Abstract
Worldwide demographic growth and the resulting increased demand of energy and a decline of natural resources prompts more and more countries to extend the use of renewable energy technologies. Renewable energy means energy converted by energy from water, biomass or solar and wind energy. One of the most promising renewable energy means is photovoltaic technology. Recent years have seen an increased demand for photovoltaic systems, because of their relatively low and constantly decreasing electricity production costs compared to other renewable energy technologies. Thus, photovoltaic technology allows for cost-effective investment in a renewable power-generating system with low environmental impact. On the other hand, a collective disadvantage of renewable energy technologies are the low energy densities of the energy sources. This means that an economically feasible use can only be achieved via large installation areas that are associated with high investment costs resulting from an extensive use of materials. In addition, the volatile character of solar irradiation impedes the operation of photovoltaic systems at an optimal economic level. This means that power plants converting energy of renewable energy sources need to be used for as long and as efficiently as possible. Low profit margins particularly require optimized operation and plant management of photovoltaic systems. An optimum operation of photovoltaic power generators can only by guaranteed by long-term performance and continuous system optimization. In order to achieve this, a loss of performance caused by abrupt failures or gradual degradations must be prevented by identifying the causes of failure in a quick and reliable manner. The identification of such failures requires an integrated concept for plant monitoring and failure detection, which shall be realised through this thesis as part of the OptPV4.0 research project. The aim of this thesis is to implement a physical photovoltaic system model in the simulation environment Simulink and use it as a type of “digital sensor” in a plant monitoring system. Furthermore, a mathematical equation shall be derived from the physical model to implement a data-driven time series analysis of photovoltaic power plants that serves the identification of failures occurring within the system. The correct identification of critical system components and the extent of the load is ensured by using white box models based on Physics of Failure. These models show the state of the system and the failure probability of components. Overall, the integrated approach shall contribute to an efficient and long-term operation of photovoltaic power plants with a maximum energy yield and shall be applied within the OptPV4.0 research project for monitoring photovoltaic plants.
Translated title of the contribution | Physical system modelling of a photovoltaic plant for condition monitoring of plant performance and failure detection in condition-based maintenance |
---|---|
Original language | German |
Qualification | Dipl.-Ing. |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 29 Sept 2020 |
Publication status | Published - 2020 |
Bibliographical note
embargoed until nullKeywords
- PV
- Photovoltaic
- PV modelling
- PV module modelling
- PV plant modelling
- PV inverter modelling
- one diode model
- Simulink modelling
- PV monitoring
- reference value model
- model-based state detection
- digital sensors
- yield prediction
- physics of failure
- failure diagnosis
- condition -
- predictive- and reliability-based maintenance
- time series analysis
- OptPV4.0 research project