This WG’s Chair is Jonathan Leloux (UPM) and its Vice Chair is dr. Marios Theristis (name organization).
Email for contact: email@example.com
The WG5 aims to contribute towards PV systems that are better integrated into the grid, that perform better, and whose operation under real-world conditions is better understood. This is done through the exploration of different complementary pathways:
- Identification of required data and appropriate simulation models
- Accurate forecasting of solar power and cloud formation to make it possible for utilities to be able to better balance their distribution grids
- Analysis of the interface between PV systems and low voltage distributed grids. Feeding of distributed solar power into grids under variable irradiance conditions.
- Matching electricity demand with supply in (smart) energy homes. Smart solar charging of electric vehicles and other storage solutions. Self-consumption and domestic uses of PV such as smart appliances.
- Development of data analysis and simulation tools
- Short and long term forecasting of irradiance and PV power generation.
- Better energy management and storage control at the interface between PV systems and low voltage distributed grids. Smart solar charging of electric vehicles. Domestic uses of PV such as smart appliances. Study of the relationship between the PV production and the local consumption, the possible use of batteries, the economic viability of alternative options,…
- Power Quality indicators at the connection of PV systems to the grid.
- Assessment of PV power fluctuations at one PV system, as well as for PV system fleets. Study of the correlation between the fluctuations in PV power between neighboring installations. Several complementary approaches are considered, including parametric and non-parametric models, peer to peer (P2P) approaches, Artificial Neural Networks (ANNs), the use of geostatistics,…
- Assessment of the PV power mitigation potential from the geographic dispersion of PV systems.
- Peer-to-peer prosumer cooperation, e.g. using blockchain technology and spatiotemporal forecasting using data from distributed PV systems.
- Better procedures for cheap and effective Operation and Maintenance (O&M) of PV systems, including soiling and cleaning strategies.
- Fault detection algorithms to improve the energy yield of grid-connected PV systems and reduce their power instability. Several complementary approaches are considered, including parametric and non-parametric models, peer to peer (P2P) approaches, Artificial Neural Networks (ANN), stochastic modeling,…
LIST OF PARTICIPANTS
|Last name||First name||Affiliation||Country|
|Albu||Mihaela||Politechnic University of Bucharest||Romania|
|Almonacid||Florencia||University of Jaen||Spain|
|Brito||Miguel||University of Lisbon||Portugal|
|Caruana||Cedric||University of Malta||Malta|
|Fernandez||Eduardo||University of Jaen||Spain|
|Fialho||Luis||University of Evora||Portugal|
|Georgieva||Penka||Burgas Free University||Bulgaria|
|Gercek||Cihan||University of Twente (UT)||Netherlands|
|Gomez||Emilio||University of Castilla-La-Mancha||Spain|
|Knoebl||Karl||University of Applied Sciences Wien||Austria|
|Leloux||Jonathan||Polytechnic University of Madrid (UPM)||Spain|
|Licari||John||University of Malta||Malta|
|Livera||Andreas||University of Cyprus||Cyprus|
|Marinkovic||Zlatica||University of Nis||Serbia|
|Micheli||Leonardo||University of Jaen||Spain|
|Mikalauskiene||Renata||College of Arts, Science and Technology||Malta|
|Munoz||Emilio||University of Jaen||Spain|
|Senabre||Carolina||Miguel Hernandez University of Elche||Spain|
|Stoilkov||Vlatko||Ss. Cyril and Methodius||FYROM|
|Theocharides||Spyros||University of Cyprus||Cyprus|
|Theristis||Marios||University of Cyprus||Cyprus|
|Todorovic||Jovan||Elektroprenos||Bosnia and Herzegovina|
|Valero||Sergio||Miguel Hernandez University of Elche||Spain|
WG5 is organized along several Tasks that are proposed by the participants. Typically, a Task starts when several participants are interested in a collaboration on a specific topic that can lead to the joint publication of an article in a scientific journal, and it ends once the paper is published. Each line is coordinated by one of the participants who are willing to take upon this responsibility and who will usually be the first author of the paper.
Several lines of work have been proposed. Some of them are currently ongoing, and other potential lines have been discussed and they will start if we can identify a team and a leader ready to attack it.
More details on each Task appear below in the text.
- Power quality at the connection of PV to the LV and MV grids
- Grazia Barchi, Sonia Pinto, John Licari, Cedric Caruana, Jovan Todorović, Karl Knoebl
- Mapping annual and seasonal soiling in Western Europe
- Leonardo Micheli, Joao Gabriel Bessa, Jonathan Leloux, Florencia Almonacid, Eduardo F. Fernandez
- Fault detection for PV system fleets using machine learning
- Andreas Livera, Marios Theristis, Jonathan Leloux
Potential future tasks
- PV in the context of smart-grids and self-consumption; Cihan Gercek, Luis Fialho, Brian Azzopardi, Karl Knoebl
- Spatio-temporal forecasting and PV monitoring; Rodrigo Amaro e Silva, Miguel Centeno Brito, Jonathan Leloux
- PV degradation vs stress factors (temp, RH%, UV, etc); Sascha Lindig, David Moser, Jonathan Leloux
- Monitoring data filtering and Quality Control (QC); Jonathan Leloux, Penka Georgieva, Zlatica Marinkovic, Sascha Lindig
Willing to join a task or create a new one? Ready to meet with the required high level of compromise? Send an email with your proposal to the WG5 leaders (see contacts below).
