Experimental and numerical studies of thermomechanical problems of solar tower power plants

Resumen

Concentrating solar power (CSP) with thermal energy storage is one of the most promising renewable energy technologies for electric generation. However, these systems are not completely mature, and several challenges must be overcome to reduce the cost and to guarantee the safe operation of the plant. This PhD thesis addresses some of these challenges studying relevant thermo-mechanical problems in different subsystems of solar tower power plants: a storage system consisting of a tank of solid particles and molten salt, a molten salt external receiver, and a particle-to-CO2 heat exchanger for the next generation CSP plants with a particle-based receiver. Simplified models capable of predicting the temperature evolution with time of these subsystems have been developed, to study important aspects related to their control and structural integrity. Concerning the thermal storage system, the critical phenomenon called thermal ratcheting related to the cyclic operation of molten salt thermocline tanks has been studied. A thermocline tank is a thermal storage system for CSP plants, which consists of a filler bed of rock with molten salt flowing through the bed. The use of a thermocline tank instead of two tanks typically used in comercial solar power plants could reduce the capital cost of the storage subsystem. However, the cyclic thermal expansion and compression of the tank wall in thermocline tanks result in thermal stresses that can lead to the mechanical failure of the steel shell, which is known as thermal ratcheting. A simplified model has been developed in this work to study the unsteady thermal response of the tank and the thermal stress in the tank shell. The model considers the fluid flow inside the tank as one-dimensional along the tank axis direction, whereas the heat conduction in the multiple layer wall is considered to be two-dimensional. The model has been validated against experimental and CFD results available in the literature, and a good agreement is observed, but the simplified model presented in this PhD thesis has a lower computational cost, which is an advantage for studying the influence of different parameters, such as the molten salt velocity, in the thermal ratcheting. Concerning the heat exchanger coupling the receiver system and the power block of a CSP plant, a simplified model of a moving packed-bed particle-to-CO2 heat exchanger in a shell-and-plate configuration has been developed to study the transient operation and control of the heat addition to the power cycle in particle-based solar towers. The particle heat exchanger is one of the technology gaps that should be addressed to demonstrate the viability of particle-based power towers with a supercritical CO2(sCO2) cycle, that have been identified as a potential pathway for the next generation CSP plants. The control system of the heat exchanger proposed in this work is based on a split valve in the sCO2 stream that allows obtaining the desired outlet temperatures of sCO2 and particles by varying the flow rates. Molten salt external receivers are subjected to high and nonuniform heat flux, which causes a high thermal stress and deformation of the tubes that can reduce its service life. In this PhD thesis, experimental work to study the uneven temperature distribution of a central receiver tube has been conducted for two different operating conditions of solar tower power plants: nominal conditions, when the molten salt is flowing through the tube, and startup conditions, when the empty tube is preheated until a given value of temperature is reached, when the molten salt starts to flow. Very few experimental facilities allow studying the heat transfer characteristics of molten salt external receivers, which is useful to validate thermo-mechanical simulations, as well as to investigate the optimal operating conditions that maximize the thermal efficiency of the receiver while preserving its structural integrity. In real CSP plants, it is complicated to accurately measure the unsteady solar flux reaching the receiver; thus, an inverse heat transfer problem has been applied in this PhD thesis to obtain the heat flux from the measurement of the outer surface temperature, which can be accomplished in practice. The inverse heat transfer problem involves the development of a numerical model that simulates the heat transfer characteristics of the receiver tube. Two different models have been developed to simulate the tube under both nominal and startup operating conditions: the first considers a 2-D heat diffusion along radial and circumferential directions of the pipe, and establishes a convective heat transfer boundary condition between the inner surface of the pipe and the molten salt flow, whereas the second considers a 3-D heat diffusion and establishes a radiative heat transfer condition at the inner surface, which is in contact with air. The solution of the inverse heat transfer problem allows for obtaining the temperature distribution along the cross section of the tube; thus, the thermal stress and deflection of the tube caused by the unilaterally heat flux has been calculated.