of the simulation. Frequency domain and time domain results can be accessed by selecting plots of a broadband farfield monitor and by the plot properties dialog. 5#?
HL
 
D/zp_9B  
Freq. samples: Choose a number of equidistant frequency samples between Fmin and Fmax where frequency domain farfields are calculated. Results for any frequencies in between are interpolated to obtain broadband information. For a more accurate interpolation over the frequency band increase this value e.g. to "31". An odd number assures that the result at mid frequency is not obtained by interpolation. =dC5q{  
ET]`  
Accuracy: Defines the desired accuracy of the farfield. Together with the Fmax and the size of the structure it determines the number of modes required to represent the farfield. nG5:H.)  
oX]c$<w5  
Note that leaving out higher order terms by choosing a lower accuracy is equivalent to low pass filtering the farfield solution. This saves memory and computation time. However, the farfield result has usually less detail. X15e~;&  
u|8V7*)3  
Transient farfields: This check button activates the additional calculation of transient farfield information which can be displayed afterwards for a certain time in the post processing. Note that in order to accurately obtain time domain farfields the computational effort is higher than for broadband farfields only. zQ eXN7$  

刚开始为了节省时间，设置的频点少。后来想增加频点，只能重新仿真了。谢谢各位。