TASK PearlPQ: Power quality at the connection of PV to the LV and MV grids
Power quality at the connection of PV to the LV and MV grids; Grazia Barchi, Sonia Pinto, John Licari, Cedric Caruana, Jovan Todorović, Karl Knobl.
The high diffusion of photovoltaics plants in medium and low voltage distribution grids might cause different issues in the integration such as reverse load flows, protection settings and power quality. In the literature several works analyzed the impact of PV in grid PQ mainly in terms of voltage distortions and harmonic contents and how these aspects can affect the proper system operation but also power losses and in certain cases economic revenue for utility or final-user. Since a significative number of power quality disturbances should be introduced into the grid by PV inverter, the focus of this team is to is to carry out an assessment of power quality indicators for various PV inverters topologies (micro-inverters, string and central inverters) and to understand their impact on the utility grid vis-à-vis the EN50160 standard. Some of the inverters’ assessment will be carried out in the laboratory (when the inverter is available) where different power operating conditions simulating various irradiance levels can be tested. Through this exercise, the levels of the current harmonics will be analyzed particularly at partial loading. It is expected that the Total Harmonic Distortion will go up as the power generated reduces. In cases were the inverters are not available in the laboratory on-site measurements will be carried out.
On-site measurements have already been collected for a single inverter and three-phase inverter and a comparative analysis to a laboratory setup is currently being done.
In addition, a comparative power quality analysis between different inverter brands can also be conducted but this is dependent on the availability of inverters.
During this work, some short scientific missions are being considered by several members of the team.
The aim of this group is to publish the final results of this work in the near future. Simultaneously, we are preparing a special issue for a journal, focusing on power quality impact of PV systems in the grid.
TASK PearlSoil: Mapping annual and seasonal soiling in Western Europe
Mapping annual and seasonal soiling in Western Europe; Leonardo Micheli, Joao Gabriel Bessa, Jonathan Leloux, Florencia Almonacid, Eduardo F. Fernandez
As part of the effort on soiling, the team at the University of Jaen is working to extract the soiling losses from the sites available on the PearlPV dataset, using standard and novel soiling extraction techniques. The work of the team is currently focused on building a model to automatically calculate the performance index and the soiling ratio of the systems, by integrating the PV power data and various climatic data available on public repositories.
Also, a prototype of DUSST, a low-maintenance and low-cost soiling detector conceived and developed by the University of Jaen in collaboration with the National Renewable Energy Laboratory (USA), is being installed on one of the roofs of the university. The comparison of the DUSST measurements with the data collected by an Atonometrics soiling station and the performance of PV modules of various technologies installed on the same roof will make it possible to validate the prototype in a high-insolation region such as Southern Spain.
TASK PearlFault: Fault detection for PV system fleets using machine learning
Fault detection for PV system fleets using machine learning; Andreas Livera, Marios Theristis, Jonathan Leloux
Application of machine learning on a comparison of the PV data from a benchmark of neighbouring PV systems
POTENTIAL FUTURE TASK: PV in the context of smart-grids and self-consumption
PV in the context of smart-grids and self-consumption; Cihan Gercek, Luis Fialho, Brian Azzopardi, Karl Knobl
Exact scope still to be defined.
POTENTIAL FUTURE TASK: Spatio-temporal forecasting and PV monitoring
Spatio-temporal forecasting and PV monitoring; Rodrigo Amaro e Silva, Miguel Centeno Brito, Jonathan Leloux
Integration of spatially distributed solar times series in forecasting and performance monitoring applications.
POTENTIAL FUTURE TASK: PV degradation vs stress factors (temp, RH%, UV, etc)
PV degradation vs stress factors (temp, RH%, UV, etc); Sascha Lindig, David Moser, Jonathan Leloux
Degradation assessment using PVlib. Switch in paradigm from assessing a limited amount of data of high quality (typically lab data) to assessing a very large amount of data of lower quality (monitoring data).
POTENTIAL FUTURE TASK: Monitoring data filtering and Quality Control (QC)
Monitoring data filtering and Quality Control (QC); Jonathan Leloux, Penka Georgieva, Zlatica Marinkovic, Sascha Lindig
Quality Control on metadata: azimuth, tilt and peak power of PV systems,…
Different approaches: physical modelling, cost function, pattern recognition, ANN.
Peer-reviewed scientific papers
- Correlating photovoltaic soiling losses to waveband and single-value transmittance measurements, Leonardo Micheli, Jose A. Caballero, Eduardo F. Fernandez, Greg P. Smestad, Gustavo Nofuentes, Tapas K. Mallick, Florencia Almonacid, Energy, Volume 180, 2019, Pages 376 386, ISSN 0360-5442, https://doi.org/10.1016/j.energy.2019.05.097. (http://www.sciencedirect.com/science/article/pii/S0360544219309740)
- Performance analysis of mechanistic and machine learning models for photovoltaic energy yield prediction, A. Livera, M. Theristis, G. Makrides, J. Sutterlueti, S. Ransome, and G. E. Georghiou, EU PVSEC, 2019.
- EU COST Action PEARL PV, Performance and Reliability of Photovoltaic Systems: Evaluations of Large Scale Monitoring Data, WG5: PV in grids, Marios Theristis, Carolina Senabre, Emilio Gómez, Sonia Pinto, Luis Fialho, Emilio Muñoz, Jonathan Leloux, PV Reliability Workshop, NREL/SNL/BNL, Albuquerque, 2018.
CONTACTS (Working Group 5 Leaders)
Chair of Working Group 5
Polytechnic University of Madrid, Spain
Vice chair of Working Group 5
University of Cyprus, Cyprus