论文部分内容阅读
设计并搭建双循环流化床冷态试验台,通过试验分析提升管二次风风量、二次风送风方式、二次风口高度和二次风口数目对颗粒循环流率的影响。建立BP神经网络预测模型,采用3种算法对颗粒循环流率进行预测,通过对比找出最优预测模型——基于LM算法的改进型BP神经网络预测模型。该预测模型很好地预测了二次风特性对颗粒循环流率的影响,试验值与模型预测值的平均绝对误差为0.23kg/(m2·s),平均相对误差仅为1.37%;最大偏差为1.23kg/(m2·s),最大相对误差5.75%。
Design and set up a double-circulating fluidized bed cold test bed. The effects of secondary air volume, secondary air supply mode, secondary tuyere height and number of secondary tuyere on the particle circulation flow rate were analyzed through experiments. BP neural network prediction model is established, and three kinds of algorithms are used to predict the circulating flow rate of particles. The optimal prediction model is found through comparison - an improved BP neural network prediction model based on LM algorithm. The prediction model predicts the effect of the secondary air characteristics on the circulation rate of the particles. The average absolute error of the experimental values and the predicted values is 0.23 kg / (m2 · s), the average relative error is only 1.37%. The maximum deviation 1.23kg / (m2 · s), the maximum relative error of 5.75%.