Renewable Energy & Advanced Power Electronics Research Lab
The REAPER project on “Characterizing Photovoltaic System Arc-Faults,” funded by the grant UREP24-023-2-010 from the Qatar National Research Fund, addresses the need to understand the characteristics of an electrical arc in a photovoltaic system in order to build better arc detectors to make these systems safer.
Arc faults have always been a concern for electrical systems as they can cause fires, personnel shock hazard, and system failure. In photovoltaic (PV) electricity wiring systems a large number of electrical connectors and long wire runs are common to connect multiple PV modules into strings, strings connected through combiner boxes into arrays, and arrays connect through long home-run wiring into inverters for grid-connection. When installed properly, PV solar systems do not cause fires. However, due to improper installation, faulty wiring or insufficient insulation, fires have been attributed to solar panels. Accelerated aging and deterioration of the wire insulation due to the hot and humid Qatar environment can exacerbate other circumstances such as rodent bites, abrasion due to chaffing against building walls and roofs, or damage from being pulled through conduit during installation. These conditions lead to an increased probability that electric arcs can occur at some point in the lifetime of the PV system.
Arc faults in alternating current (AC) systems have been well studied and arc fault detectors are now commonly required by electrical codes such as The National Electric Code in the USA. Manufactures of AC arc fault detectors test their products using well-established safety standards, such as UL 1699, to ensure safe and reliable operation. Comparatively, a much smaller body of work pertaining to characterization and understanding arcs in DC electrical systems exists and commercialization of sensing and protection devices has only recently begun. A significant complication in a DC system is that there is no periodic zero-cross of the voltage and current, as which occurs in a 50Hz AC system. Thus, unlike an AC system, a DC system has no natural way for arc current to be interrupted. Complicating the detection of the arc in a DC system is the presence of power electronics, which generate electrical noise that can mask the electrical signal of an arc. The result is that an arc fault in a DC system can go undetected, and uncorrected, dissipating energy into the surroundings until ignition occurs and a fire results. Fire and smoke detectors only signal once the fire has already started. Arc detectors provide much earlier warning of the underlying root-cause problems that can lead to a fire – before ignition occurs.
Qatar has already suffered from the tragic loss of life and economic devastation from electrical problems that lead to fire. Perhaps most notably was the May 2012 Villagio shopping mall fire that left 13 children and 6 adults dead. In September 2013, a fire allegedly caused by the rooftop PV system destroyed a food warehouse in the USA.
The results of this project will improve the technology of arc detection to prevent future loss of life. This project will produce a database of electrical arc fault signatures that will later be used to develop and test arc fault detector technology and products. Since electrical arcs are not expected to occur frequently in a PV system, and there are numerous operating parameters that influence the occurrence and nature of an electrical arc, it is difficult to create repeatable arcs in an in-situ real-world environment. Hence, this project will design and develop a mechatronics testbed for precise and repeatable computer control of the arc generation process. It is hypothesized that the electrode geometry, electrode material, electrode gap distance and separation velocity, and electrical voltage and current all can change the characteristics of the direct current (DC) arc. The proposed mechatronics testbed will consist of a stepper motor driving a lead screw to control very precisely the gap between two electrodes. The testbed will allow interchanging electrodes so that different electrode geometries machined from different conductive materials can be studied. The computer controlling the testbed will also precisely adjust the driving voltage and available current for the arc, so that arcs at different power levels can be studied. Since arcs are chaotic, numerous experiments will be performed for each combination of the electrode geometry, electrode material, electrode gap profile, and arc voltage/current. The arc voltage and arc current will be measured by an oscilloscope, also under the precise control of the testbed computer. The acquired electrical signals will be analyzed using the Fast Fourier Transform to create spectral plots as the gap profile changes. The time-domain waveforms, along with the FFT spectral analysis will be curated for off-line use. The entire database will be made freely available to researchers and product developers.
The scope of work and level of technical depth are appropriate for undergraduate students as the various topics including Fourier analysis, MATLAB programming and stepper motor control are all part of undergraduate courses. The unique aspect of this proposed project is the application and how the students will integrate these concepts to synthesis a computer-controlled mechatronic platform to generate and record the data and characteristics of different electrical arcs. Computer-controlled test equipment is becoming commonplace in industrial settings. Thus, students will learn technical skills that translate into industry and can help them become better prepared to enter the workplace. Students will also learn the soft skills of working together on a team in a multi-disciplinary project. The project is expected to generate at least two scholarly papers on the test bed methodology and the data generated, which will allow the undergraduate students to learn how to write a technical research paper.