You can check the log files in /var/log/* to see if there is anything unusual there (focusing on 'dmesg' and 'messages' log files).
Try to figure out if there is any pattern before, during, and after the server crash. For ie. crashing at same time in the day, crashing when cron is running something, crashing when same file is accessed.
I guess you can ask your service provider to replace/check your hardware saying that it crashes multiple times.
If it's a power supply you'll likely need to wait it out about a week or so, if it is bad the PSU will die. If it does not die after a week, look for other causes.
.._(_)_.. Black Lotus Communications - AS32421 (_)@(_) >> Carrier grade DDoS mitigation for service providers and enterprises
....(_)..... >> 480 Gbps active DDoS filtering capacity available in Los Angeles and Ashburn facilities
Guessing it is a hardware issue is not a necessary thing with linux and a tool like hotsanic. Use real logging utilities to actively monitor system performance, temperature, loads and you will know what is going on.
Guessing that it's a power supply issue is bad practice, when you could be monitoring the rail voltages directly off the PSU.