Thank you, I thought having a more stable input would contribute to a more stable output.
Hi MAntunes: interesting PSU project.
What is the secondary current rating of your transformer?
As the other members say, an LM317 stage is not required. In fact it would limit your output voltage range considerably and would add heat and unnecessary complication. I like your nice neat schematic though. 
Afraid you will not be able to get an output of 32V with a 24V AC transformer. If you let us know the secondary current rating of your transformer and the maximum PSU output current that you want, I will work out what your maximum output voltage can be.
Can we take it that the maximum current you want is 1A?
There are standard techniques to reduce switching noise. I would not think that an active filter would be necessary or desirable.
The 1N400x type rectifier diodes in your schematic are not the best for this application. I can recommend a better type if you like.
I'm not sure about the current rating. I'll check the transformer later.
If the LM317 is not really needed the better for me: less components = less money.
I don't really need the the full 32V of the DPS3003 module, I'm happy with 30V. I'm not even sure I'll need that much.
Yes, 30VDC@1A is what I'm aiming for.
Could you tell me these techniques to reduce switching noise? What you be better? LC filter, Pi network? What should be the cut-off frequency?
About the diodes, I'm always ready to learn so yes, I'd like it if you could recommend better components. I know I have a diode bridge at home somewhere, but I can't find it at the moment
I know, I should have explained myself better. I pretend to build a AC-DC converter to power a DPS3003 module (
https://goo.gl/hii2nP) which is a buck converter. I have a 230VAC RMS to 24VAC RMS (~34V peak) transformer so in the output, taking into account the voltage drop of the diodes, I will have around 32V.
I also want to design an output filter to remove the high frequency components of the switching converter.
Thank you all.