A methodology to investigate the cause of quenching in once-through tower type power plant boilers

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Due to the penetration of variable renewable energy (VRE) sources, conventional coal fired power plants need to operate with greater flexibility via two-shifting or low load operation whilst remaining reliable and conserving the lifetime of components. Thick sectioned components are prone to thermal fatigue cracking as a result of through-wall temperature gradients during start up and shutdown. These temperature gradients can be significantly amplified during quenching when components at high temperature are unintentionally exposed to colder liquid or steam. Such quench events are known to occur during two-shift operation of a large once-through coal fired tower type boiler, which is the subject of this study. The purpose of this study is to develop and demonstrate a methodology to determine the root cause of quenching in a once-through tower type boiler and provide information that can be used to predict the impact on thick-walled components by estimating the through-wall temperature gradients. The first modelling element in the methodology is a simplified transient heat transfer model for investigating condensation of steam in the superheater. The model is presented and verified by comparison with real plant data. The second element is a liquid tracking model that approximates the liquid level in the superheater as a function of time to predict the location and magnitude of through-wall temperature gradients. The complex geometry of the superheater was divided into a number of control volumes and a dynamic thermo-fluid process model was developed to solve the transient conservation of mass and energy equations for each volume using a semi-implicit time wise integration scheme. The liquid tracking model was verified by comparison with a similar model constructed in Flownex and also by comparison with plant data. Varying levels of discretisation were applied to a particular quench event and the results are presented. The third modelling element is a two-dimensional transient pipe wall conduction model that is used at selected localities to evaluate the temperature gradients within the pipe wall. The temperature gradients and internal heat flux were verified by temperature measurements from the outer surface of a main steam pipe undergoing quenching. The stresses associated with the temperature gradients were also briefly considered. The real plant quenching problem is analysed in detail and found to be caused by liquid overflow from the separators. A particular plant configuration creates a previously unidentified siphon of water from the separating and collecting vessel system into the superheater. This situation is not recognised by plant operators and thus persists for some time and causes flooding of the superheater. Analysis of the resultant through-wall temperature gradients show that quenching causes significant stresses which can be avoided. By understanding the causes and preventing the occurrence of quenching, the life of thick-walled high temperature components can be conserved.