What is the benefit of using a physical model to model the power generation (i.el solar field with parabolic trough collectors)? A data-driven method could potentially achieve similar accuracy with massive training data?
All physical models are based on the laws of physics and especially on thermodynamics. They are mostly used for design calculations. The benefit is, that you only need input data, no output data.
All empirical models are based on experimental results. The influencing parameters on the results are then weighted with different coefficients and incorporated into the solution algorithm as a mathematical equation. The coefficients are then adapted so that the calculation result matches the sea values. You need input and output data.
However, there are also the combinations of both models. E.g. in the partial load case mostly corresponding efficiency characteristics or characteristic arrays are used. The same applies to the fouling of heat exchangers. The extent of contamination cannot be measured, only the result, so that the overall heat transfer coefficient is supplemented by the fouling parameter.
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training