On the basis of actual operation conditions of coal-fired boilers

a controlled auto-regressive integrated moving average(CARIMA)model for boiler's combustion system was deduced.Moreover

the fuzzy predictive controller for this model was designed.According to the features of self-tuning multistep prediction and feedback compensation of the generalized predictive control

and regarding the fuzzy control as the correction regulation

the boiler combustion control system was divided into three subsystems

such as the fuel control system

air flow control system

and furnace pressure control system.Finally

the fuzzy PID controller was employed to optimize each control subsystem.The simulation results showed that

after the optimization

the fuel control system had a stable response and a faster tracking speed

on the premise of ensuring the control precision;the air flow control system and the furnace pressure control system

which had a short adjusting time

and both the overshoot and steady state error were zero

had a strong anti-interference ability.