提高鞍钢冷连轧机效能和产品精度的途径

通过对鞍钢1680mm冷连轧机力能参数、结构参数、带钢厚度和板形精度的大型综合测试，采用综合遗传神经网络建立轧制力模型，与实测误差不大于5％；采用综合改进遗传算法与冷连轧机辊型系统优化相结合，板形废品率从2．24％降至1．19％；运用缩合改进遗传算法与冷连轧机大系统参数相结合优化轧制规程，相同厚度来料成品由1mm扩大至0．8mm；运用“机电结合故障诊断法”，打出带钢厚差无法消除和易断带的原因。综合措施提高了轧机的效能和产品的精度。

WAYS OF INCREASING EFFICIENCY AND PRODUCT PRECISION OF COLD TANDEM STRIP MILL AT ANSHAN STEEL

By measurement of capacity parameters, structure parameters, strip thickness and shape precision of the product on 1680 mm cold tandem strip mill at Anshan Steel, a rolling force model has been built by use of Mended Genetic Algorithms BP Networks, with a tolerance less than 5 % comparing with measured By optimizing roll system by use of Mended Genetic Algorithms, strip rejection has been reduced from 2 24 % to 1 9 % for flatness Optimizing rolling schedule by use of combining Mended Genetic Algorithms with rolling mill big system parameters, product thickness can be reduced from 1 0 mm to 0 8 mm for same entry strip thickness The reason why the strip thickness tolerance is difficult to reduce and strip is easy,to break has been found by use of mechanical and electrical fault diagnosis method Finally, rolling mill efficiency and product precision has been increased by use of all above mentioned measures