Advances in Steam Quality Monitoring Systems in Power Plants

Introduction

The US Energy Information Administration (EIA) predicts that despite the wide availability of low power consuming lightbulbs and high power efficient appliances, demand for electricity will reach 351 GW over the next 25 years, mainly driven by the increasing installations of HVAC (Heat, Ventilator, and AC) equipment in homes or industrial plants in the world. Part of this energy will be provided by renewable energy sources such as wind energy and solar or nuclear, but most of it is expected to be supplied by natural gas and to lower extend geothermal plants. These sources of energy usually use/produce a high quantity of steam, the quality of which may affect considerably the output power level. Impurities carried in the steam can cause hydrogen damage and fouling of superheaters, re-heaters, and turbines. In addition, a condensation due to energy losses results in liquid droplets of sub-micrometer size carried in the steam. The size of droplets increase as the steam travels to the turbine which can cause damage to various components of the system including the turbine itself and also erosion of the internal wall of the pipe which in turn generates a stream of solid contaminants that can be even more critical to the system such as failure of the turbine blades and also deposit of solid particles at some locations of the pipeline, causing pressure drops of the steam and hence can substantially lower the output power. A thickness grow of only 3 mils can cause an increase of 1 to 2% in the fuel bill revenue. Hence, carryover of solid impurities and droplets of water in a superheated steam turbine remains a major concern for steam generators which may lead to turbine unbalance and threat of the integrity of the other plant equipment such as the condenser, heater, pumps, boilers, and turbines. This challenge is more crucial in geothermal power generation plants which contains naturally-occurring contaminants besides pure water than in other conventional steam powered power generation (e.g. coal, natural gas, and nuclear). Nevertheless, these latest technologies are also subject to substantial amount of contaminants created in the pipeline due to some mechanical causes. One way to mitigate these contaminants is to use fluid filters and separators (E. Al Hosani, 2014; M. Rehman, 2012; S. Teniou, 2012; Z. Piyushsinh, 2014). However this requires recurrent replacement of the filters and also would cause some increasing pressure drops of the steam. Another approach is to monitor the quality of the steam with or without sampling. Online monitoring has already been considered by some power plant operators (e.g. energy Development Corporation and Mercury) by measuring either a single compound (e.g. amount of sodium) or multiple variables (e.g. silica, gas content, pH, and conductive). Offline monitoring is also widely considered in power systems by acquiring isokinetic samples in a way that the sample represents the flowing steam. This book chapter presents at first a background of steam-based power plants, together with different sensors used for steam quality monitoring. This is followed by a presentation of a new hybrid device for real-time measurement and imaging of moving solid and liquid contaminants that may occur in steam generators. The device explores the fact that the dielectric of the steam is approximatively one while the one of water droplets when exposed at high temperature can range between 8 and 65 (M. Meribout, 2011a, E. Zhang, 2008). This contrast of dielectric values can help track water droplets as well as other solid contaminants. The device uses a dedicated Near Infra-Red device to determine the type of contaminants (i.e. water droplets and iron oxide particles) and a THz imaging system which measures the amount of contaminants as well as its flow rate. The NIR subsystem uses a pattern recognition method based on a combination of principal component analysis and least squares support vector machine (LS-SVM) (Fukutomo, 1997; Kothare, 2000; Jean, 2008; Lord, 1980). The usage of image processing techniques together with NIR spectrometry constitutes a new promising step in flow metering (Baker, 2003; Bunce, 2011; Meribout, 2010b; Grimmelius, 1999). This is demonstrated by the extensive experiments which have been conducted for different scenario where the NIR subsystem system could determine the concentration of water droplets and solid contaminants with a maximum uncertainty of +/- 1.45% and +/- 1.16% respectively. With the NIR subsystem, pixel-level accuracy of motion vector was achieved, while the concentration of solid contaminants showed consisted proportionality with the average pixel intensity (Meribout, 2009c; Gao, 2009; Meribout, 2002d).