An Online Dwell Time Optimization Method Based on Parts Performance for Hot Stamping

Abstract Hot stamping for high strength steel is an effective method for lightweight of vehicles. Hot stamping parts provide better structural strength while reducing weight, so as to improve safety performance of cars. But due to dramatic temperature changes for high-strength steel sheets and molds in hot stamping process, the final performance of parts is determined by a variety of complex factors. The dwell time is closely related to the part performance. If dwell time is too short, parts performance will decrease significantly. But reducing the dwell time can effectively shorten the production cycle of hot stamping. In the actual production, the dwell time is always longer in order to ensure the part performance. It is found that for a given part, there is a Critical Dwell Time (CDT); if the dwell time is greater than CDT, the performance of the part reaches the expected performance; if the dwell time is slightly smaller than CDT, the performance of the part decreases slightly. This paper presents an online optimization method of hot stamping process based on part performance. This method uses online non-destructive sensor, which is a Magnetic Barkhausen Noise sensor, to monitor the performance of hot stamping parts, and adjust the dwell time in smaller steps according to the performance of parts. When the performance of the part fluctuates around the expected performance, CDT will be determined. The part performance is monitored continuously, if part performance is lower than expected, the dwell time is increased. By using this method within a control simulation, the performance of the parts keeps as expected, the threshold of the dwell time for the specific part and its performance is obtained, and the forming cycle of the parts (including deformation process and dwelling process) is shortened by about 8~21%, compared with the original one, and the production efficiency of forming equipment in the hot stamping production line will be increased obviously.