Learning from industry: David Osuna Ruiz reports on his placement with Shouhang European S.L
Third year PGR David Osuna Ruiz was recently awarded funding from EPSRC for industry placements in national or international companies whose activities are aligned with the UK National Industry Strategy. David’s field was Renewable Energies.
His two-week placement, made possible by the EPSRC’s funding award, was with Shouhang European S.L., which develops energy resources saving technologies such as air cooling systems, water treatment technologies, waste heat utilization, heating supply and Concentrated Solar Power (CSP) Stations. In China, the company has been awarded for many pioneering achievements on these topics and was the first company in that country to produce the largest amount of kilowatts from its projects based on those technologies.
David was assigned the task of developing an interactive GUI (Graphic User Interface) through Matlab to let the user perform the following actions:
• After introducing the technical parameters and physical dimensions of the station, show a simulated sunspot on the target of the solar tower (where rays are collected) with a frontal view (a piece of code with the mathematics of the ray optics was provided)
• Import a real image of that sunspot provided by an IR camera
• Draw an isometric schematic of the system (tower+plane of incidence+sun rays) to give an idea to the user of how the orientation changes as parameters are modified in real time
• A difference image to spot differences more easily. The purpose of this program is to infer from simulations a parameter called the ‘slope sigma error’, an error related to the aiming of the heliostats (mirrors) to the solar tower. The real sunspot must have the same ‘sigma error’ that the one in simulations when both sunspots look similar, which helps to modify the aiming settings of the heliostat in real time, compensate the error and get a more efficient aiming to the collection target. This can be checked by taking orthogonal cuts to both spots, simulated and real and comparing the width of the Gaussian distributions of normalized power intensity.
David talks about his experience:
The Matlab code I implemented was mainly based on ‘call-back’ functions linked to buttons and toggles that can be pressed on the main window. The imported data also needed image processing and geometrical transformations (trigonometry, perspectives and projections on cylindrical planes… etc.) to make it suitable for comparison with simulations. In summary, I learned how to implement interactive GUI’s in Matlab and more specifically for image processing. This is very helpful for developing a user-friendly and clean GUI for any other scientific, control or data processing purpose in my research project. It will also make so much easier for new students/users to use for the first time any Matlab program that incorporates such an interface. Objectives were accomplished within the two weeks and the Matlab code was given to the company for further improvement and future use.
I also attended a few meetings where the team discussed about the state-of-the-art of some of their projects and about a visit to an international congress, which was very insightful and I could hear about the latest news on solar technology and the company’s plans moving forward.