An experimental study of forced flame response in technically premixed flame in a lean premixed gas turbine combustor

Open Access
Author:
Orawannukul, Poravee
Graduate Program:
Mechanical Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
February 28, 2014
Committee Members:
  • Domenic Adam Santavicca, Dissertation Advisor
  • Thomas Litzinger, Committee Member
  • Richard A Yetter, Committee Member
  • Sarma V Pisupati, Committee Member
Keywords:
  • Forced flame response
  • technically premixed flames
  • combustion instabilities
  • flame transfer function
Abstract:
Flame response in a technically premixed, swirl-stabilized, turbulent, lean-premixed natural gas-air flame is investigated in a single research gas turbine injector. The main purpose of this research is to understand flame behavior in the context of combustion instabilities under technically premixed conditions, which is characteristic of industrial operation. In this operating mode, flames are exposed to both velocity and equivalence ratio fluctuations. Previous studies focused on heat release rate fluctuations due to velocity fluctuations, typically using CH* chemiluminescence emission as an indicator of heat release rate. However, chemiluminescence emission can only be used to measure heat release rate fluctuations in the presence of velocity fluctuations or equivalence ratio fluctuations, not both. Therefore two important questions need to be addressed. First, how does a flame respond to equivalence ratio fluctuations? Second, how can heat release rate fluctuations be measured in technically premixed flames? Flame response to equivalence ratio fluctuations is studied with operating conditions similar to industrial gas turbine conditions, with the exception of combustor pressure. The response is primarily due to a convective mechanism. Fuel flow rate fluctuations produce equivalence ratio fluctuations at the fuel injection location and then travel through the injector and arrive at the flame with the mean flow. The flame response decreases with increasing frequency (Strouhal number). A previously proposed "reconstruction technique" is experimentally validated and applied to measure heat release rate fluctuations in technically premixed flames. The heat release rate fluctuations in technically premixed flames are a vector superposition of the heat release rate fluctuations due to velocity fluctuations and heat release rate fluctuations due to equivalence ratio fluctuations. This technique assumes that the flame’s response to velocity fluctuations and equivalence fluctuations are both linear and independent. Since there is no independent measurement of heat release rate fluctuations in technically premixed flames that can be used to validate the reconstruction technique, the technique is indirectly validated over a range of operating conditions using chemiluminescence intensity fluctuation measurements. It was found that this technique is valid in both global and local measurements. The accuracy is within 20% in amplitude and 10o in phase. Strouhal number has the greatest contribution to the accuracy. Flame response in technically premixed flames is measured. Flame transfer functions in technically premixed flames are the result of the constructive and destructive interactions between the flame’s response to velocity and equivalence ratio fluctuations, and depend on both the relative amplitude and phase of these contributions.