Can I submit an article just for the sake of useful feedback?
Yes and no. Of course nothing stops you from sending an article just to get feedback, and this is indeed a suggestion that I sometimes see given to young researchers when they are unsure about their paper.
However, there are a few issues that, in my opinion, makes this approach not overly effective.
- You will get feedback, but you may not get useful feedback. Reviews fundamentally tend to assess the current state of a research project. It should not be like that, but many reviews focus on what's wrong with a paper, and not on how it could be fixed. Think of it as an evaluation more than a feedback procedure.
- Additionally, there is little incentive for the PC to be overly constructive. My impression has always been that the anonymity of peer review leads people to be more dismissive than what they would ever be if they talked to you personally.
- The turnaround time of good conferences is so long that you need to expect to wait a few months for your reviews. This is wasted time if you are already pretty sure that the paper won't be accepted.
- As you say yourself, an argument can be made that submitting a paper that you yourself are not sure is ready for publication to be an unethical abuse of the peer review system. I would consider this a fairly mild problem myself, but it is obvious that the peer review system would break down (even more?) if many people started to adopt your approach.
More promising may be to find a collaborator who is aware of community standards of wherever you want to publish to take on board for finishing and publishing this work. Such a person can not only help you writing a paper using the conventions of the field, (s)he will also be able to point you and help you with any additional work that may make your work stronger in the eyes of your targeted community. Most importantly, this person has then skin in the game and will strive to be constructive rather than just telling you that your work isn't going to cut it.
This of course leaves the question how to find such a person. I am sad to say that personal connections (of you, your supervisor, or other friends or fellow faculty) are by far the best approach, but failing that you can try to establish ties by attending workshops or working conferences, as proposed by lighthouse keeper in his answer.