【摘要】 针对采用炉排上下方压差来衡量炉排炉干燥段垃圾厚度时测量值波动过大的问题，提出了一种垃圾料层厚度软测量模型。首先，采用皮尔逊相关性分析，消除模型输入变量之间的迟延；然后，结合专家经验，通过现有垃圾厚度模型计算得出垃圾厚度等级；最后，建立基于布谷鸟搜索算法优化长短期记忆的垃圾料层厚度软测量模型。实验结果表明，相比反向传播神经网络、Elman神经网络与长短期记忆模型，所提出的软测量模型具有较高的精度，可实现对料层厚度的准确判断。更多还原

【Abstract】 Aiming at the problem that the measured value fluctuates too much when the differential pressure between the top and bottom of the grate is used to measure the garbage thickness in the drying section of the grate furnace, a soft measurement model of garbage layer thickness is proposed. Firstly,Pearson correlation analysis is used to eliminate the delay between model input variables. Then,combined with expert experience, the thickness grade of the waste layer is calculated through the existing waste layer thickness model. Finally, a soft measurement model of garbage layer thickness is established based on cuckoo search algorithm to optimize long short-term memory. The results show that compared with the back propagation neural network, Elman neural network and long short-term memory model, the proposed soft measurement model has higher accuracy and can realize accurate judgment of material layer thickness.更多还原