APPLICATION OF MULTIVARIABLE MODEL PREDICTIVE ADVANCED CONTROL FOR A 2×310T/H CFB BOILER UNIT
When a CFB boiler is in automatic control, there are strong interactions between various process variables and inverse response characteristics of bed temperature control target. Conventional PID control strategy cannot deliver satisfactory control demand. Kalman wave filter technology is used to establish a non-linear combustion model, based on the CFB combustion characteristics of bed fuel inventory, heating values, bed lime inventory and consumption. CFB advanced combustion control utilizes multivariable model predictive control technology to optimize primary and secondary air flow, bed temperature, air flow, fuel flow and heat flux. In addition to providing advanced combustion control to 2×310t/h CFB + 1×100MW extraction condensing turbine generator unit, the control also provides load allocation optimization and advanced control for main steam pressure, combustion and temperature. After the successful implementation, under 10％ load change, main steam pressure varied less than ±0.07MPa, temperature less than ±1℃ , bed temperature less than ±4℃ , and air flow （O2）less than ±0.4％.
circulating Fluidized bed boiler system multivariable predictive model advanced control
Zhao Weijie Dai Zongliao Gou Rong Gong Wengang
Sinopec Shanghai Petrochemical Company Limited Thermoelectricity Division, Shanghai,200540, China Ho Sinopec Shanghai Petrochemical Company Limited Thermoelectricity Division, Shanghai,200540, China Sh
国际会议
第20届国际流化床燃烧会议(20th International Conference on Fluidized Bed Combustion)
西安
英文
817-823
2009-05-18（万方平台首次上网日期，不代表论文的发表时间）