Theoretical and experimental studies of the spatial sensitivity of an electrostatic pulverised fuel meter

Abstract

In coal-fired power plants, the coal is pulverised and pneumatically conveyed to the burners. It is essential to measure and control pulverised fuel (PF) to improve combustion efficiency, reduce pollution and lower operating costs. Unlike single-phase flow, where flow density is generally considered to be uniform, air–solids two-phase flow in pneumatic conveying systems can undergo inhomogeneous concentration distributions across a given pipe cross-sectional area, especially around bends and bifurcators or trifurcators. The “roping” flow regime is an extreme example of inhomogeneous solids distribution, where highly concentrated solids form a column-like (rope-like) flow. For such complex flow regimes, solids concentration or solids mass flow meters will give different outputs when inhomogeneous flow occurs at locations where these meters are installed if they are sensitive to the flow patterns. This problem can be solved in two ways. The first and simplest approach is to restrict the meter's installation to locations where solids are relatively uniformly distributed over the cross-sectional area of the pipe. Alternatively, it can be solved by using a meter having uniform spatial sensitivity, so that the measurement would not be affected by the flow regimes. Because the installation locations of PF meters must suit other requirements, such as accessibility for the convenience of maintenance, it is ideal if the meters have uniform spatial sensitivities. Electrostatic mass-flow meters are typically used to give a measure of the fuel mass flow rate. One design employs ring-shaped electrodes which have non-uniform sensitivity, resulting in variations in meter output for the same flow stream passing through the sensing volume at the different radii relative to the meter's central line. This paper presents a theoretical analysis of the spatial sensitivity of the electrostatic meter with ring-shaped electrodes in the time and frequency domains. One goal of the study is to improve its performance and to achieve uniform sensitivity. The experimental data presented in this paper support the overall mathematic modeling, based on electrostatic field theory, using the finite-element method (FEM). Although the FEM analysis provides useful results, a more rigorous investigation is recommended for future work.