The demand for hot rolled strips with a better quality has been continuously grown in recent years. In particular, the uniform thickness of the strips is a key issue in determining the strip quality. In the present study, a fuzzy algorithm to calculate the roll speed variations was developed and a looper position control model was newly proposed to improve the thickness uniformity of hot rolled strips. Also, a set-up system that prescribes appropriate hot rolling process parameter values such as roll speed, rolling temperature, roll separating force, and front and back tensions was developed for the hot finishing mill system that consists of seven stands. Since the strip thickness is affected by roll separating force depending on the roll speed, the strip thickness deviation can be reduced by controlling the roll speed. Therefore, slab analyses were carried out to determine the relation between roll separating force and roll speed for various process parameters such as roll speed, reduction ratio, strip entry thickness, and front and back tensions. From the production data flow stress - strain rate relation used in the slab analysis was acquired. Based on the analytical results, the relation between roll separating force and roll speed was approximated by a log function and a fuzzy algorithm was developed to determine variations in roll speed due to variations of roll separating force, depending on various ranges of rolling temperature, reduction ratio, front and back tensions, and strip thickness. In addition, roll speed variation during the process may cause changes in strip tension and unbalance of mass flow of the strip. When such phenomena occurs, the looper must be controlled accordingly. In order to solve this problem, a looper position control model was also developed based on the geometric condition of the looper system. To integrate the developed fuzzy algorithm of roll speed variation with the looper position control model, values of process parameters at the initial stage of hot rolling must be known in advance. Therefore, the roll speed and temperature at each stand of the finishing mill system were calculated by predicting the accurate neutral position in the roll bite and by considering heat loss due to convection, respectively. The roll force was predicted by fuzzy control algorithm based on the analysis of production data and a neural network system was constructed to determine the front and back tensions. It was found out that the roll speed variation fuzzy algorithm, looper position control model, and a process parameter set-up system were successfully integrated in reducing the thickness deviation of steel strips and maintaining stability of the process.
