**1. Introduction**

The operation and maintenance of a photovoltaic (PV) power plant is of extreme importance to guarantee its optimal performance. Effective maintenance involves at the least semi-automatic analysis and alerts. In this way, the maintenance operator is capable of making immediate decisions to solve safety problems and minimize power losses [1].

performance of a plant can be analyzed. Predictive techniques can then be used to analyze deviations in the behavior of supposedly viable components and forecast possible outage in

Degradation Monitoring of Photovoltaic Plants: Advanced GIS Applications

http://dx.doi.org/10.5772/intechopen.75650

95

This chapter is organized into four further sections. First, a systematic review of the most common PV faults is undertaken. Then, the fundamentals of GIS and how they can be applied to PV maintenance and fault supervision are presented. In section three, the application of a GIS tool to both a large PV plant (100 kWp) and a domestic installation (9 kWp) is fully described and results of the implementation of both examples are shown. Finally, the last sec-

Prior to going deeper into this topic, the term *"*failure*"* in the PV context must be properly defined. The bibliography defines a failure in a photovoltaic module as *an effect that degrades the module power output which is not reversed by normal operation or which, on the other hand, creates* 

As such, degradation of wiring or modules, PV cell defects and malfunctions, dust, and mismatches can be considered to be failures of PV modules [5] and purely esthetic problems are not.

This consists of the loss of adhesion between the glass, the encapsulation, the active layers, and the subsequent layers [6], which can cause loss of current (power) in the photovoltaic modules. Loss of adhesion may occur for various reasons. Large thin film modules and certain other types of modules sometimes contain an additional transparent conductive oxide (TCO) layer, which may lose adhesion with the adjacent glass layer [6]. If the loss of adhesion is due to contamination, perhaps from cleaning, or environmental factors, then delamination will often take place, followed by moisture entering and, in due course, corrosion. Delamination

In most cases delamination can be detected by visual inspection, with the degree of layer detachment being quantified by use of a reflectometer. Some delaminations, however, cannot be identified in this way and so methods such as pulsed active thermography or lock-in thermography can be used, while smaller delaminations can be detected with ultrasound scanners and X-ray tomography. The latter methods are slow [7] but provide a much higher resolution.

The multilayer composites that make up PV module backsheet films comprise three or more polymer layers. Outer layers provide resistance to weathering factors such as sunlight and humidity and are often made from fluoropolymers with polyvinyl fluoride (PVF), polyamide

tion includes the main conclusions and some future research directions.

*a safety issue.* Evidently, both of these effects can occur at the same time.

The following subsections describe briefly all potential PV module failures.

leads to reflection of light and a subsequent loss of power in the modules.

(PA), or polyethylene terephthalate (PET) [8] being popular choices.

**2.2. Loss of adhesion in backsheet films**

**2. Photovoltaic module defects and faults**

areas of the PV plant.

**2.1. Delamination**

Moreover, monitoring is not just a regular recording of data but involves a more detailed analysis in order to prevent possible malfunctions associated with power and, in the end, economic losses.

Thus, the development of automatic supervision tools to help maintainers to carry out an effective supervision of the power plant and, what is more important, to monitor the evolution of the modules behavior in an easy and feasible way is of great interest in the industry. Furthermore, the detection of operating failures in a timely fashion through the evaluation of panels over time means that automated monitoring is, in fact, absolutely necessary. Such monitoring will lead to the early replacement of poorly performing components, preventive maintenance policies, and better management of plants.

For device manufacturers, performance evaluations of their products can be used as a benchmark of their quality manufacturing processes. On the other hand, researchers and R&D divisions from companies can take advantage of such evaluations, using that information to identify future needs in the industry and to test real working conditions. For PV plant promoters and owners, realistic performance data is essential for investment decision-making.

Although several techniques for the detection of real and/or potential failures in the PV field can be found in the literature [2–4], few of them attempt to detect anomalies such as power loss, module failure, or health and safety dangers until after they occur and the effects are felt. These techniques, moreover, do not address preventative maintenance strategies or effective economical programs for the replacement of components. In addition, these systems do not integrate geo-references that may help to improve the application of preventative maintenance strategies.

New methodologies are needed to locate and analyze performance and malfunction of plant components on a global scale. Despite the great value of analysis carried out in the laboratory, it can often be of little help when applied to the real operation of maintenance plans. Test conditions in a laboratory may allow for a complete analysis of PV components, but owners can ill afford to close an entire installation or part thereof for equivalent testing in the field. Besides, laboratory test conditions are unlikely to fully typify working conditions in a real field. As such, laboratory test results and, more importantly, any conclusions drawn from them are likely to be decontextualized. On the other hand, carrying out systematic procedural techniques in the field under changing environmental and climatic conditions is in no way easy.

Integrated geographic information system (GIS) platforms will allow test-related information to be comprehensively organized and geo-referenced, providing significant benefits. One such benefit is the fact that the impact of a single defective component on the overall performance of a plant can be analyzed. Predictive techniques can then be used to analyze deviations in the behavior of supposedly viable components and forecast possible outage in areas of the PV plant.

This chapter is organized into four further sections. First, a systematic review of the most common PV faults is undertaken. Then, the fundamentals of GIS and how they can be applied to PV maintenance and fault supervision are presented. In section three, the application of a GIS tool to both a large PV plant (100 kWp) and a domestic installation (9 kWp) is fully described and results of the implementation of both examples are shown. Finally, the last section includes the main conclusions and some future research directions.
