@     @
TITLE Development of advanced combustion control system that predicts optimum combustion conditions
(Joint Program to Promote Technological Development with the Private Sectors)
AUTHOR

RITE - SAMUKAWA LABORATORY in Yamatake Corporation
RITE - FUJISAWA LABORATORY in Yamatake Corporation

SOURCE @@
ABSTRACT

‚PjDevelopment of a high efficiency combustion control system

A highly effective combustion control system that controls the amount of air and the fuel flowrate was developed for use in general combustion equipment where an analysis meter and an advanced control algorithm were not adopted. This control system was developed by acquiring data during combustion on the temperature, pressure, and flowrate of the combustion exhaust gas.  Yamatakefs original data mining techniques, TCBM (Topological Case-Based Modeling) and RSM-S(Response Surface Method by Spline), were utilized for the combustion control system. Data mining and TCBM were to find and model the causal relations of combustion.  RSM-S uses the O2 concentration in exhaust gas from modeled causal relation and calculates the control condition required to control of concentration of CO. When the combustor efficiency was evaluated with a boiler and the amount of CO2 exhaust was controled by adjusting the air fuel ratio, the boiler efficiency was usually improved by about 4% compared with a fixed air fuel ratio. The amount of the CO2 reduction can be provisionally calculated as 4,320(thousand ton/year) when applying this development technology to a steam boiler of 7 (metric ton)/h or less.

‚QjDevelopment of a high reliability combustion sensor

For the soft sensor to predict control values properly, high accuracy and long-term stability of the input signal reading of the composition of the exhaust gas will be demanded. The O2 concentration and CO concentration, which are important parameters in the composition of the combustion exhaust gas, sensors with high reliability that can continuously be measured were developed. For the O2 sensor, the zirconia tube with platinum coating was integrated with the ceramic heater. For the CO sensor, which uses RTDs and contact combustion to detect CO, the catalystfs protective layer continued to improve and this has improved the sensorfs durability when exposed to caustic gases such as SO2. Both sensors were left in SO2 2000ppm to evaluate long-term stability to the causticity element seen in a boilerfs chimney exhaust gas. The long term (about 180 days) result is that both sensors have not decreased in output. In addition a small, lightweight detector prototype in which both sensors were housed was built and installed in the boiler test plant.  After being installed the test plant, a 3 (metric ton)/h steam boiler, both sensors output were steady.