(113d) Design Optimisation for Optimal Shape Selection of Solar Parabolic Trough Receiver
- Conference: AIChE Spring Meeting and Global Congress on Process Safety
- Year: 2016
- Proceeding: 2016 AIChE Spring Meeting and 12th Global Congress on Process Safety
- Group: Computing and Systems Technology Division
- Time: Tuesday, April 12, 2016 - 2:36pm-2:58pm
In this study, a procedural approach for the geometric parameterisation, robust design and multi-objective shape optimisation of a parabolic photovoltaic trough receiver. The shapes considered for optimisation is based on a half-circle parabola, an equilateral triangle elliptic major axis profile solar radiation heat exchanger. Two objectives of minimum heat transfer area of the receivers and the minimum pressure drop of the shapes were used to obtained or determine the optimum shape with the lowest heat transfer area corresponding to the lowest friction factor or lowest resistance to flow of the heat transfer fluid, without discriminating or favouring either of the objectives . The multi-objective optimisation of the three dimensional shapes was based on the Pareto dominance concept of the Elitist Non-dominated Sorting Genetic Algorithm, popularly referred to as NSGA-II. The approach adopted completely automated starting from a what-if analysis of the initial shape best guess, followed by a correlation studies that provides parameters that matters to analysis of the shape and it the output. A design of experiment takes the parameters from the correlation studies based on the design space and bind them within the design space. A response surface provides the interpolation algorithms that produce the line for 2D and surface for 3D visualisation of how the input parameters predict the output parameters, based on mathematical model. A multi-objective GA provides a design set of shape base on 3-star criteria for minimum heat transfer area and pressure drop. Finally, a Six-Sigma or robust design was used to benchmark the shape to ensure that shape works well both in the virtual world and the real world, based on uncertainty such as ease or difficulty of manufacture and operator handling during the manufacturing process.