Here's what I like to do. I wouldn't call it "planning" necessarily, but more "sketching". I take a piece of my main themes that I particularly enjoy or think are rich with possibilities, and I just start coming up with variations on them. This is sketching. I am not trying to construct a finished piece at this stage, but simply generating ideas, the way a comic book artist might draw different versions of a character in his sketchbook.
IMPORTANT: Do not try to create a masterpiece at this stage. You are just generating content, not creating a final draft. Don't over think it. If you have an idea you like, follow it to it's logical conclusion. Go where ever you like with the music. Development is free composing, so let your stream of consciousness take over. Try it in minor, try it in 6/8, try taking the main motif and inverting it, try the melody in the bass, try putting theme 1 in treble and theme 2 in the bass and fix up the harmony until it sounds musical and not just gimmicky. If you start an idea and run out of juice, set it aside and try another idea. Do not under any circumstances delete anything. No critics allowed at this stage. Just generate content and set it aside.
Do this multiple times until you have more ideas than you can possibly use. Then take a break. Later, once you've forgotten what all the stuff you wrote sounds like, go back and listen to it all or play it through. You may hear some ideas that jump out at you, that excite you. Find an idea you created that you like, and picture where in the grand scheme you might like to see it appear. Does it feel like a climax or a lull? Is it a good ending, a good beginning, wacky all-over-the-place material that just sounds cool, transitional material, a sequence? Consider your best ideas to be like puzzle pieces and imagine how they might fit together.
Then begin to craft them into actual art. Connect the ideas, develop the ideas so they aren't just vignettes but instead well-crafted parts of the larger story. Write spin-offs from the spin-offs. Try different ideas with different accompaniment. Try Alberti bass, then block chords, then sparse texture, then whatever you want to try. If you have created two ideas you wish to connect but they are in distant keys, transpose them (there is no shame in that). Take an idea you like and do one version in major and one in minor, or make the powerful final chord of that idea a 7th chord so that it wants to lead to a key change. Like creating a samurai sword, you must fold the steel over and over until it is just right. Don't worry about rules. You can begin and end in any key you wish. Even Beethoven wrote developments that ended up in different keys than he "should" have been in. He ended his developments when they sounded done, not when the key signature told him he was done. (see mvmnt. 1 of his sonata Op. 10 No. 2. The main theme is in the key of F, but when the development is over, he plays the main theme in the key of D. Why? Cuz he felt like it).
I am a big fan of this sketching process, but it takes patience. If you sit down and try to plan your piece from beginning to end from scratch, you may feel overwhelmed by the task. But sketching will give you more material than you can use, and then you can edit yourself rather than sitting there wishing you had more musical material to work with. Plus you may end up with material you can spin into a different piece, or a different movement of the same sonata. Do that and, voila, your sonata will be united by similar motifs across all the movements. Good luck and keep writing music!