Abstract:  Using the existing technologies of furnace flame temperature field detection and reconstruction, it is convenient to obtain section temperature field, which can be used as output in stochastic distribution control (SDC) for the fuel control system. Based on the characteristics of four corners tangentially fired furnace, a model of section temperature field is built. Then the corresponding output probability density function (PDF) dynamic model based on spline functions is identified. Under the framework of SDC theory, conventional control and predictive control are adopted to make the section temperature field track the ideal distribution. The simulation results show the effectiveness of the control algorithms. Compared with the traditional furnace combustion control strategy, this method without problems from large delay and large inertia has good application prospects.

Key words: furnace temperature field, probability density function, dynamic modeling, predictive control