论文部分内容阅读
在试验台上对双循环流化床的物料循环进行系统的试验研究,以试验数据为基础建立加入动量的BP神经网络预测模型并进行有效模拟。定义平均偏离度来评价模型预测值相对于试验值的平均偏离情况,通过对比分析试验数据与神经网络模型预测值,表明神经网络模型预测值相对于试验值偏差不超过5kg/(m2.s),相对误差在±20%以内,平均偏离度仅为8%。结果表明建立的神经网络模型具有较好的预测效果。
On the test bench, a systematic experimental study was conducted on the material circulation of the double-circulating fluidized bed. Based on the experimental data, a BP neural network prediction model for adding momentum was established and validated. Define the average deviation to evaluate the average deviation of the model predictive value from the experimental value. By comparing the experimental data with the predictive value of the neural network model, it shows that the deviation of the predictive value of neural network model from the experimental value does not exceed 5kg / (m2.s) , The relative error within ± 20%, the average deviation of only 8%. The results show that the established neural network model has a good prediction effect.