A European analysis staff has proposed a brand new methodology for the optimum design of hybrid wind-solar energy vegetation. Their technique is predicated on 1-minute decision knowledge, which the scientists say gives extra correct details about potential curtailment and clipping losses in comparison with 1-hour decision knowledge.
A European analysis staff investigated the impression of various time decision knowledge on the design of hybrid PV-wind energy vegetation (HyPPs) and located that low decision may end up in an overestimation of the online current worth (NPV) in long-term amenities. to three%.
“When doing feasibility research for HyPPs, knowledge at an hourly decision, or decrease, are often used, as a result of that is the decision that’s often obtainable in meteorological databases,” the group stated. “However, when optimizing the topology of HyPPs with capability constraints of the interconnection (POI), the techno-economic efficiency of those energy vegetation will be overestimated as a result of the discount and lack of -clip shall be diminished when utilizing hourly as an alternative of upper decision knowledge.”
The analysis is predicated on knowledge obtained from an operational HyPP in East Germany. This plant has an put in PV capability of 11.64 MW and a DC-AC ratio of 1.13. The wind generators have an put in capability of 24 MW, and whereas they’re put in 1.6 km from the PV panels, they don’t shade them. In 2020, the yr the researchers carried out their analysis, wind and PV belongings generated 57.58 GWh and 12.80 GWh, respectively.
Weather and technology knowledge from the German HyPP have been recorded at 5-second decision and down-sampled for one minute and one hour for comparative evaluation. As a part of the evaluation, totally different DC-AC ratios are assumed for the PV half, which successfully adjustments the actual capability of the plant.
“It is proven that for a HyPP consisting of 1 per unit (pu) of wind, PV, and POI capability, respectively, the discount loss is estimated to be only one.45% for five s and 1.09% for 1- time. decision dataset, translating right into a lack of 0.77% and 0.51% of the German day ahead market, respectively,” they discovered. “These outcomes present a distinction between the losses that hourly estimates and high-resolution knowledge.”
In addition, the researchers discovered that utilizing knowledge with a decision of 1 min seems to be ample to estimate the discount losses predicted by the 5-second dataset. “Data acquisition at 1-minute decision appears to be compromise between accuracy and measurement effort,” they defined.
Conducting a techno-economic estimate to search out the optimum dimension of the HyPP, the researchers discovered that the usage of the 1-hour dataset results in a 1.86% overestimation of the entire NPV in comparison with within the 1-minute dataset. They additionally say that if the rise in PV shares and wind capacities is bigger than the POI capability, this overestimate shall be even better.
“The techno-economic mannequin reveals that the cost-optimal HyPP topology for this website consists of wind and PV capacities which can be every barely bigger than, however carefully aligned with, the POI capability. utilizing 1-hour as an alternative of 1-minute decision knowledge seems to have little impression on figuring out the optimum HyPP design,” they stated. “However, for the optimum HyPP topology obtained from the 1 h dataset, the NPV is overestimated by 2.99% when calculated for the 1-hour as an alternative of the 1-minute dataset.”
Their findings are introduced in “Influence of high-resolution knowledge on the correct estimation of attenuation loss and optimum design of hybrid PV-wind energy vegetation,” printed in Applied Energy. The group contains teachers from Norway’s Institute for Energy Technology (IFE), the University of Oslo (UiO), and Germany’s Fraunhofer Institute for Solar Energy Systems (ISE).
This content material is protected by copyright and might not be reused. If you need to cooperate with us and need to reuse a few of our content material, please contact: [email protected].