做氟谱的时候在峰的两边会有一些倒立的峰，怎样避免？

First of all, I don't know if your sample is solid or liquid.
The first possible reason is the T1 relaxation times among those peaks are different and the delay time (the time between each pulse sequence) you set is not long enough, try a longer delay time (in Varian machine, it's the "d1" parameter).
The second possible reason is that when the pulse goes into different regions in a solid sample, the phase of the pulse probably changed a bit, leading to different phases of the signals.
Another thing you can try is to type "lp=0 rp=0" and then "wft" to see if it works.