What statistics?  Are you saying your PLC does something you do not like while it is shutdown (and not running and solving PLC logic)?

You might consider using a 24VDC UPS, with a configurable digital output wired to a PLC dig input, where in the PLC logic, if 24V power going OFF condition is detected, appropriate PLC logic can save data (or statistics) to a RETAINED DB, and then restored in OB100 (startup OB) when PLC powers back up.

U_310d9f61-4866-4978-8972-8edd0f5582d4

When we switch off the power of cabinet statistics moves to previous day (not after date). Is this any solution for this 

U_310d9f61-4866-4978-8972-8edd0f5582d4

When we switch off the power of cabinet statistics moves to previous day (not after date). Is this any solution for this 

The solution depends on what you have done. It sounds as if some rework of the logics is needed. Or perhaps some retaining DB should be used.

Our volunteers are blind to what you have. The focus of our volunteers shifts into your favour if they have something to work with. A sample project as ZIP attachment would be very helpful.

^{NOTE:
  Sample code is STL coded in Step7 V16
  GSD imbedded
  Code does not compile
  No indication of which code is being evaluated}

You are not doing yourself any favours by not telling us which part of the program is your responsibility. Should I answer in spite that your problem is line 5 of the STL code? Or perhaps 42?

The "dbSt1ProdCounters".Intsantaneous_counterActual.Cnt.Station1 tag should be retentive, IMHO. Consider creating another DB where you have the option of making this count retentive.

Sorry for my delay and too general an approach to the issue. I search solution for this: Turning off the cabinet transfers the statistics to the previous day (not after the date). What could help? 

Forgive me as I do not generally help on STL based requests.  However, you did ask "What could help?".  To that, I suggest following the Siemens S7-1200/1500 Style and programming guide and avoiding STL, and in your case, switching to SCL.

If you posted the same question in SCL, I think 30-300 other regular forum users could almost instantly tell you what is wrong with the logic.  I dare say you would not even have the problem if it was written in SCL because it would be relatively obvious what was at fault for the situation you are seeing.

When you say "turning off the cabinet", what time of day does this happen?  Is the PLC powered off when the cabinet is turned off - or does it simply turn off high voltage and leave non-controlling 24V voltage on?

If the cabinet (and PLC) are turned off before midnight, the prior day is all it knows.

If the cabinet (and PLC) are powered off after midnight and the daily statistics have not yet been calculated (presumably done once a day AT midnight rollovers), then there are no present day statistics yet calculated.

However, because I do not stare at STL long enough to understand it, I can simply guessing.

U_310d9f61-4866-4978-8972-8edd0f5582d4

Sorry for my delay and too general an approach to the issue. I search solution for this: Turning off the cabinet transfers the statistics to the previous day (not after the date). What could help?