**1. Introduction**

The trend of the last decade to reduce greenhouse gas emissions and decarbonize the energy sources is mainly due to the increase of the temperature of the planet, has driven the development and improvement of technologies for renewable energy sources. As well known, solar energy is the most available renewable energy source in the world with respect to the other sources and an interesting topic of research is the evaluation of its economic convenience. For this reason, in many studies reported in literature, the economic convenience of grid-connected PV systems has been evaluated with respect to other technologies. The traditional approach is based on the use of the "levelized cost" method that represents the per unit value of total costs (i.e., capital, operation and maintenance, fuel) over the economic life of the power plant [1,2]. The notion of economic convenience of PV plants emerged from these studies.

In the European Union (EU), the number of the PV plants is increasing significantly thanks to the policies of the different countries that have led to economic benefits for private citizens and in particular for small size PV plants called "residential plants" (from 1 kWp to 10 kWp) through diverse incentive systems. The international policies to encourage the installation of the PV plants usually consist of an incentive for the total "green energy" produced and in an incentive for the energy injected into the grid only for the grid-connected plants.

As well known, a typical issue of PV plants is the power loss due to the differences of irradiation (partial shading or wrong design) among the cells of the same module or among different modules of an array. This phenomenon, known as mismatch, can generate a considerable power loss of the total system with a consequent economic loss. In [3] an interesting estimation of PV mismatch losses caused by moving clouds is reported. Moreover, this phenomenon can be caused from a fixed obstacle that may have appeared years after the installation of the PV plant. The presence of a fixed obstacle after the installation is a frequent case in residential PV plants.

In order to limit the mismatch phenomenon, monitoring systems are extensively used in renewable energy applications to track the performance of the generation plant. A monitoring system for PV arrays is usually needed to collect power production and performance data as well as weather condition information and relate them [4–6]. These systems allow detecting fault conditions, but they are not effective for estimating the power reduction of a PV plant.

The issue of the different irradiation levels among the cells of a module has been studied in [7], where an investigation on partially shaded modules with different PV cell connections was reported. The authors compared five different connection configurations in order to find the best solution to increase the maximum power production and the fill factor of a module. The same problem has been studied in [8] and [9].

The different irradiance causes also problems in maximum power point tracking (MPPT) algorithms because the P-V curves of the PV module exhibit multiple maximum power points due to the bypass diodes, which are used to exclude the module of an array. In [10], the authors classified MPPT techniques for different PV array configurations. Obviously, each method presents advantages and drawbacks. Again, an interesting MPPT strategy for PV arrays under uniform and non-uniform irradiance condition is described in [11].

A recent solution proposed in the literature to reduce the power losses is the use of a dynamic reconfiguration system (DRS). The DRS allows one to change the configuration of the PV plant in order to increase the power production. Different solutions have been recently proposed in literature to optimize the power output adopting dynamic reconfiguration systems for PV module interconnection [12–19].

An interesting topic about DRS concerns the economic benefits introduced by the use of these systems. In [20] a technical-economical evaluation on the use of a DRS in some EU countries for PV plants is reported. In particular, by considering the incentive policies and others technical aspects of a 3 kW PV plant the NPV have been evaluated for each country taken into account. Nevertheless, in this study the technical aspects and different configurations of the DRS were not considered. For this reason, it is necessary to extend the economic analysis by introducing the real technical considerations of the DRS for different configurations reported in literature.

The aim of this paper is to evaluate the economic benefits accrued by using a DRS in a residential PV plant. First, an economic analysis of different DRSs according to the costs of the components and to the adopted topological schemes, is carried out; to the authors' knowledge, this issue has never been addressed in the technical literature. Architectures involve switching matrix, sensing network and driving circuit, the choice of switches affects the electro-technical and, electrical endurance. As will be shown in the following sections, the choice of a more flexible DRS comprises higher initial cost due to the number of switches required by the adopted architecture, but at the same time a less exploitation and therefore a longer useful life.

In particular, the study takes into account different technical and economic aspects of a PV plant in order to present a complete economic analysis. In other words, the study is focused on the use of different DRS configurations reported in literature, in some EU countries in order to evaluate the performances of the investment. The economic tools are the net present value (NPV) and payback time.

This paper is organized as follows: Section 2 provides a brief description of the DRS topology taken into account in this work and technical considerations to estimate the costs and lifetime of DRS. Section 3 describes the experimental set-up to perform the evaluation of performance of DRSs. In Section 4 the economic data are reported and in Section 5 the economic results are presented. Finally, Section 6 concludes the paper.

## **2. Dynamic Reconfiguration Systems**

A DRS allows changing the connections among PV modules in order to increase the total power production from a PV plant under poor irradiance conditions or other situations, that determine the degradation of the performance. In this way, the hardware complexity of the DRS depends on the possible connections among the modules. Generally, in a reconfiguration algorithm each panel is a considered a node of the dynamic array; the number of the nodes is identified as *m* while *n* switches perform the dynamic connections among the panels. A plant with a high number of panels requires a DRS with a high number of switches in order to connect all the nodes. Thus, a topology with more switches guarantees a high number of possible configurations for connection of the panels. In this section, a brief state of art of the dynamic reconfiguration systems (DRS) and a technical-economic analysis of the four cases studied, are reported.
