1-The choice between these two types of solar inverters depends on a number of factors such as total cost and energy production. For centralized architectures, they are most common for commercial and utility projects because of their efficiency and interactive control requirements.

On the other hand, for small-scale commercial projects, distributed architectures are generally preferred because of the ability of string inverters to optimize energy production in photovoltaic systems.

2-Distributed architectures that use multiple three-phase string inverters in a grid are the typical architecture and are increasingly common in the commercial market.

Kind Regard

The matter of one large inverter vs small inverters on each panel (string inverters vs micro-inverters) have been researched well in the past. A summary is available here:

https://www.solarreviews.com/blog/pros-and-cons-of-string-inverter-vs-microinverter

Also, the Wikipedia article has a great summary:

https://en.wikipedia.org/wiki/Solar_micro-inverter

It really depends upon what you need and for which application you are using it. If cost and size is not an issue. Moreover, the reliability of the application is very important than DMPPT. For large PV applications, it is very expensive to use DMPPT. Additionally, partial shading and other similar non-ideal conditions are not very common in large PV plants therefore, central MPPT. DMPPT should be preferred for uneven PV applications.

For the first question: there are several factors to be taken into consideration when selecting photovoltaic system architecture:

1-Efficiency: distributed power processing (i.e., power electronics at module, submodule, and cell levels) mitigate partial shading conditions and are normally adopted in PV systems with high partial shading (e.g., building integrated photovoltaic BIPV). The finer the power processing level is, the lower power loss is expected. However, with higher complexity, cost, and control overhead. My conference paper includes more details ( M. K. Al-Smadi and Y. Mahmoud, "Comparative Analysis of Partial Shading Power Losses in Photovoltaic Topologies," 2019 International Conference on Clean Electrical Power (ICCEP), Otranto, Italy, 2019, pp. 699-704, doi: 10.1109/ICCEP.2019.8890218).

2-Reliability: series architectures achieve the desired voltage, however, they suffer from single point of failure. In this sense, parallel architectures are more reliable (and modular), nevertheless, the power electronic stage must be designed to step up the PV voltage up to the grid level which necessitates, sometimes, adopting isolated topologies.

3-Cost: centralized architectures require only one inverter for the entire PV array, thus, reducing the overall system cost. The cost of PV system with distributed power processing is higher as clarified in the first point.

For the second question: both centralized and distributed architecture are used (Commercialized solution are available), taking into account the factors mentioned in the fist question.

Hello everyone. Thank you very much for your comments. Your answers are very enriching and interesting. I am currently working on monotoring conventional PV systems to detect faults (such as hot spots)and automatically provide protection. I have already read that DMPPT configurations do not have Hot Spot problem since the PV panels (shaded cells) do not operate in their second I-V quadrant. I wanted to make sure how competitive my product is.

Hello Khedidja,

Even with DMPPT configurations, hot spots may occur as follows:

Assuming that each sub-panel (i.e., submodule) is interfaced with a DC-DC converter, if the cells within one submodule are partially shaded (not all the cells receive the same irradiance and/or work on the same temperature), then the shaded cells behave as load for the non shaded cells and consume power which eventually lead to hot spots.

The ultimate solution that decouples the partial shading effects is to use a small, very high frequency power electronic converter for each PV cell. In this sense, each cell produces its maximum power according to its irradiance and temperature. However, this solution is complex and expensive.

I suggest this paper for more details:

""Mitigation of Hot-Spots in Photovoltaic Systems Using Distributed Power Electronics ""

Dear Mohammad K.Al-Smadi,

Thank you for the follow-up. I have already read this article. In fact, HS phenomena is not only due to the presence of partial shading, but also to the operating point. If we keep the operating point at the maximum available power point which corresponds to the maximum current offered by the weakest shaded cell, the shaded cells never become loads and no hot spot are formed.