Scientists have used a CPLEX-based MIP model and tested it on part of the 10 MW Masdar City Solar Fotovoltaic plant. In their simulation, they assume the use of two robot -like cleaning agents to work on a horizon of 90 days. The total cleaning costs were estimated at $ 7,987.
A research team led by scientists from the University of Sharjah of the United Arab Emirates (VAE) has developed a new model to optimize the cleaning schedule of PV panels in power plants in deserts. They have demonstrated the technology using data from a 10 MW solar energy plant in Masdar City, located in the Emirate of Abu Dhabi.
“Our approach is intended to minimize both cleaning costs and financial losses due to power reduction as a result of pollution. The proposed method offers valuable insights for decision makers in the VAE, so that they can make informed decisions that are tailored to their specific priorities and limitations,” the scientists said. “Ultimately, this research wants to improve the performance of solar energy plants in the VAE and support the progress of sustainable and efficient energy systems.”
The new technology uses a commercial optimization solver (CPLEX)-based Mixed-Integer Programming (MIP) model. The CPLEX-based MIP model is fed with environmental data, including radiation, temperature and pollution speed; Operational data, such as parameters of the cleaning system; and economic factors, including cost data. The model comparisons are then used to calculate the energy output of the system, the loss speed and the cleaning restrictions. All this is done in view of the objective function of balancing energy loss costs against cleaning costs.
Image: Sustainable Futures, University of Sharjah, CC by 4.0
To demonstrate the model, the team took the full operational data of the Fotovoltaic factory of Masdar City from 2022. According to the collected points, the 10 MW park per year generates 17,564 MWH, from 87,780 PV modules. The facility area is 210,000 square meters, with 4,389 rows, each with 20-100 panels. The pollution rate is considered 0.02% to 0.9% per day and the energy rate is $ 0.1/kWh.
Their simulation was performed on a part of the park, consisting of 40 rows of 100 panels each. They assume that the use of two robot -like cleaning agents works over a horizon of 90 days, with cleaning intervals ranging from 3 to 20 days. The cleaning price was simulated at $ 0.012 per square meter or $ 0.05 per square meter. The optimization model was then used to calculate when and which rows of panels must be cleaned during the three months.
According to the results, the cleaning process is optimized by cleaning three rows a day. The number of cleaned solar panels per row during the 90 days varied from five to nine, with most rows being cleaned six or seven times. The cleaning costs amounted to $ 4,833 when taking a rate of $ 0.012 per square meter and amounted to $ 7,987 when the cleaning costs were $ 0.05 per square meter.
“By bridging quantitative optimization with practical considerations, the model offers a valuable decision -making support instrument for industry and policy makers,” the group concluded. “Restrictions include an optimality gap of 21.65%, dependence on its tropical pollution data without real-time variability and absence of empirical validation via tools such as PVsyst or SCADA systems.”
Their work appeared in “Optimization of the cleaning plan of a solar PV system factory“Published in Sustainable future. Scientists from the University of Sharjah in the VAE, the University of Jean-MonTet in France and the University of Tunis El Manar in Tunisia participated in the study.
This content is protected by copyright and may not be reused. If you want to work with us and reuse part of our content, please contact: editors@pv-magazine.com.
Popular content
