It is an emergency power supply. The battery, grid charger and inverter are the place to put money in this type system, especially if the person is wanting to upgrade or have the option.
Op says balcony mount, totally doable and that they have good exposure. ( i left this on my clipboard and awoke to stunningly spaceous balcony photos!)
if you want to run a freezer and the other critical loads indicated the freezer is priority and system power and capacity are largely going to be enough to allow opportunity charging and use of all kinds of devices and appliances.
Lights and comms/computing are really a byproduct of having a system capable of electric water delivery and refrigeration. Even though water delivery is not mentioned, the freezer represents all kinds of spare capacity in sizing of power equipment.
the proposed system is small, a micro-system or cabin system. The desire to deep freeze should utilize extreme efficiency there then supply a battery that can do it plus the other loads at 80%dod. And add 20% for all kinds of reasons unless it only has to do it a few times or will not be expected to still work several years from now even if it lives at float.
The battery is the core module in the system. If you can deploy any pv watts, you will at least offset consumption and offer opportunity use. The pv is an adjunct system to a standby ups.
Of course the system could perform load defection from utility supply and increase value or cost avoidance as opposed to just a backup which is a luxury, a household appliance. $$
Daily system capacity needs to be defined and specified in duration of autonomy in order to spend good money on storage and power-electronics.
Inverters are cheap if need be. Have an inverter/charger if possible. The cost will be close an inverter and transfer switch combo but you get a grid charger and ups function out of it. Plug and plays will be clunky.
Im not sure which ones people are liking but the mpp amd sigineer stuff are low cost, good quality. Mpp even has high functionallity and stunning monitor interface both onboard and computer terminal/phone app.
I would shop low standby as much as i would pure sine wave, and i actually use msw, but that is for low budgets like mine.
if you set a goal to be able to replace the power used by the fridge under 5% solar conditions your system will rarely not charge during daylight. There are days where the control may as well not turn on. That eventuality and overnight is wher the battery comes in. If you have an ample array and the system is for emergency 2.5 or more days autonomy is a decent number with a budget.
i dont know your insolation data, nor do i know the likely duration of outage, nor if you intend it to supply beyond that.
if you get the core of the system in, adding ample comtroller and pv modules is trivial over time and at 149$ apiece for 320w modules cheap to puff up the budget as you can.
think of it as a bank account.
my best system is one where the client does not beat around the bush on budget. Just state how much you are willing to spend iand ill maximize the dollar or say directly it cant be done.
i your budget were set for now at 1000cdn id invest in enough battery and inverter charger directly and add solar later. Or start with a single panel, a real residential/commercial power one.
yu will save money and actually be able to use the nameplate in the design.
i have to step in when someones dreams and aspirations in solar are getting beaten down in cautionary blather.
Too many projects succumb to the negative hearsay on the matter as opposed to direct investigation of what can and really should be done with 1000$
