高压密相气力输送固相流量的实验与GRNN网络预测

Experimental Study of High-pressure Dense-phase and Pneumatically Transported Solid-phase Flow and its Prediction Based on a GRNN(generalized regression neural network)

煤粉高压密相气力输送是气流床加压气化的关键技术之一。在输送压力可达3.7MPa,管路固气比可达660kg/m3的气力输送实验台上进行系统的研究,考察输送压力、输送差压、流化风量、充压风量、补充风量、煤粉含水率等条件对固相质量流量的影响。结果表明:固相流量随着输送差压的增大而增大;随着流化风量的增大而先增大后趋向于某一定值;注入风量一定时,随着充压风量的增大而先减小后增大;与补充风量的大小基本无关;随着煤粉含水率的增大而减小。同时建立广义回归神经网络(GRNN)对固相流量进行了有效预测,最大预测误差在2.3%以内。上述工作将为系统的控制和运行提供一定的指导,同时为深化高压密相气力输送的研究奠定基础。

Pulverized-coal high-pressure dense-phase pneumatic transmission represents one of the key technologies for the pressurized coal gasification of a gas fluidized bed.A systematic study was performed on a pneumatic transmission test rig featuring a transmission pressure up to 3.7 MPa and a pipeline solid-gas ratio of 660 kgm3 to investigate the influence of such conditions as transmission pressure,transmission pressure difference,fluidized air quantity,pressurized air quantity,supplementary air quantity and water content of pulverized coal etc.on the solid-phase mass flow rate.The results of the study show that the solid-phase flow rate increases with the increase of transmission pressure difference.It first increases with an increase in fluidized air flow rate,and then tends to be a constant value.When the injected air quantity reaches a constant one,the flow rate in question will first decrease and then increase with the increase of the pressurized air quantity,basically independent of the supplementary air quantity.The flow rate will decrease with an increase of the water content of the pulverized coal.In the meantime,a generalized regression neural network(GRNN)was established to effectively forecast the solid-phase flow rate with the maximal prediction error being within 2.3%.All these efforts will somewhat provide guidance for system control and operation,and at the same time lay a solid foundation for an in-depth study of the high pressure dense-phase pneumatic transmission.