A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Environment, Resources and Development

Thesis (M.Eng.) - Asian Institute of Technology, 1981

A mathematical model has been formulated to predict the performance of a solar thermal system. The model is developed for use in hour by hour computer simulations. Experimental systems exhibiting various sizes of storage tank volume with 4 typical load patterns were installed and tested at AIT energy demonstration park for a 10 weeks period. The experimental results and the results predicted by the model utilizing actual solar radiation and ambient temperature recorded in the meteorological station located at the experiment site are in good agreement. The simulation model can, therefore, be used with confidence to study the long term performance of solar energy systems with various types of load characteristics. Two different solar radiation simulation models are used as the solar energy input to the systems, one is the probability binomial distribution model derived by Dr. Exell and the other is a daily sinusoidal distribution model . The two models outputs show not so much difference , then the latter model is adopted since it involves much shorter computer execution time than the Dr. Exell's model. Th e long term performance resulted from the simulations are presented and the optimal storage tank volume of various systems are discussed. The actual work provides the solar engineering engineers with one of the necessary mathematical tool for undertaking the cost-benefit ratio optimization prior to the design and installation of large solar water heater in Thailand. The model is easy to adapt to other countries if local solar radiation availability models are known.