金矿浮选回收率预测模型

浮选回收率是金矿选矿过程重要的生产指标,目前主要是通过人工化验的方法检测获得,人工检测周期较长,造成金矿厂不能及时把握浮选工艺水平.在大量现场生产数据的基础上,分别采用多元线性回归和BP神经网络的方法,建立了金矿厂浮选回收率的预测模型.预测误差分析表明,BP神经网络预测模型能较好地预测金矿厂的浮选回收率,当预测相对误差在±3%范围内时,模型的预测精度达到91%,对于实际生产具有良好的参考作用.     

Intrinsic ZnO films fabricated by DC sputtering from oxygen-deficient targets for Cu(In,Ga)Se_2 solar cell application

Intrinsic zinc oxide films,normally deposited by radio frequency(RF) sputtering,are fabricated by direct current(DC) sputtering.The oxygen-deficient targets are prepared via a newly developed double crucible method.The 800-nm-thick film obtaines significantly higher carrier mobility compareing with that of the 800-nm-thick ZnO film.This is achieved by the widely used RF sputtering,which favors the prevention of carrier recombination at the interfaces and reduction of the series resistance of solar cells.The optimal ZnO film is used in a Cu(ln,Ga) Se2(CIGS) solar cell with a high efficiency of 11.57%.This letter demonstrates that the insulating ZnO films can be deposited by DC sputtering from oxygen-deficient ZnO targets to lower the cost of thin film solar cells.     

LF精炼终点钢水温度灰箱预报模型

LF精炼工序在炼钢过程起着调节温度的关键作用,准确预报LF精炼终点钢水温度对实际生产有重要意义.传统的LF精炼预报模型包括机理模型与黑箱模型.机理预报模型能够体现各工艺因素对终点钢水温度的影响,但由于LF精炼传热机理研究尚不完善,依靠机理模型预报终点钢水温度,难以达到预期效果;黑箱预报模型能够准确预报终点钢水温度,但不能反映精炼过程各工艺因素对钢水温度的影响,尤其当生产工艺条件发生改变时,黑箱模型在应用上会受到限制.本文以方大特钢LF精炼炉为研究对象,建立一种机理预报模型与黑箱预报模型(BP神经网络预报模型)相结合的LF精炼终点钢水温度灰箱预报模型.该模型既能反映各工艺因素对终点钢水温度的影响,又能准确预测终点钢水温度,其终点钢水温度预测误差在±5℃以内的命中率可以达到95%以上.