| 摘要 |
<p>PROBLEM TO BE SOLVED: To perform proper learning according to the change of an operation condition, and to properly cool the head part of the steel strip by obtaining a deviation from the true value of a thermal conductivity according to a difference between the predicted value of the winding temperature of the steel strip and an actual value by the sequential type least square, method setting it to a learning value, and correcting the thermal conductivity of a cooling control part. SOLUTION: The forecast cooling speed of the steel strip 1 is calculated by a cooling controller 20 in consideration of a cooling speed from the signals of finish outlet side thermometer 8, finish speed detector 13, winding thermometer 9 and winding speed detector 12. By using the thermal conductivity, water injection is controlled so that this predicted value and a target value coincide. The information of a cooling control part 20 and the information of the operation condition are inputted in a learning calculation part 30, the learning value of the thermal conductivity of the head part of the steel strip 1, is calculated previously for each group of various steel strips 1, and they are stored. According to the operation condition, the thermal conductivity of the cooling control part 20 is corrected by the stored learning value by the learning calculation part 30. For the thermal conductivity to be used for the calculation of a predicted cooling temperature in the divided area of a water injection bank 4, the sequential type least square method is applied.</p> |
