Carlos Giron internship
Company: National Institute of Wind Energy (NIWE), Solar Resources department (SRRA)
Location: Chennai, India
Intern: CAfukcnRLOSheeb ARMANpehflDO 8lnbGIRÓN ROSA (email@example.com+rsv/e+7m6m) (PPRE 2015/2017)
The NIWE is in charge of promoting and studying the wind energy development in India. In order to achieve this goal (within other activities), they own a turbine test site with an installed capacity of 4.4 MW, and also a wind forecast system. Now with the SRRA department they want to integrate wind energy with solar energy, for this they have more than 100 solar monitoring stations all over India and they also want to implement a solar forecast system. The internship was divided in two tasks, the first task was to define (according to literature given by them), which one are the the considerations that need to be done to select a solar forecast technology. The second task was longer than the first one (took 80% of the duration of the internship), was to do a prefeasibility study for a Virtual Power Plant (VPP) in India. A Virtual Power Plant, (also called Combined Power Plant) is a power plant compounded by small decentralized power plants of different renewable energy resources, which are all controlled by one remote control center, for supplying a variable load. This type of plant is aimed to substitute non-renewable plants.
The result of the first task was a 6 pages document where it is stated that considering that in India the solar power plants need to:
1. Have a forecasting of more than 70% accuracy (see Indian Grid Code Annexure 1, section 5),
2. Area to forecast: 1-5 km point, depending of area of power plant,
3. Time Horizon: 15 min intervals and day ahead forecast (see Indian Grid Code Annexure 1, section 4.i. ),
It was concluded that the selection of a forecasting technology type depends mainly in the accuracy needed for the application. And for this, should be contemplated the following considerations (Pelland et al. 2013):
1. Local climate and weather conditions
2. Single-site or regional forecast
3. Forecast horizon
The second task was divided also in two parts, the first one was to develop a software that can simulate the dispatch of the Virtual Power Plant, and the second part was to develop a
prefeasibility study for a Virtual Power Plant using the 4.4 MW wind farm (owned by NIWE), a 2 MW solar plant (under construction during this 2016 also for NIWE), a biomass and biogas plants (already on operation with a third party) and a storage plant (not yet constructed). This task was done with a 4 people team from different institutions: Engineering Research Division (National Research Centre, Egypt), Power Energy Dedicated Advanced Centre (University of Malaya, Malaysia), NIWE (India), and myself from Oldenburg University (Germany).
For the first part of this task, after researching in IEEE documents about VPP, we used the software Power Factory and Matlab to create the software showed on the Figure 1.
After doing this software, it was used like a tool for testing several scenarios (more than 108) to choose the best suitable scenarios (those that import zero energy making the best of storage) changing the values of storage and bioenergy plants. This results were added to the prefeasibility study, in this study we also did an assessment of all the technologies that were planned to use in the VPP. For the solar assessment, the PVsyst software was used. And for the wind energy, it was hard to get the actual generation due to delays on getting the information, then an exercise with Matlab was done (with the power curve of the turbines) in order to get the generation of the wind plant. After selecting 4 scenarios (see Figure 2) with the best installed capacity mix, an economical analysis had to be done. However, the time of the internship finished at the beginning of this phase, but the team continued with this phase. At this moment, Im still helping with this project and we did a paper that we want to present on the conference Solarpaces 2016 that it will take place at Abu Dhabi, UAE from 11-14 in October 2016.